What’s moving now
Grok Bot or NotSep 10 · My tests, the hype cycle, and why I'm not pivoting.
34.4× the usual response · 1,512 engagements
18.2× the usual response · 22,050 engagements
17.6× the usual response · 2,611 engagements
9.4× the usual response · 99 engagements
6.7× the usual response · 8,117 engagementsDevin↑6 places in Coding Agents & Harnesses for GTM
LangGraph↑29 places in GTM Developer Tools
Neon↑28 places in GTM Developer Tools
OpenCode↑2 places in Coding Agents & Harnesses for GTM
Together AI↑24 places in GTM Developer ToolsTrending Topics
Market Moves
568 engagements · 1.5× usualPost ↗
deepseek ↗99 engagements · 9.4× usualPost ↗
U371 engagements · 6.1× usualPost ↗
308 engagements · 1.3× usualPost ↗
173 engagements · 3.4× usualPost ↗
Post ↗GTM Tech Index
Latest Posts
Sep 10Grok Bot or NotMy tests, the hype cycle, and why I'm not pivoting.
Sep 3Everything I measured in AI x GTM -- August readoutfull EOM report: 47 market moves, SaaSpocalypse fading, signal-based slowdown, AI theory hype cycles, and the AI slop conversation shift.
Aug 20Watermarking worriesTry this instead of hacking it.Tracking key moves in GTM tech since June 2026. See methodology. Something missing? Send it to adam[at]adamgtm.com
All topics
BetaFigures move as the panel grows and definitions sharpen.
-14.7%vs prior 30d+14.4%vs prior 30d+21.8%vs prior 30d>16×vs prior 30d-1.6%vs prior 30d+2.3%vs prior 30d+17.1%vs prior 30d
+12.2%vs prior 30d+836%vs prior 30d+29.2%vs prior 30d+86.6%vs prior 30d+2.1%vs prior 30d+9.7%vs prior 30d+12.4%vs prior 30d+104%vs prior 30d+13.9%vs prior 30d+38.6%vs prior 30d+3.1%vs prior 30d+3.0%vs prior 30d+45.9%vs prior 30d
+37.7%vs prior 30d-10.4%vs prior 30d-12.9%vs prior 30d
-78.5%vs prior 30d
-10.0%vs prior 30d+4.7%vs prior 30d-11.2%vs prior 30d±0%vs prior 30d-5.3%vs prior 30d+22.0%vs prior 30d
-32.8%vs prior 30d
-40.6%vs prior 30d+28.3%vs prior 30d
-20.4%vs prior 30d+9.8%vs prior 30d-16.3%vs prior 30d+96.3%vs prior 30d-15.2%vs prior 30d
+74.1%vs prior 30d
+40.2%vs prior 30d
+38.1%vs prior 30d-26.8%vs prior 30d-64.2%vs prior 30dNEWvs prior 30d
-16.0%vs prior 30d-29.3%vs prior 30d-40.0%vs prior 30d-21.3%vs prior 30d
-2.5%vs prior 30d
+0.9%vs prior 30d
-10.7%vs prior 30d
-42.9%vs prior 30d
-44.0%vs prior 30d
-45.8%vs prior 30d
-47.5%vs prior 30d
-41.1%vs prior 30d-35.1%vs prior 30d-67.0%vs prior 30dNothing matches this filter.
The tools and practices teams use to build agents for sales, marketing and RevOps.
Still the largest share of the conversation, but off its April high of 27.5% and down 17.8% over 90 days. The talk moved from whether to build agents to the plumbing under them: DeepSeek open-sourced an agent harness, Vercel shipped a plugin standard with Cursor and GitHub, Sierra started benchmarking agents that build agents. Roughly flat over the last 30 days.
Posts to understand this · as of Sep 8
Inside Building GTM Agents
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Codex | 79 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Grok Bot | 57 |
| 5 | Clay | 90 |
| 6 | Lovable | 57 |
| 7 | Salesforce | 87 |
| 8 | HubSpot | 83 |
| 9 | Replit | 54 |
| 10 | OpenClaw | 50 |
How brands earn visibility, citations and recommendations inside AI answers.
The largest block here, holding near 11% all year with a June peak. Spring brought the buildout: HubSpot shipped an AEO tool and Daydream raised $15M to be the AI search agency. Lately the fight is over the words, with Eli Schwartz quoting Google that there is no GEO or AEO without SEO fundamentals. Down 9% over the last 30 days.
Posts to understand this · as of Sep 12
Inside AEO/GEO
| Brand | GTM Index | |
|---|---|---|
| 1 | Ahrefs | 49 |
| 2 | Profound | 50 |
| 3 | Semrush | 49 |
| 4 | HubSpot | 83 |
| 5 | Claude Code | 99 |
| 6 | Clay | 90 |
| 7 | Webflow | 46 |
| 8 | Codex | 79 |
| 9 | Salesforce | 87 |
| 10 | AirOps | 36 |
How software exposes its data and actions so agents can use it directly.
Jumped from 5.5% to 9.1% in April and has drifted down since, now near 7.4% and flat over 30 days. The building blocks kept landing through the decline: Anthropic acquired Stainless, ElevenLabs shipped an MCP inside Claude, Cloudflare opened a waitlist to charge for MCP tools in stablecoins. The work is real; the novelty is spent.
Posts to understand this · as of Sep 8
Inside Headless & Agent Interfaces
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | Salesforce | 87 |
| 4 | Codex | 79 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | HubSpot | 83 |
| 7 | Instantly | 61 |
| 8 | Grok Bot | 57 |
| 9 | Lovable | 57 |
| 10 | Nooks | 46 |
The people, tools and operating model behind custom go-to-market systems.
No single event explains this one. Clay put its engine in the terminal in July with an API and CLI, Alex Vacca turned that into a twelve skill playbook, and Vercel shipped agent plugins as an open standard with AWS, Cursor and GitHub. Share still fell about 15% over ninety days and held flat over the last thirty. The tooling moved; the talk did not follow it.
Posts to understand this · as of Sep 12
Inside GTM Engineering
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Claude Code | 99 |
| 3 | HubSpot | 83 |
| 4 | Prospeo | 51 |
| 5 | Instantly | 61 |
| 6 | Codex | 79 |
| 7 | Salesforce | 87 |
| 8 | HeyReach | 52 |
| 9 | n8n | 59 |
| 10 | Lemlist | 50 |
Marketing that delivers value in the feed or answer instead of forcing a click or form fill.
The climb topped out in May with Dave Gerhardt arguing that MQLs and gated PDFs are finished. AutoRFP put numbers on it in April: 22.4% of inbound leads crediting ChatGPT. By August Rand Fishkin was asking whether the website still deserves the spend. No single event turned it down; the case is now assumed rather than argued. Off 13% in 30 days.
Posts to understand this · as of Sep 8
Inside Zero-Click Marketing Strategies
| Brand | GTM Index | |
|---|---|---|
| 1 | Ahrefs | 49 |
| 2 | Profound | 50 |
| 3 | Semrush | 49 |
| 4 | Clay | 90 |
| 5 | HubSpot | 83 |
| 6 | Webflow | 46 |
| 7 | Claude Code | 99 |
| 8 | Codex | 79 |
| 9 | Salesforce | 87 |
| 10 | AirOps | 36 |
The choice between open models and closed APIs, including cost, control, quality and lock-in.
This climbed from 2.19% in February to a July peak of 8.56%, up 87.3% over 90 days. Jensen Huang's open-models letter was the loudest single moment, and arguments about control and cheap open-weight tokens kept it fed. The last 30 days are down 41.8% off that peak.
Posts to understand this
Inside Open vs. Closed Models
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Claude Code | 99 |
| 4 | OpenRouter | 51 |
| 5 | Salesforce | 87 |
| 6 | Vercel | 58 |
| 7 | Together AI | 31 |
| 8 | Grok Bot | 57 |
| 9 | Baseten | 44 |
| 10 | Clay | 90 |
How companies prove AI delivers enough value to move beyond a pilot.
Share climbed from 6.5% in February to 7.9% in July, with a dip in May along the way, then fell to 6.1% in August. The last 30 days are down 28.8%. The talk has moved from whether AI works to what it costs and what the freed-up time actually buys.
Posts to understand this · as of Sep 8
Inside Proving AI ROI
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Salesforce | 87 |
| 3 | Codex | 79 |
| 4 | OpenRouter | 51 |
| 5 | Clay | 90 |
| 6 | Cursor (Anysphere) | 78 |
| 7 | HubSpot | 83 |
| 8 | Vercel | 58 |
| 9 | Lovable | 57 |
| 10 | Replit | 54 |
How large companies move AI from small pilots into secure, useful production work.
Adoption talk peaked in April at 7.5% and has drifted down every month since, with L30 off 24%. The posts show where it went. Rippling sells spend visibility, Mistral sells production orchestration, Informatica sells governed context. Deployment is assumed now, and the argument has moved to the plumbing that holds it up.
Posts to understand this · as of Sep 8
Inside Enterprise AI Adoption
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Salesforce | 87 |
| 3 | Codex | 79 |
| 4 | Cursor (Anysphere) | 78 |
| 5 | Clay | 90 |
| 6 | OpenRouter | 51 |
| 7 | Lovable | 57 |
| 8 | HubSpot | 83 |
| 9 | Vercel | 58 |
| 10 | Grok Bot | 57 |
How teams use buyer signals to decide who to target, when and why.
A long slide, from 7.9% in February to 4.0% now, down 31% across 90 days. Yamini Rangan's January case against ICP-only outbound was the high-water mark. Since then the idea got absorbed into agent-stack posts like Dima Bilous's seven-agent GTM, and the counter-position lands: Raouf Lemouchi selling matcha with Google Maps and a spreadsheet.
Posts to understand this · as of Sep 12
Inside Signal-Based Sales & ABM
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Claude Code | 99 |
| 3 | HubSpot | 83 |
| 4 | Salesforce | 87 |
| 5 | Instantly | 61 |
| 6 | Prospeo | 51 |
| 7 | Apollo.io | 54 |
| 8 | HeyReach | 52 |
| 9 | Lemlist | 50 |
| 10 | Apify | 53 |
How companies keep agents secure, accountable and under control as they take on real work.
Share climbed from February through July, then dropped in August and is down 34% over the last 30 days. The posts arriving now come after that peak. Automation Anywhere names governance as one of three gaps in running agents at scale, Lyzr puts governance in the build pipeline, and Conexiom sells AI that checks its work.
Posts to understand this · as of Sep 8
Inside Agent Governance
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Claude Code | 99 |
| 4 | Salesforce | 87 |
| 5 | OpenRouter | 51 |
| 6 | Clay | 90 |
| 7 | Grok Bot | 57 |
| 8 | OpenClaw | 50 |
| 9 | Vercel | 58 |
| 10 |  Cline | 24 |
How companies build trust and stand out when AI makes content cheap and abundant.
No event drives this one. It sits in a 4 to 6% band all year, L30 down 7% and L90 down 3%, and that flatness is the point. The complaint is chronic, not seasonal. Ann Handley on humanizer tools, Sjeel K.'s week with a CEO who found Claude, and a December post from Dave Gerhardt still pulling.
Posts to understand this · as of Sep 8
Inside Brand & Authenticity
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | Salesforce | 87 |
| 4 | HubSpot | 83 |
| 5 | Substack | 51 |
| 6 | Semrush | 49 |
| 7 | Codex | 79 |
| 8 |  Clarify | 29 |
| 9 |  Docket | 28 |
| 10 | Grok Bot | 57 |
How teams create, distribute and judge marketing content made with AI.
Share peaked at 5.2 percent in February, fell to 4.0 in March, then climbed back to 5.1 by June. July and August sat close together, 4.4 then 4.6. September is only twelve days in, so read the last 30 days instead: down 24 percent, with current share at 3.7.
Posts to understand this · as of Sep 12
Inside AI Content
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Claude Code | 99 |
| 3 | Midjourney | 40 |
| 4 | Higgsfield | 38 |
| 5 | Clay | 90 |
| 6 |  Runway | 33 |
| 7 | Salesforce | 87 |
| 8 | Codex | 79 |
| 9 | Lovable | 57 |
| 10 | Cursor (Anysphere) | 78 |
Marketing built on real people, earned trust and experiences AI cannot easily copy.
Flat is the story, and the flatness is the point. Share has held between 3.3% and 4.8% every month since February, with the L30 up 5.2% and no spike behind it. It runs on recurring complaints and proof points: alinav's list of 2026 marketing contradictions, Eoghan McCabe crediting Fin's brand awareness win to authentic superiority, Samantha McKenna on company pages losing to people.
Posts to understand this
Inside Human Experience
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Claude Code | 99 |
| 3 | HubSpot | 83 |
| 4 | Salesforce | 87 |
| 5 | Codex | 79 |
| 6 | Semrush | 49 |
| 7 | Grok Bot | 57 |
| 8 | Nooks | 46 |
| 9 | Ahrefs | 49 |
| 10 | Clarify | 29 |
How AI changes software pricing, usage costs and gross margins.
Share climbed from February to a June peak, then slipped in July and fell in August to roughly where it started the year. September is only 12 days in, so it is not a finished month, but the last 30 days are down about a third. The cost story moved from vendor price cuts to buyers cutting their own bills.
Posts to understand this · as of Sep 12
Inside AI Pricing & Margins
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | OpenRouter | 51 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Claude Code | 99 |
| 5 | Clay | 90 |
| 6 | HubSpot | 83 |
| 7 | Vercel | 58 |
| 8 | Codex | 79 |
| 9 | Replit | 54 |
| 10 |  Twilio | 37 |
How AI changes what software companies build, charge for and deliver.
This cluster halved from February and sits 26% below its 90-day mark, with the last 30 days up slightly. The argument did not stop, it got more specific. Eoghan McCabe doubles down on Intercom's help desk while Fin grows, Rob Litterst traces monday.com's 18-month path to credit pricing, Jake Dunlap costs out replacing Gong and Salesforce.
Posts to understand this · as of Sep 12
Inside Software Business Model Disruption
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | HubSpot | 83 |
| 4 | Claude Code | 99 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | ZoomInfo | 52 |
| 7 | Replit | 54 |
| 8 | Codex | 79 |
| 9 | Intercom | 46 |
| 10 | Twilio | 37 |
How teams track and control the cost of models, tokens and agent work.
This more than doubled between February and July, 1.6% to 3.6%, then cooled: L30 down 31%, L90 still up 28%. Tooling drove the rise, not theory. Rippling shipped a console mapping AI spend to teams and pull requests, Cognition's Devin Fusion routes work to a smaller sidekick agent and cut cost 35%.
Posts to understand this · as of Sep 8
Inside Token Spend & Cost Controls
| Brand | GTM Index | |
|---|---|---|
| 1 | OpenRouter | 51 |
| 2 | Claude Code | 99 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Codex | 79 |
| 5 | Salesforce | 87 |
| 6 | Vercel | 58 |
| 7 | Clay | 90 |
| 8 |  GitHub Copilot | 35 |
| 9 | Replit | 54 |
| 10 | Grok Bot | 57 |
Software that performs SDR work, and how it changes the human SDR role.
Share peaked near 3.7% in March, held in April, then slid to 1.5% by July. The last 30 days are up 23% off that floor, putting share back at 1.8%, about half the spring peak. April's posts are a standalone launch and the pushback on it. By June the same pitch comes from Salesforce.
Posts to understand this · as of Sep 12
Inside AI SDR
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | Claude Code | 99 |
| 4 | HubSpot | 83 |
| 5 | ZoomInfo | 52 |
| 6 | Lovable | 57 |
| 7 | Outreach | 44 |
| 8 | Zapier | 52 |
| 9 |  11x | 25 |
| 10 | Airtable | 48 |
How people and agents share work, hand off tasks and make decisions together.
The May high came from two loud moments: Dust's argument that single-player AI traps its gains inside one person's instance, and StackAI joining Asana to build an operating system for human-agent execution. Neither started a durable run. Share has fallen every month since, down 14.7% over the last 30 days. The framing landed, the conversation moved on.
Posts to understand this
Inside Multi-Player AI
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Claude Code | 99 |
| 3 | Salesforce | 87 |
| 4 | HubSpot | 83 |
| 5 | OpenClaw | 50 |
| 6 | Clay | 90 |
| 7 | Cursor (Anysphere) | 78 |
| 8 |  Buffer | 28 |
| 9 | Grok Bot | 57 |
| 10 | LangChain | 43 |
Growth driven by people using the product before they talk to sales.
PLG has more than halved since February and sits down 27% over 90 days. The idea is not out of favor, it has folded into other people's stories: Elena Verna at Lovable through a $400m round, Dharmesh Shah going back to selling HubSpot himself, and Anthropic closing 54% of new enterprise logos without a salesperson.
Posts to understand this · as of Sep 12
Inside Product-Led Growth
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Claude Code | 99 |
| 3 | Codex | 79 |
| 4 | HubSpot | 83 |
| 5 | Clay | 90 |
| 6 | Cursor (Anysphere) | 78 |
| 7 | Intercom | 46 |
| 8 | Canva | 48 |
| 9 |  Calendly | 37 |
| 10 | Instantly | 61 |
How revenue teams give agents the knowledge, memory and context they need to work.
Peaked in June, down since, with the last 30 days off about a third. The June high was product news, led by Perplexity's Brain memory system for Computer. What followed is quieter and more operational: Databricks selling a context engineering certification, and practitioners describing the in-house GTM engines they wired to Snowflake and Gong.
Posts to understand this · as of Sep 8
Inside Context Engineering in GTM
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | Salesforce | 87 |
| 4 | Codex | 79 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | HubSpot | 83 |
| 7 |  Sumble | 32 |
| 8 | Grok Bot | 57 |
| 9 | Lovable | 57 |
| 10 |  Attention | 32 |
The choice between building software internally and buying another product.
The April peak at 2.8 percent came from buyers saying it out loud, like Zipline's Mallory Lee telling Salesforce her team is trying to use it less. July was the floor at 1.2. Benioff's August quarter post, with attrition near an all-time low, is the loudest answer back. L30 is up 17.8 percent off that floor.
Posts to understand this · as of Sep 12
Inside Build vs. Buy
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | Claude Code | 99 |
| 4 | HubSpot | 83 |
| 5 | ZoomInfo | 52 |
| 6 | Airtable | 48 |
| 7 | Cursor (Anysphere) | 78 |
| 8 | Replit | 54 |
| 9 | Clarify | 29 |
| 10 | Codex | 79 |
Low-quality AI content and the fight to make automated work worth reading.
No single event moves this one. Share has drifted down all year, from 2.4% in February to 1.8% in August, and the L30 is off 37.6%. What is left is argument about the policing rather than the slop: Ann Handley on detection tools turning writers against each other, Lulu Cheng Meservey naming who still writes their own posts.
Posts to understand this
Inside AI Slop
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | HubSpot | 83 |
| 3 | Clay | 90 |
| 4 | Substack | 51 |
| 5 | Salesforce | 87 |
| 6 | Codex | 79 |
| 7 |  Optimizely | 37 |
| 8 | Semrush | 49 |
| 9 | Airtable | 48 |
| 10 | Clarify | 29 |
The argument that agents and custom software will break the classic SaaS model.
Share peaked at 3.3 percent in February and slid through spring and summer, holding flat from April to May, then bottoming at 1.05 percent in August, the last completed month. The last 30 days are close to flat, down 1.5 percent. What is left of the conversation is mostly rebuttal, from a vendor, a buyer, and payments data.
Posts to understand this · as of Sep 12
Inside SaaSpocalypse
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | HubSpot | 83 |
| 4 | Claude Code | 99 |
| 5 | ZoomInfo | 52 |
| 6 | Nooks | 46 |
| 7 | Sumble | 32 |
| 8 | Airtable | 48 |
| 9 | Clarify | 29 |
| 10 | Codex | 79 |
Ways to verify whether digital content or online activity came from a person or AI.
Regulation made August. Anthropic's watermarking FAQ pointed at the EU AI Act, Synthesia marked every relevant video with C2PA credentials as Article 50 and California's transparency act came into force, and Lily Ray turned compliance into a practitioner test. Share is flat over the last 30 days, which means the spike held rather than faded.
Posts to understand this · as of Sep 12
Inside Proof of Human
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Codex | 79 |
| 3 | Substack | 51 |
| 4 | Cursor (Anysphere) | 78 |
| 5 | HubSpot | 83 |
| 6 | Clay | 90 |
| 7 | ElevenLabs | 49 |
| 8 | Grok Bot | 57 |
| 9 | OpenRouter | 51 |
| 10 | Salesforce | 87 |
AI tools that write, edit, test and operate software alongside a human builder.
Down 24.7% over 90 days from a March high near 21.5%, but up 4.3% in the last 30. The category stopped being news and became infrastructure. Notion defines a workspace in TypeScript, ElevenLabs put its whole API in the terminal, ColdIQ runs ad spend from Claude Code. The floor near 14% has held since July.
Posts to understand this · as of Sep 8
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Codex | 79 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Grok Bot | 57 |
| 5 | Clay | 90 |
| 6 | Lovable | 57 |
| 7 | Replit | 54 |
| 8 | HubSpot | 83 |
| 9 | Salesforce | 87 |
| 10 | OpenClaw | 50 |
The frontier models and labs that provide the base layer for AI products.
Three labs kept this loud on price and capability. Anthropic's Project Glasswing and the Mythos preview drew the biggest single response of the year here. Then Sonnet 5's introductory pricing went permanent in August, and Grok 4.6 arrived the same week at half the price of comparable frontier models. August cooled off, and the L30 move is back up about 9%.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Claude Code | 99 |
| 4 | Grok Bot | 57 |
| 5 | OpenClaw | 50 |
| 6 | OpenRouter | 51 |
| 7 | HubSpot | 83 |
| 8 | Salesforce | 87 |
| 9 | Vercel | 58 |
| 10 | Clay | 90 |
An open protocol that lets agents connect to a product's data, tools and actions.
MCP's share more than doubled from 3.5% in February to a 7.8% peak in May, then slipped each month through August. September is only eight days in, so read the last 30 days: down 5.5%. The posts underneath are all shipping notices. Vendors are wiring MCP into products, gateways, and orchestration layers.
Posts to understand this · as of Sep 8
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | Salesforce | 87 |
| 4 | Codex | 79 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | HubSpot | 83 |
| 7 | Instantly | 61 |
| 8 | Prospeo | 51 |
| 9 | Grok Bot | 57 |
| 10 | Lemlist | 50 |
People asking AI assistants for answers instead of using a traditional search engine.
The June launches did the lifting. HubSpot's AEO product and Jasper's GEO agent both pitched brands on being findable inside AI answers, and June is where the curve tops out. Since then the argument has moved to measurement, like Supermetrics splitting visibility into four countable metrics. L30 is down about 12%.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Semrush | 49 |
| 2 | HubSpot | 83 |
| 3 | Claude Code | 99 |
| 4 | Ahrefs | 49 |
| 5 | Profound | 50 |
| 6 | Webflow | 46 |
| 7 | AirOps | 36 |
| 8 | Optimizely | 37 |
| 9 | Substack | 51 |
| 10 | Codex | 79 |
The practice of helping a brand appear, earn citations and get recommended inside AI answers.
May was the high point, and the argument has turned since. Eli Schwartz quotes Google's John Mueller saying there is no GEO or AEO without SEO fundamentals. Wil Reynolds says once a category settles on four or five brands, hacks cannot break in. Down 9.6% in the last 30 days.
Posts to understand this · as of Sep 8
| Brand | GTM Index | |
|---|---|---|
| 1 | Ahrefs | 49 |
| 2 | Profound | 50 |
| 3 | Semrush | 49 |
| 4 | Clay | 90 |
| 5 | Webflow | 46 |
| 6 | HubSpot | 83 |
| 7 | Claude Code | 99 |
| 8 | AirOps | 36 |
| 9 | Salesforce | 87 |
| 10 | Codex | 79 |
Email sent to people with no prior relationship to start a sales conversation.
Cold email share peaked near 5.5 percent in May, fell back to about 3.2 percent by July, and held there through August. The last 30 days are up about 20 percent, still well under the spring high. The talk now is craft and plumbing: what actually earns a reply, and where the mail sends from.
Posts to understand this · as of Sep 12
inRelevance gapArgues personalized openers are observations, not reasons for a prospect to reply.WA
inPushbackOutbound veteran tells builders of one-shot AI email loops to temper the math.W
inSending infrastructureVendor argues spreading mailboxes across providers limits the damage from one blacklisting.How large organizations adopt AI, move beyond pilots and change how work gets done.
A slow fade, down roughly 19% over ninety days and 22% over the last thirty. The posts show why: the subject moved from adoption to plumbing. Mistral shipped Workflows for production orchestration, Databricks made Unity AI Gateway generally available for governing an estate of agents, and the risk talk turned to who owns the decision.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Salesforce | 87 |
| 3 | Codex | 79 |
| 4 | Lovable | 57 |
| 5 | OpenRouter | 51 |
| 6 | Clay | 90 |
| 7 | Vercel | 58 |
| 8 | HubSpot | 83 |
| 9 | Replit | 54 |
| 10 | AirOps | 36 |
The work of retaining customers, helping them succeed and expanding the relationship.
Down 23% across 90 days but flat since July, and the flatness is the point. These posts are not about growth, they are about where CS belongs. Vanta put success, account management and SDR under one sales leader. Mark Kosoglow argues CSMs should not carry revenue at all. Same question, opposite answers.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | HubSpot | 83 |
| 4 | Claude Code | 99 |
| 5 | Cursor (Anysphere) | 78 |
| 6 |  Salesloft | 42 |
| 7 | Twilio | 37 |
| 8 | Intercom | 46 |
| 9 | Replit | 54 |
| 10 | Gong | 62 |
Measuring whether AI answers mention, cite and recommend a brand.
Share peaked at 4.07% in June, fell to 3.41% in July, and has sat near 3.5% since. The last 30 days are down 4.4%, so the topic is cooling off that peak rather than holding it. The current posts show where the argument went: past dashboards, into whether AI visibility work produces business results.
Posts to understand this · as of Sep 8
| Brand | GTM Index | |
|---|---|---|
| 1 | Semrush | 49 |
| 2 | Ahrefs | 49 |
| 3 | Profound | 50 |
| 4 | Claude Code | 99 |
| 5 | HubSpot | 83 |
| 6 | Peec AI | 28 |
| 7 |  Ranked AI | 35 |
| 8 | AirOps | 36 |
| 9 | Docket | 28 |
| 10 | Salesforce | 87 |
The team that runs go-to-market systems, processes, data and reporting across revenue functions.
RevOps has slid all year and sits 26% below its 90-day mark, with the last 30 days flat. No single event explains it. What lands now is the job changing name and shape: Andy Mowat joining Inflection as CMO after running RevOps at four unicorns, xAI posting a GTM systems and agents role, Davide Grieco telling founders they are not ready for a GTM engineer.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | HubSpot | 83 |
| 4 | Claude Code | 99 |
| 5 | Attio | 50 |
| 6 | Gong | 62 |
| 7 | Codex | 79 |
| 8 | Zapier | 52 |
| 9 | n8n | 59 |
| 10 | Apollo.io | 54 |
Finding, researching and qualifying the people and accounts a sales team should contact.
The level stepped down in June, from roughly 4% to under 3%, and has stayed there. Launches no longer move it. Origami's automatic lead finder in April, HubSpot's dollar-per-qualified-lead agent pricing in May, and Clay's rebuilt sequencer this month all landed into the same flat line. Up 4% over 30 days.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Claude Code | 99 |
| 3 | Prospeo | 51 |
| 4 | Salesforce | 87 |
| 5 | HubSpot | 83 |
| 6 | Instantly | 61 |
| 7 | Apollo.io | 54 |
| 8 | n8n | 59 |
| 9 | Apify | 53 |
| 10 | Lemlist | 50 |
Voice and chat software that talks with buyers and customers.
This one moves in steps, not a trend. June dipped, July and August ran near 2.9%, and the last 30 days are down 16% while the 90-day read is still up 15%. The posts are deployment announcements rather than launches: ElevenLabs with Genesys, Sierra with Fiserv, Cartesia with Parallel for live web search.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 |  AssemblyAI | 34 |
| 2 | beehiiv | 47 |
| 3 | Salesforce | 87 |
| 4 | Codex | 79 |
| 5 | Docket | 28 |
| 6 | Cursor (Anysphere) | 78 |
| 7 | Claude Code | 99 |
| 8 | Clay | 90 |
| 9 | ElevenLabs | 49 |
| 10 | Grok Bot | 57 |
Marketing programs that create and capture demand, then turn it into pipeline.
Demand generation is down 26% over 90 days, with the last 30 up 5%. The conversation moved from running programs to questioning them. Joe Chernov says founders no longer hire demand first, Brigitta Ruha says Meta is being trained to find form fillers instead of buyers, and Webflow launched Campaigns for the same audience.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | HubSpot | 83 |
| 3 | Salesforce | 87 |
| 4 | Webflow | 46 |
| 5 | Docket | 28 |
| 6 | RB2B | 48 |
| 7 | Claude Code | 99 |
| 8 | Optimizely | 37 |
| 9 | Instantly | 61 |
| 10 |  Vector | 42 |
Cold calls, messages and coordinated outreach used to create new sales opportunities.
Cold outbound holds a narrow band all year and drifts down, off 11% in the last 30 days. No launch resets it. Nooks extended its dialer into AI sequencing in February, Armand Farrokh praised Gong for a five-minute human follow-up, and Rippling's story is 150 SDRs working the same accounts harder.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Claude Code | 99 |
| 3 | Instantly | 61 |
| 4 | Prospeo | 51 |
| 5 | Gong | 62 |
| 6 | HeyReach | 52 |
| 7 | Nooks | 46 |
| 8 | Salesforce | 87 |
| 9 | Apollo.io | 54 |
| 10 | HubSpot | 83 |
Buying attention through paid search, paid social, display and other advertising channels.
Paid media share bottomed at 1.8% in April, climbed back through the summer to 2.5% in August, still under February's 2.8%. The last 30 days are down about 11%, so September is easing off that peak. Two of the posts date to winter; the newest is one marketer's case study of paid search share falling inside one company's CRM.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | HubSpot | 83 |
| 3 | Claude Code | 99 |
| 4 | Codex | 79 |
| 5 | Instantly | 61 |
| 6 | Salesforce | 87 |
| 7 | beehiiv | 47 |
| 8 | HeyReach | 52 |
| 9 | Lemlist | 50 |
| 10 | RB2B | 48 |
The technical work that keeps outbound email out of spam and inside the inbox.
Deliverability never spikes, it returns. The curve has bounced between roughly 1 and 2.2 percent all year with no defining event, carried by stack posts like Brigitta Ruha's waterfall verification math. The recent lift comes from people staffing and betting on it, including Resend hiring a deliverability engineer. Up 39% over 30 days.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Prospeo | 51 |
| 3 | Instantly | 61 |
| 4 | HubSpot | 83 |
| 5 | Claude Code | 99 |
| 6 | Salesforce | 87 |
| 7 | Lemlist | 50 |
| 8 | Smartlead | 44 |
| 9 | Codex | 79 |
| 10 | n8n | 59 |
Programmable tools used to build custom go-to-market systems, agents and automations.
Share climbed from under 1% in February to a 2.59% peak in May, then fell three straight months to 1.42% in August. September is still partial, so the current read is the last 30 days, up 50%. The selected posts sit along that run, from Eric Nowoslawski's April skills repo to Tim Yakubson's August giveaway.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Prospeo | 51 |
| 2 | Clay | 90 |
| 3 | Instantly | 61 |
| 4 | Claude Code | 99 |
| 5 | HeyReach | 52 |
| 6 | HubSpot | 83 |
| 7 | Lemlist | 50 |
| 8 | Apollo.io | 54 |
| 9 | n8n | 59 |
| 10 |  FullEnrich | 42 |
Tests and benchmarks used to judge whether a model or agent works well enough.
Share climbed from February to a July high near 3%, then gave ground. August fell back and the last 30 days are down 38.9%, while the trailing 90 days are still up 35%. These posts show where evals sit now: inside company build loops, inside enterprise workflows, and inside vendors' own benchmarks.
Posts to understand this · as of Sep 8
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Claude Code | 99 |
| 4 | OpenRouter | 51 |
| 5 | Salesforce | 87 |
| 6 | Cline | 24 |
| 7 | Grok Bot | 57 |
| 8 | Vercel | 58 |
| 9 | Clay | 90 |
| 10 | Baseten | 44 |
Adding and verifying contact, company and buying-signal data used by go-to-market teams.
Enrichment held above 3 percent through June and then dropped hard, down 37 percent over 90 days. The argument moved rather than died. Adam Robinson's post about Clay's largest user replacing 17.3 million weekly calls with a $200 Claude Code plan, and Maja Voje's agent-ready CRM, both put the work upstream of the enrichment vendor.
Posts to understand this · as of Sep 12
The warehouses, pipelines and data models that keep customer and revenue data usable.
A long, even slide from 4.7% in February to about 2% now, with no single event bending it. Real moments landed along the way: Pocus joining Apollo in March, Rippling Data Cloud in June, Clay shipping Workflows in August. The plumbing keeps shipping while the argument about it moves elsewhere. Down 26% over 30 days.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | HubSpot | 83 |
| 4 | Claude Code | 99 |
| 5 | Gong | 62 |
| 6 |  GrowthLoop | 28 |
| 7 | Zapier | 52 |
| 8 | Vercel | 58 |
| 9 |  ActiveCampaign | 29 |
| 10 | Attio | 50 |
Using dinners, meetups, conferences and field events to build demand and trust in person.
Share dipped to 0.8 percent in April, then climbed back through the summer to 1.66 percent in August. The last 30 days are up 23 percent, measured against that mid-July to mid-August stretch, not against the April low. September is only 12 days in, so read the 30-day move rather than the month.
Posts to understand this · as of Sep 12
inPeak scaleDatabricks opens its summit with 30,000-plus attendees and 800-plus sessions.AF
inTurnout mechanicsVendors hosting dinners in San Francisco got poor RSVPs when the conference sat elsewhere.N
inSponsor spendNetlify sponsors International Generalist Day: 100-plus free meetups across 40-plus countries.| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Salesforce | 87 |
| 3 | Claude Code | 99 |
| 4 | Codex | 79 |
| 5 | HubSpot | 83 |
| 6 | Vercel | 58 |
| 7 |  Chili Piper | 32 |
| 8 | Clarify | 29 |
| 9 | Docket | 28 |
| 10 | Grok Bot | 57 |
Tools and workflows for generating video, avatars and synthetic presenters with AI.
Share has held in a narrow band, roughly 1.3 to 2.3 percent, since February. The highest month is June at 2.276, and the loudest June post is a creator handing out 1,000 images, not a vendor launch. August sat at 2.203 with a routine model release. The last 30 days are down 18.7 percent.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Midjourney | 40 |
| 2 | Higgsfield | 38 |
| 3 | HubSpot | 83 |
| 4 | Runway | 33 |
| 5 | Claude Code | 99 |
| 6 | Lovable | 57 |
| 7 | Salesforce | 87 |
| 8 | Cursor (Anysphere) | 78 |
| 9 | Codex | 79 |
| 10 | Grok Bot | 57 |
Selling large, complex deals through long cycles, procurement, security review and many stakeholders.
Down 22 percent over 90 days, but August pushed back to 1.69 percent on craft posts: Jen Abel's fifteen-step process via Lenny Rachitsky, and Ryan Walsh charting 80 sales orgs on deal size against cycle length. L30 is up 12.7 percent. When enterprise sales rises here, it rises on process detail, not on tools.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Clay | 90 |
| 4 | HubSpot | 83 |
| 5 | Claude Code | 99 |
| 6 | Grok Bot | 57 |
| 7 | Lovable | 57 |
| 8 | Gong | 62 |
| 9 | Attention | 32 |
| 10 | Attio | 50 |
The training, fine-tuning and modeling work underneath AI products.
This is the line moving up over 90 days, and no single release explains it. Skild's S1 one-demo robot model, Poolside's open-weight Laguna S 2.1, and Together AI's 10,000-GPU build with Larsen & Toubro in Chennai are three different stories carrying the same topic. The last 30 days are flat, so the rise is holding rather than accelerating.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Claude Code | 99 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Salesforce | 87 |
| 5 |  Groq | 25 |
| 6 | OpenClaw | 50 |
| 7 | OpenRouter | 51 |
| 8 | ElevenLabs | 49 |
| 9 | Exa.ai | 43 |
| 10 |  Guru | 21 |
Sales and marketing coordinating around a defined set of target accounts.
The February high came from people and product news at once: Kelly Hopping moving from Demandbase to 6sense as CMO, and Clay opening ads inside the platform. Nothing since has matched that, and share is down about a quarter over 90 days. The ABM posts that land now are agent builds, not platform bets.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | HubSpot | 83 |
| 3 | Salesforce | 87 |
| 4 | Claude Code | 99 |
| 5 |  6sense | 39 |
| 6 |  Demandbase | 31 |
| 7 | Apify | 53 |
| 8 | Vector | 42 |
| 9 | ZoomInfo | 52 |
| 10 | FullEnrich | 42 |
The choice of go-to-market motion and the plan for making it work.
The spring peak did the damage. March ran at 3.2% behind big agent-run GTM posts, Amos Bar-Joseph claiming $1.5M monthly pipeline with no SDRs and majavoje packaging a whole methodology into 12 Claude Code skills. Nothing has replaced those, so share settled near 1.6% and the L30 is flat at plus 0.8%.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Claude Code | 99 |
| 3 | HubSpot | 83 |
| 4 | Salesforce | 87 |
| 5 | Grok Bot | 57 |
| 6 | HeyReach | 52 |
| 7 | Lemlist | 50 |
| 8 | 6sense | 39 |
| 9 | Clarify | 29 |
| 10 | Cursor (Anysphere) | 78 |
The work of connecting marketing activity to pipeline and revenue.
No event moves this line. It drifts down steadily, and the posts say where the subject went: Rand Fishkin and Amanda Natividad's zero-click book, Ethan Smith's n8n study of conversions that never produce a click, Kieran Flanagan folding attribution models into a paid and AEO playbook. Attribution now shows up inside AI search talk rather than on its own.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Salesforce | 87 |
| 3 | Claude Code | 99 |
| 4 | Clay | 90 |
| 5 | Docket | 28 |
| 6 | Intercom | 46 |
| 7 | GrowthLoop | 28 |
| 8 | 6sense | 39 |
| 9 | ActiveCampaign | 29 |
| 10 | Apollo.io | 54 |
The positioning, messaging, launches, competitive work and sales tools owned by product marketing.
Share has roughly halved since February, from 2.8% to about 1.3%, and the April marketing engineer wave did not stop the slide. Profound introduced the role and a Google marketing exec said he was hiring one, in a month that still fell. Recent volume is craft work like Aatir Abdul Rauf's PMM lessons. L30 is up 3.5%.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Claude Code | 99 |
| 3 | Clay | 90 |
| 4 | Gong | 62 |
| 5 | HubSpot | 83 |
| 6 | Canva | 48 |
| 7 | Semrush | 49 |
| 8 | Lovable | 57 |
| 9 | Profound | 50 |
| 10 | beehiiv | 47 |
Planning, producing and distributing useful content to attract and educate buyers.
Flat over 30 days and down about a quarter over 90, with no launch behind either move. The volume comes from practitioners arguing with each other: whether AI can do content without someone feeding it judgment, whether one LinkedIn post beats $300k of paid registrations, and Kyle Lacy's push toward third-party voices over in-house production.
Posts to understand this · as of Sep 12
Mapping and winning the committee of people involved in a business purchase.
The steepest recent drop in this set, down about 24% over 30 days. Nothing broke; the framework conversation thinned and what is left is specific work. Anthony Pierri argues homepages get written for executives who never read them, and Becca Lindquist describes the Slack channel Clay uses to pull executives into accounts.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | HubSpot | 83 |
| 3 | Clay | 90 |
| 4 | Claude Code | 99 |
| 5 | Gong | 62 |
| 6 | Instantly | 61 |
| 7 |  Customer.io | 37 |
| 8 | HeyReach | 52 |
| 9 | Vector | 42 |
| 10 | Apollo.io | 54 |
Predicting revenue from the pipeline and managing the commitments behind that forecast.
Forecasting stepped down after February, from 2.6% of share to 1.7% in March, and has stayed in a narrow band since. May and July both rose, August fell, so there is no direction after that one drop. September is only twelve days in. The last 30 days moved -2.8%, essentially flat near 1.3%.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | HubSpot | 83 |
| 3 |  Clari | 34 |
| 4 | Claude Code | 99 |
| 5 | Clay | 90 |
| 6 | Codex | 79 |
| 7 | Gong | 62 |
| 8 |  Rox | 30 |
| 9 | Cursor (Anysphere) | 78 |
| 10 | Gamma | 42 |
Starting and running a business, including the economics and experience of the founder path.
Entrepreneurship never spikes and never leaves. It has held near one percent of the conversation all year, dipped in July, and is up 44% over the last 30 days. What fills it rotates: Greg Isenberg calling the "GPT wrappers are useless" line the biggest lie of the past year, Nick Velkovski's bootstrapped run to $17m ARR, founder origin stories.
Posts to understand this · as of Sep 8
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Claude Code | 99 |
| 3 | Salesforce | 87 |
| 4 | Codex | 79 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | Lovable | 57 |
| 7 | Substack | 51 |
| 8 | Grok Bot | 57 |
| 9 | HubSpot | 83 |
| 10 | Instantly | 61 |
The debate over AI risk, alignment, regulation and responsible development.
AI Safety is up 31% over 90 days, and the lift came from labs shipping rather than arguing. Anthropic's Model Hardware Standard preview pulled the biggest reaction, with Genentech and QuEra results attached. Mistral's open-weights Shieldstral and Microsoft's 2026 transparency report kept the volume up. September is running high with only 12 days observed.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Claude Code | 99 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Grok Bot | 57 |
| 5 | OpenClaw | 50 |
| 6 | Vercel | 58 |
| 7 |  Frontier | 21 |
| 8 | Lovable | 57 |
| 9 | OpenRouter | 51 |
| 10 | Salesforce | 87 |
How sales leaders hire, organize, coach and manage teams against a revenue target.
The spring conversation was about who to hire, with Ghazi Masood rejecting three quarters of VP of Sales applicants for not wanting grunt work. From June on, share has sat about a third lower and the posts have narrowed to pipeline before headcount and building an AI-native org. Down 36% over 30 days.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Salesforce | 87 |
| 3 | Clay | 90 |
| 4 | Replit | 54 |
| 5 | Attention | 32 |
| 6 | Codex | 79 |
| 7 | Cursor (Anysphere) | 78 |
| 8 | Gong | 62 |
| 9 | Outreach | 44 |
| 10 | Twilio | 37 |
The work of earning awareness, trust and preference for a company over time.
Share dipped in March, climbed back to 1.85% by June, then fell through July and August. The spring posts argue about what brand strategy even is and how credibility gets built. The August post puts brand up against a quarterly budget test it cannot pass. The last 30 days are down 3.8%, so the slide has slowed, not turned.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Semrush | 49 |
| 3 | Clay | 90 |
| 4 | Gong | 62 |
| 5 | Salesforce | 87 |
| 6 | AirOps | 36 |
| 7 | Claude Code | 99 |
| 8 | Lovable | 57 |
| 9 | 6sense | 39 |
| 10 | Ahrefs | 49 |
How companies create demand, build trust and reach buyers at other businesses.
B2B Marketing is down 23% over the last 30 days and 36% over 90. No launch or campaign explains the fall, and the decline runs straight through from March. The posts still landing argue about the function itself: Kieran Flanagan's modern playbook, Jon Miller on the CMO tax, Melissa Rosenthal on the company blog as a lease.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | HubSpot | 83 |
| 3 |  CHEQ | 15 |
| 4 | Demandbase | 31 |
| 5 |  TechTarget | 23 |
| 6 | 6sense | 39 |
| 7 | Customer.io | 37 |
| 8 |  HG Insights | 27 |
| 9 |  Influ2 | 23 |
| 10 | Optimizely | 37 |
How companies sell to other companies and how business buyers make decisions.
A plateau, and the plateau is the point. Nothing launches here, so the topic holds a steady share on craft and diagnosis instead of events: Morgan Ingram on reps working the booth, Tyrona Heath on creator credibility, Lillian Pierson on 83% of opportunities ending in no decision. Up about 6% over the last 30 days.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | HG Insights | 27 |
| 4 | Instantly | 61 |
| 5 | Lovable | 57 |
| 6 | Semrush | 49 |
| 7 | Supabase | 45 |
| 8 |  TrustRadius | 26 |
| 9 |  Amazon Bedrock AgentCore | 22 |
| 10 | Apify | 53 |
Agents that prospect, enrich data, send outreach and update revenue systems.
Revenue Agents is down 52% over 90 days, off a spring run above 3%. That run was announcement-driven: Intercom giving Fin a specialized sales role, HockeyStack raising $50M for a revenue agent. Once those cleared, the talk moved to who does the work, like Nico Druelle reading Anthropic's GTM engineer and applied AI engineer specs against each other.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | HubSpot | 83 |
| 4 | Salesforce | 87 |
| 5 | Rox | 30 |
| 6 | Codex | 79 |
| 7 | Sumble | 32 |
| 8 | Attention | 32 |
| 9 | Clarify | 29 |
| 10 |  Clearskies | 21 |
Recording and analyzing customer calls to improve coaching, deal execution and revenue decisions.
A long slide from February, down 19% in the last 30 days and 25% over 90. The work did not stop, it moved: one seller ran lost discovery calls through Claude to build archetypes, and Ryan Staley scheduled call prep in Cowork. Observe.AI's G2 badges are one of the few posts still treating this as its own category.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Gong | 62 |
| 2 | Salesforce | 87 |
| 3 | Claude Code | 99 |
| 4 | HubSpot | 83 |
| 5 | Clay | 90 |
| 6 | Instantly | 61 |
| 7 | ZoomInfo | 52 |
| 8 | 6sense | 39 |
| 9 |  AI Ark | 43 |
| 10 | Apollo.io | 54 |
The software stack marketers use to plan, run, measure and improve their work.
No launch drives this one. The March high came from argument: Scott Brinker's Databricks report asking whether the stack needs a different architecture, and Joe Chernov comparing agent unbundling to cord-cutting. Share dipped through the summer and picked up again in August as Dreamforce sessions were announced. The curve tracks reports and conference stages, not products.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | GrowthLoop | 28 |
| 3 | Codex | 79 |
| 4 | Docket | 28 |
| 5 | HubSpot | 83 |
| 6 |  Augment Code | 17 |
| 7 | Claude Code | 99 |
| 8 |  Gradial | 21 |
| 9 | Intercom | 46 |
| 10 | Twilio | 37 |
Command-line tools that let people and agents run go-to-market work without a dashboard.
CLIs for GTM went from 0.07% in February to above 1% by summer, up 37% over 90 days with the last 30 easing. Clay shipping a CLI and API is the anchor. Jordan Crawford, who left Clay for Claude Code, says the missing CLI is now fixed, and Felix Frank reads the launch as proof the terminal is standard.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Claude Code | 99 |
| 3 | Codex | 79 |
| 4 | Cursor (Anysphere) | 78 |
| 5 | HubSpot | 83 |
| 6 | Salesforce | 87 |
| 7 | Apify | 53 |
| 8 | Grok Bot | 57 |
| 9 | Instantly | 61 |
| 10 | Nooks | 46 |
How marketing teams use AI, where it helps and how it changes the job.
No single event drives this one. The topic slid through the first half of the year and has held near one percent since, with the last 30 days up 25 percent. What carries it is jobs and method: Andy Mowat taking the CMO seat at Inflection, Tom Wentworth arguing SaaStr's own events business proves marketing is not replaceable, Liza Adams sorting AI work into faster, better, different.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Claude Code | 99 |
| 3 | Profound | 50 |
| 4 | Buffer | 28 |
| 5 | LangChain | 43 |
| 6 | Salesforce | 87 |
| 7 | 1mind | 37 |
| 8 | ActiveCampaign | 29 |
| 9 | Airtable | 48 |
| 10 | Calendly | 37 |
How buyers search and how marketers plan, measure and improve organic discovery.
Search Strategy is still up 17% over 90 days, but the last 30 are down about 12%, so the June high has cooled. The argument turned from claiming AI visibility to testing it. Gaetano DiNardi covers Otterly's experiment where 2,000 slop pages were de-indexed by Google while LLM citations kept climbing, and Eli Schwartz watches brands come back for abandoned SEO traffic.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Semrush | 49 |
| 2 | Ahrefs | 49 |
| 3 | HubSpot | 83 |
| 4 | Ranked AI | 35 |
| 5 | AirOps | 36 |
| 6 | Substack | 51 |
| 7 | Amazon Bedrock AgentCore | 22 |
| 8 |  AthenaHQ | 18 |
| 9 | Claude Code | 99 |
| 10 | Clay | 90 |
Choosing which model should run each task based on cost, quality, speed or risk.
Model routing sat below 0.4 percent share from February through May. June more than doubled it to 0.86, July peaked near 1.0, and August held at 0.94. The last 30 days are up under 3 percent, so this is a plateau near the top rather than a fresh climb.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | OpenRouter | 51 |
| 2 | Claude Code | 99 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Codex | 79 |
| 5 | Vercel | 58 |
| 6 | Clay | 90 |
| 7 | GitHub Copilot | 35 |
| 8 | ElevenLabs | 49 |
| 9 | Replit | 54 |
| 10 | Runway | 33 |
The durable knowledge and memory an agent can use across tasks and sessions.
June was the high, and Perplexity's Brain launch explains it: a continuously learning memory system with numbers attached, including 25 percent better correctness on tasks that need past context. The line has roughly halved since and the last 30 days are down 40 percent. What is left is definitional, like Writer's Palmyra X6 pitch and Neo4j separating knowledge graphs from context graphs.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | Salesforce | 87 |
| 4 | Codex | 79 |
| 5 | HubSpot | 83 |
| 6 | Sumble | 32 |
| 7 | Attention | 32 |
| 8 | Clarify | 29 |
| 9 | Clearskies | 21 |
| 10 | Cursor (Anysphere) | 78 |
The team that owns marketing systems, campaign execution, data and reporting.
March was the high point. Since then a long drift down, with no single event behind it. The pull now is jobs and reframes rather than tools. NVIDIA is out hiring a director of strategic marketing operations, and Scott Brinker's tech-versus-org gap argument landed in late August. The last 30 days moved up 27%.
Posts to understand this · as of Sep 7
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Docket | 28 |
| 3 | GrowthLoop | 28 |
| 4 | Clay | 90 |
| 5 | HubSpot | 83 |
| 6 | 1mind | 37 |
| 7 | Gradial | 21 |
| 8 | Augment Code | 17 |
| 9 | Lovable | 57 |
| 10 | ActiveCampaign | 29 |
Reusable instructions that teach an agent how to complete a specific kind of task.
Share climbed from February to a May peak near 1.7 percent, then fell through June, July and August. The last 30 days are down about 30 percent, so the fall is still running. September is only 12 days in, so it is not a completed month yet.
Posts to understand this · as of Sep 12
Training, coaching, playbooks and content that help sales reps perform better.
A long erosion rather than a drop. Sales enablement has lost more than half its share since February and the last 30 days are down 26 percent. No launch is holding it up. What remains is the argument about why it is not working: Sam Jacobs on quota attainment falling to 20 percent against 7x more spend, kddorsey3 on the coaching gap, Stevie Case on Hyperbound at Vanta.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 |  Highspot | 28 |
| 3 | HubSpot | 83 |
| 4 |  Seismic | 29 |
| 5 | Claude Code | 99 |
| 6 | Clay | 90 |
| 7 | Codex | 79 |
| 8 | Gong | 62 |
| 9 | Salesloft | 42 |
| 10 | Zapier | 52 |
A discipline built around experiments, channels and loops that drive efficient growth.
Spring was the peak and share has roughly halved since. Andrew Chen's March post asking what channel AI actually opened sits at the top of the curve. What is left is narrower: Holly Chen's team running 40 to 50 disconnected agents, Anthropic hiring a growth PM. Less theory of the channel, more staffing and cleanup.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Webflow | 46 |
| 3 | HubSpot | 83 |
| 4 | Optimizely | 37 |
| 5 | Clay | 90 |
| 6 | Gamma | 42 |
| 7 | Semrush | 49 |
| 8 | 6sense | 39 |
| 9 | Calendly | 37 |
| 10 | Cursor (Anysphere) | 78 |
Pricing based on what a customer uses, such as credits, tokens or transactions.
Share halved from February to July, then flattened, and the last thirty days are up 52 percent. No launch drives it. Three arguments do: Lemkin using Navan's 92 percent usage revenue as proof, DocuSign's pricing lead explaining the move to credits, and Tyler Denk listing the vendors beehiiv still pays for.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | HubSpot | 83 |
| 3 | Clay | 90 |
| 4 | Intercom | 46 |
| 5 | Vercel | 58 |
| 6 | Claude Code | 99 |
| 7 | Cursor (Anysphere) | 78 |
| 8 | OpenRouter | 51 |
| 9 | Replit | 54 |
| 10 | Twilio | 37 |
Business software that exposes data and actions through APIs, MCP and command-line tools.
April was the spike, off Salesforce's TDX push and Headless 360. The topic then faded through July before turning back up, with the last 30 days up 83 percent from a low base. Martin Kihn's post on the Salesforce MCP server going GA carried August, and Rahul Pathak supplies the argument underneath: agents call APIs and endpoints, so the interface moat is evaporating.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | ZoomInfo | 52 |
| 4 | Claude Code | 99 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | HubSpot | 83 |
| 7 | Replit | 54 |
| 8 | beehiiv | 47 |
| 9 | Airtable | 48 |
| 10 | Apify | 53 |
The tools, data and workflows that turn outbound sales into a repeatable operating system.
This peaked in May and has fallen since, down 46 percent over ninety days and 28 percent in the last thirty. The spring run was build-your-own-machine content, Hans Dekker's AI GTM team the largest of it. What is left is narrower: Clay's CLI, Claude Code skill bundles, one customer story.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | Prospeo | 51 |
| 4 | Salesforce | 87 |
| 5 |  Explorium | 27 |
| 6 | Zapier | 52 |
| 7 | AI Ark | 43 |
| 8 | Apify | 53 |
| 9 | Apollo.io | 54 |
| 10 | Codex | 79 |
Writing designed to persuade someone to click, reply, sign up or buy.
The line wobbles near one percent with a June high and a slow drift down since. The recurring move is to point past the words. Corey Haines shipped an /offers skill on the premise that no headline saves a weak offer, and Pierre Herubel put messaging, offer, and audience ahead of copy. The subject keeps redirecting to something underneath it.
Posts to understand this · as of Sep 12
Growing through partners, channels, co-marketing, co-selling, resellers and other shared routes to market.
The clearest climb here, from the smallest share in February to the largest in August, and still rising over the last thirty days. Programs did it. Lovable opened a solutions partner program with revenue share in July, Moloco launched an agency program with more than a dozen founding agencies in August, and Crossbeam published Adobe's numbers.
Posts to understand this · as of Sep 12
Engineers embedded with customers to make complex AI products work in the customer's environment.
Flat is the story. Share has barely moved over ninety days and sits mid-band, down slightly in the last thirty. Sierra traced the role back to Palantir and called it one of the fastest-growing in AI. Rox launched Teams on the opposite pitch, revenue agents with no FDEs required. Both arguments keep the term in circulation.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Salesforce | 87 |
| 3 | Claude Code | 99 |
| 4 | Codex | 79 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | ElevenLabs | 49 |
| 7 | Gamma | 42 |
| 8 | Replit | 54 |
| 9 | Twilio | 37 |
| 10 | Baseten | 44 |
A go-to-market operating model where agents execute work across the revenue motion.
A sawtooth all year, with discrete announcements setting each step: Salesforce's prospecting agent with ZoomInfo as native data provider in April, then Clay's $115M raise at a $7.1B valuation in September. The last thirty days are up 93 percent. Grafe's BCG read is the counterweight, with only 8 percent of CMOs running hands-off campaigns.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Salesforce | 87 |
| 3 | Claude Code | 99 |
| 4 | HubSpot | 83 |
| 5 | n8n | 59 |
| 6 | Codex | 79 |
| 7 | Cursor (Anysphere) | 78 |
| 8 | 1mind | 37 |
| 9 | Clari | 34 |
| 10 | Common Room | 44 |
Scoring leads and accounts to decide who deserves attention first.
Lead scoring ran between 1.36 and 1.85 percent of the conversation from February through May. It has fallen every month since, 1.38 in June to 0.71 in August, and the last 30 days are down another 18 percent. The substantive posts still date to the February and March peak.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Salesforce | 87 |
| 3 | Clay | 90 |
| 4 | Attio | 50 |
| 5 |  Bombora | 29 |
| 6 | FullEnrich | 42 |
| 7 | ActiveCampaign | 29 |
| 8 |  Adobe Marketo | 25 |
| 9 | Apify | 53 |
| 10 | Claude Code | 99 |
The territories, quotas, compensation plans, capacity models and reporting behind a sales organization.
No launch drives this one. The rise into a May peak and the slow slide since both run on the same argument, restated: quota attainment stuck near 40% for a decade, comp plans that punish the top seller, whether an $870k rep is a cost or a bargain. The last 30 days are down 14%, and the line still sits above where February started.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | HubSpot | 83 |
| 4 | Claude Code | 99 |
| 5 | ElevenLabs | 49 |
| 6 |  Monaco | 27 |
| 7 | 6sense | 39 |
| 8 | Ahrefs | 49 |
| 9 | Chili Piper | 32 |
| 10 | Cline | 24 |
Rules, triggers and workflows that execute marketing programs across systems.
Flat is the story here. The category bottomed near 0.47% in June and has held since, up 0.8% over the last 30 days. What fills the line is vendor supply rather than argument: G2 badge posts, webinar invites, and the occasional sharp take on what HubSpot's campaign object actually measures.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Salesforce | 87 |
| 3 | ActiveCampaign | 29 |
| 4 | Adobe Marketo | 25 |
| 5 | Clay | 90 |
| 6 | Gong | 62 |
| 7 | HeyReach | 52 |
| 8 | Instantly | 61 |
| 9 | Intercom | 46 |
| 10 |  Zendesk | 39 |
Rule-based software that moves business work through predefined steps and systems.
A long fade from February to a July floor, then a lift, up 26 percent in the last thirty days. The return is framed around people, not plumbing: Stevie Case telling closers to run the agent stack, Salesforce profiling sellers who built their own tools, Automation Anywhere arguing a single copilot misses the cross-system work.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | n8n | 59 |
| 3 | Salesforce | 87 |
| 4 | Clay | 90 |
| 5 | HubSpot | 83 |
| 6 | Nooks | 46 |
| 7 | Zapier | 52 |
| 8 | Apollo.io | 54 |
| 9 | Gong | 62 |
| 10 | HeyReach | 52 |
Systems where several agents divide a job, coordinate their work and combine the result.
Steady near one percent through June, then a slide, down 32 percent in the last thirty days. The posts are vendor-led and mostly architecture: Mercor's 100-subagent run, Workato's Airo expansion. Jonathan M K. names what the curve shows, that "AI agent" describes a build, not a reason to buy.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | HubSpot | 83 |
| 3 | Claude Code | 99 |
| 4 | Cursor (Anysphere) | 78 |
| 5 | ElevenLabs | 49 |
| 6 | Buffer | 28 |
| 7 | Granola | 46 |
| 8 | Grok Bot | 57 |
| 9 | OpenClaw | 50 |
| 10 | Replit | 54 |
The point where a product solves a real need for a market that will buy it.
A long fade rather than a break: 1.76% in February down to 0.58% by July, roughly flat since. The term is being redefined, not retired. ClickUp says AI refreshes fit every few weeks, Andrew Chen argues the ship-fast-validate theory will not survive hardware and deeptech, Mark Roberge treats a second ICP as a full reset. Down 5% on the month.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Gamma | 42 |
| 3 | OpenClaw | 50 |
| 4 | OpenRouter | 51 |
| 5 |  Bland | 20 |
| 6 | Clay | 90 |
| 7 | Codex | 79 |
| 8 | Cursor (Anysphere) | 78 |
| 9 | Devin | 36 |
| 10 | Grok Bot | 57 |
Approval and review steps that keep a person involved in an agent's work.
This one bounces rather than trends, between 0.37 and 0.84 every month this year, with the last thirty days up 32 percent. It moves when someone tries to make the phrase mean something specific. Brinker splits it into expertise and thinking time, Healey into seven stages of delegation, McNamara into a plain refusal.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | OpenClaw | 50 |
| 4 | Salesforce | 87 |
| 5 | Ahrefs | 49 |
| 6 | Cursor (Anysphere) | 78 |
| 7 | HubSpot | 83 |
| 8 | Lovable | 57 |
| 9 |  PhantomBuster | 31 |
| 10 | Apify | 53 |
Companies that reorganize how they work around AI instead of stopping at a few AI features.
Share peaked near 0.87 in April and has fallen since, down about 28 percent over 90 days. The May posts from Stevie Case and Tom Wentworth landed while share was still near the top, arguing over what AI-native companies actually staff. By July the talk had moved to Sierra describing its own rollout, and the last 30 days are down only 2.7 percent.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | HubSpot | 83 |
| 4 | Salesforce | 87 |
| 5 | Codex | 79 |
| 6 | Rox | 30 |
| 7 | Grok Bot | 57 |
| 8 | ZoomInfo | 52 |
| 9 | 1mind | 37 |
| 10 | Attention | 32 |
Buying, onboarding and expanding inside a product without talking to a salesperson.
March's 1.27% was the high, and the spring drop to 0.57% held until a slow summer recovery. August's biggest single moment was Bending Spoons agreeing to buy Airtable for $1.285B, down from an $11B valuation, which Elena Verna read as the PLG-to-enterprise trap. The economics arguments follow the deal. Down 10.6% over 30 days.
Posts to understand this · as of Sep 12
Founders, executives and employees using their own voices to distribute the company's ideas.
Share sat between 0.77% and 0.94% from February through July, six straight months of steady founder and employee posting. August broke the band at 0.57%, and the last 30 days are down another 13%. The newer posts argue method and proof rather than a founder stepping into the selling himself.
Posts to understand this · as of Sep 12
B2B brands paying creators and independent voices to reach trusted audiences.
This one runs on announcements, not a trend. August lifted to 0.81% behind two of them: Alex Lindahl taking a creator-in-residence role at Clay, and Maja Voje publishing a five-type program taxonomy drawn from 36 brand collaborations. Jordan Crawford's May post, arguing paid opinions are simply for sale, is the standing counterweight. Down 15.9% over the last 30 days.
Posts to understand this · as of Sep 12
Embedded marks and signed metadata that show where digital content came from and how it was made.
August did all the work. Anthropic's watermarking FAQ on August 14 explained the EU AI Act requirement and pulled the topic from near zero to 0.93%, with Synthesia following on Article 50 and California's transparency act. Liza Adams drew the line between the Claude watermark and LinkedIn's slop button. L30 is roughly flat at minus 5%.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Codex | 79 |
| 3 | Substack | 51 |
| 4 | HubSpot | 83 |
| 5 | Clay | 90 |
| 6 | Cursor (Anysphere) | 78 |
| 7 | Salesforce | 87 |
| 8 | ElevenLabs | 49 |
| 9 | Grok Bot | 57 |
| 10 | Midjourney | 40 |
Websites and docs structured so AI crawlers and agents can read and use them.
Share peaked at 0.71 in April, then fell through the summer to 0.448 in August. The fade tracks Google's May guide calling AEO still SEO and telling sites to skip llms.txt, then June research finding 97 percent of llms.txt files are never fetched. The last 30 days are up 3.9 percent, and the live argument is which crawler directives actually separate AI from search.
Posts to understand this · as of Sep 12
Time-limited access that lets a buyer test a product before paying.
Fell from 1.13% in February to 0.38% by April and has mostly sat there. What lift there is comes from AI products using the trial as distribution, like Grok Bot opening a limited free trial beyond its paying subscribers. The rest is pricing mechanics: funnel conversion benchmarks, and 511 pricing changes across 564 companies in one quarter. Up 2.7%.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | HubSpot | 83 |
| 3 | Codex | 79 |
| 4 | Salesforce | 87 |
| 5 | Clay | 90 |
| 6 | Cursor (Anysphere) | 78 |
| 7 | Instantly | 61 |
| 8 |  Lindy | 24 |
| 9 | Calendly | 37 |
| 10 | Firecrawl | 39 |
Advantages that grow stronger as more people, data or participants join a product.
The moat conversation did not die, it changed owners. Classic network-effects talk fell from 1.87% in February to 0.442% in August while Salesforce, Sierra and platform bulls reframed defensibility around proprietary data, permissions and long-horizon agent relationships. Up 7.5% over the last 30 days, so the floor is holding under a lower line.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Claude Code | 99 |
| 3 | Clay | 90 |
| 4 | Gamma | 42 |
| 5 | Calendly | 37 |
| 6 | Codex | 79 |
| 7 | Cursor (Anysphere) | 78 |
| 8 | Guru | 21 |
| 9 | HubSpot | 83 |
| 10 | Intercom | 46 |
Systems for producing and distributing large volumes of marketing content, often with AI.
Peaked at 0.96% in May, then halved by July and stayed there. Volume stopped being the story. Peter Caputa has production automated and measurement still done by hand, Kieran Flanagan sends up the agent-swarm content machine, and Lily Ray points at a programmatic SEO exemplar crashing further in a Google spam update. Flat on the month, down 41% over 90 days.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Clay | 90 |
| 3 | Codex | 79 |
| 4 | Buffer | 28 |
| 5 | Claude Code | 99 |
| 6 | Granola | 46 |
| 7 | Salesforce | 87 |
| 8 | Semrush | 49 |
| 9 | Substack | 51 |
| 10 | beehiiv | 47 |
Products distributed as apps or connectors inside AI assistants and agent marketplaces.
Spiky through spring, swinging from 0.98% to 0.30% and back, then three steady months near 0.63%. The plateau is the point: it holds because product keeps shipping. n8n's Claude Code connector now writes workflows, Grok added a Figma connector, and Salesforce took hosted MCP servers to GA. Episodic news turned routine. Up 41% over 90 days, down 12.3% on the month.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Canva | 48 |
| 3 |  Cargo | 30 |
| 4 | Prospeo | 51 |
| 5 | Airtable | 48 |
| 6 | Apollo.io | 54 |
| 7 | Clay | 90 |
| 8 | Codex | 79 |
| 9 | Cursor (Anysphere) | 78 |
| 10 | Exa.ai | 43 |
AI agents that discover, compare and buy products on a person's behalf.
Share peaked in March at 0.93% and has not gone back. Since April it has sat in a narrow band near 0.5%, and the last 30 days are flat. The talk matches that shape: people keep arguing about where agent buying actually stops, and the push now comes mostly from vendors telling brands to get ready.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Exa.ai | 43 |
| 4 | Hermes Agent | 37 |
| 5 | Midjourney | 40 |
| 6 | Profound | 50 |
| 7 |  Similarweb | 35 |
| 8 |  StackAdapt | 18 |
The software stack sales teams use to prospect, engage buyers and manage deals.
The category label keeps losing share, down 21.6% over 90 days and 9.5% in the last 30, even as the deals underneath stay big. Aligned raised a $60M Series B, Zoom bought Common Room, and stack roundups still circulate. People increasingly name a specific layer instead of talking about the stack.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Gong | 62 |
| 3 | Claude Code | 99 |
| 4 | Clay | 90 |
| 5 | Attention | 32 |
| 6 | Explorium | 27 |
| 7 | FullEnrich | 42 |
| 8 | Instantly | 61 |
| 9 | Nooks | 46 |
| 10 | Outreach | 44 |
A permanently free product tier designed to attract users and convert some to paid plans.
Freemium share peaked at 0.89 percent in February, swung back up to 0.44 in May, then bottomed at 0.07 percent in July. August recovered to 0.35 percent, and the last 30 days are up 224 percent while the last 90 days are still down 32 percent. September has only twelve days observed.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Claude Code | 99 |
| 3 | Codex | 79 |
| 4 | Canva | 48 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | Intercom | 46 |
| 7 | Salesloft | 42 |
| 8 | ActiveCampaign | 29 |
| 9 | Adobe Marketo | 25 |
| 10 | Ahrefs | 49 |
Using AI to tailor outreach and marketing to each person at scale.
The slide is steady, not sudden: eight months down from 1.09% in February to 0.54% in August, with no single event marking a top. What changed is the argument. Clay's ChatGPT app sold personalization at scale, but by spring Kyle Asay was cutting 1,000 accounts down to 50, and Nick Palasz argued an observation is not a reason to reply.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Salesforce | 87 |
| 3 | ElevenLabs | 49 |
| 4 |  Lightfield | 23 |
| 5 | Salesloft | 42 |
| 6 | 6sense | 39 |
| 7 | Apify | 53 |
| 8 | Clay | 90 |
| 9 | Codex | 79 |
| 10 | Cursor (Anysphere) | 78 |
Deciding what to build, why it matters and which product bets come first.
Share has swung between 0.39% and 0.72% all year with no steady trend, and it sits near the low end now at 0.43%. The last 30 days are down about 11%. What people are posting is concrete: where to place a bet in a crowded category, how agents change a roadmap, and who owns product.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Salesforce | 87 |
| 4 | Claude Code | 99 |
| 5 | Devin | 36 |
| 6 | Exa.ai | 43 |
| 7 | Grok Bot | 57 |
| 8 | Hermes Agent | 37 |
| 9 | OpenClaw | 50 |
| 10 | Bland | 20 |
Protecting models and agents from prompt injection, data leaks, unsafe access and attacks.
The flattest curve here, holding between 0.43 and 0.66 every month this year, with the last thirty days off 51 percent. No viral moment, just steady vendor cadence: OpenAI's gpt-red red teamer, Cursor's AIUC-1 certification, Vanta crossing $300M ARR on shadow-AI risk. The plateau is built from many small proofs.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Hermes Agent | 37 |
| 4 | Canva | 48 |
| 5 | Claude Code | 99 |
| 6 | Devin | 36 |
| 7 | ElevenLabs | 49 |
| 8 | Grok Bot | 57 |
| 9 | Groq | 25 |
| 10 | Midjourney | 40 |
Marketing inside large companies with bigger budgets, stricter governance and more complex brands.
The one line going up. It roughly doubled from 0.24% in April to 0.42% in July, held there through August, and is up 57% over 90 days. Platforms did the sustaining rather than any one launch: LinkedIn's Adobe AI skills partnership in June, Canva and LinkedIn's ads workflow session in August, Writer's agentic GTM fireside.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Vercel | 58 |
| 3 | ActiveCampaign | 29 |
| 4 | Adobe Marketo | 25 |
| 5 | CHEQ | 15 |
| 6 | Claude Code | 99 |
| 7 | Codex | 79 |
| 8 |  DemandScience | 23 |
| 9 | Docket | 28 |
| 10 | ElevenLabs | 49 |
The runtime around an agent that provides tools, state, controls and stopping rules.
The term was absent from the panel in February and has swung between 0.27% and 0.53% ever since. August's high came with DeepSeek's open-source harness preview and Perplexity's fully local runtime on NVIDIA DGX Spark. The last 30 days give back 17%. That is what a topic carried by releases looks like, not one carried by a standing argument.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Claude Code | 99 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Vercel | 58 |
| 5 | OpenRouter | 51 |
| 6 | Cline | 24 |
| 7 | ElevenLabs | 49 |
| 8 | Grok Bot | 57 |
| 9 | Lovable | 57 |
| 10 | OpenClaw | 50 |
Companies reducing their go-to-market stack by replacing point tools with broader platforms.
The topic peaked at 1.50% share in February, then more than halved in March and slid to a low of 0.30% in August. May was the one rebound, 0.93% and the second-highest month. September is only 12 days in, and the last 30 days are down 18.6%, so the line is still falling.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | ZoomInfo | 52 |
| 4 | ActiveCampaign | 29 |
| 5 | Adobe Marketo | 25 |
| 6 | Bombora | 29 |
| 7 | Codex | 79 |
| 8 | Cursor (Anysphere) | 78 |
| 9 |  Fireflies | 35 |
| 10 | Gong | 62 |
Tracing and monitoring what a model or agent did during a production run.
Eight months inside a 0.29% to 0.46% band, and down 4% over the last 30 days. A plateau this flat means steady vendor supply rather than debate: Intercom's Fin suite in March, Parloa Lens in July, ClickHouse joining the Open Secure AI Alliance with Langfuse. The releases keep arriving and almost nobody argues about them.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Salesforce | 87 |
| 4 | Claude Code | 99 |
| 5 | LangChain | 43 |
| 6 | Grok Bot | 57 |
| 7 | HubSpot | 83 |
| 8 | OpenClaw | 50 |
| 9 | Cline | 24 |
| 10 |  CrewAI | 30 |
The strategy of naming, defining and leading a new market.
Category share slid from 0.50% in February to 0.32% in July, then August pulled it back to 0.40%. The last 30 days are up 38.6%, while the 90-day trend is still down 8.2%. The rebound is an argument, not agreement: operators are launching new categories and warning founders off inventing one in the same window.
Posts to understand this · as of Sep 12
The SDR and BDR function that creates pipeline through prospecting and outreach.
The clearest decline in the set, down 52 percent over ninety days and 49 percent in the last thirty, from near two percent share in February. What remains is about replacement rather than practice: seven agents running GTM with no SDR team, Notion asking applicants to build an agent instead of sending a resume.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Nooks | 46 |
| 2 | Salesforce | 87 |
| 3 | Claude Code | 99 |
| 4 | Clay | 90 |
| 5 | HubSpot | 83 |
| 6 | Clari | 34 |
| 7 | Gong | 62 |
| 8 | Instantly | 61 |
| 9 | Lemlist | 50 |
| 10 | Outreach | 44 |
AI agents that create, launch, buy and optimize advertising.
Up 71.7% over 90 days, and the shape is infrastructure arriving rather than one launch. StackAdapt opened early access to ads in ChatGPT, then Hightouch shipped conversions API support and AppsFlyer added install and purchase measurement. Each vendor wires the same new channel into existing plumbing. The last 30 days are flat.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | HubSpot | 83 |
| 3 | OpenClaw | 50 |
| 4 | Similarweb | 35 |
| 5 | Cursor (Anysphere) | 78 |
| 6 |  Deepgram | 24 |
| 7 | Higgsfield | 38 |
| 8 |  Lambda | 22 |
| 9 | Midjourney | 40 |
| 10 | Runway | 33 |
Independent advisors and fractional executives selling experience, judgment and hands-on help.
Advisory share slid from an April peak through a July low, then more than doubled in August. The last 30 days are up 66.8 percent, while the three-month trend is still down. August was solo-operator craft: pipeline habits and the life a practice buys. The buyer-math case, fractional against a full-time hire, shows up on September 1.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | DemandScience | 23 |
| 4 |  Framer | 35 |
| 5 |  Reevo | 24 |
| 6 | Webflow | 46 |
Writing and structuring the instructions a model follows.
Share has fallen most months since February's 1.09, and the L30 is down 47%. No single event broke it. All three posts make the same move, retiring prompts in favor of context: Glean cut its own system prompt by 45% by pushing instructions into skills, Weaviate says better prompts won't fix a broken pipeline, DevRev sells a session on context engineering.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Cursor (Anysphere) | 78 |
| 2 | Claude Code | 99 |
| 3 | Cline | 24 |
| 4 | Codex | 79 |
| 5 | Vercel | 58 |
| 6 | Canva | 48 |
| 7 | Devin | 36 |
| 8 | Granola | 46 |
| 9 | HubSpot | 83 |
| 10 | LangChain | 43 |
Marketing to existing leads and customers through onboarding, nurture, retention and expansion programs.
April peaked at 0.89 and share has drifted down since, with the L30 off 32%. This one runs on vendor supply, not practitioner argument. Hightouch's Lifecycle Studio in June was the last real launch; what has come after is a Dreamforce session from Gainsight and a customer story from Bloomreach, which is what the gap between launches looks like.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | ActiveCampaign | 29 |
| 3 | AI Ark | 43 |
| 4 | Adobe Marketo | 25 |
| 5 | Ahrefs | 49 |
| 6 | Airtable | 48 |
| 7 | Apify | 53 |
| 8 | Apollo.io | 54 |
| 9 |  BetterContact | 42 |
| 10 |  Browserbase | 32 |
Founders running early sales themselves before handing the motion to a sales team.
Thin and jumpy. It nearly vanished in May at 0.06%, then reached 0.44% in July behind two posts: Peter Kazanjy opening Founding Sales to every founder, and Scott Leese's account of a $200k hire who closed zero deals. Down 18% over the last 30 days. With a base this small, one post moves the whole line.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Codex | 79 |
| 3 | AI Ark | 43 |
| 4 | AirOps | 36 |
| 5 | Apify | 53 |
| 6 | Apollo.io | 54 |
| 7 | BetterContact | 42 |
| 8 | Chili Piper | 32 |
| 9 | Claude Code | 99 |
| 10 | Clay | 90 |
A shared revenue knowledge layer that people and agents can query and update.
Share has sat in a 0.34 to 0.45 band most of the year, with one June spike that fell straight back. August held the band while ZoomInfo put its context layer inside Slack. The last 30 days are down about 10 percent, and September is still running, so the band is the story, not the spike.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Cursor (Anysphere) | 78 |
| 3 | Salesforce | 87 |
| 4 | Apify | 53 |
| 5 | Attio | 50 |
| 6 | Clay | 90 |
| 7 | Codex | 79 |
| 8 | Lovable | 57 |
| 9 | Ahrefs | 49 |
| 10 | Exa.ai | 43 |
Multi-step work planned and executed by agents instead of fixed automation rules.
HubSpot's Agent Hub and Agent Builder beta in July drew the biggest post here by far, and share still fell to 0.34 in August with the L30 down 38%. What followed the launch was doubt. Elaine Zelby has worked with hundreds of marketing teams and says none run fully agentic workflows, and brendan short puts the block on context rather than the workflow itself.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Claude Code | 99 |
| 3 | Apollo.io | 54 |
| 4 | Attio | 50 |
| 5 | Codex | 79 |
| 6 | Explorium | 27 |
| 7 | FullEnrich | 42 |
| 8 | Gong | 62 |
| 9 | Lemlist | 50 |
| 10 |  PredictLeads | 38 |
How AI changes headcount, entry-level work and the jobs people still need to do.
This fell hardest, off 55% over 90 days from a 1.44% March high. The spring peaks were argument posts, like Melissa Rosenthal reading Gartner's finding that 80% of enterprises cut jobs with no matching return. By July the frame had turned operational: Notion collapsed nine specialized GTM roles into four builder roles.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Salesforce | 87 |
| 3 | Claude Code | 99 |
| 4 | Grok Bot | 57 |
| 5 | Lovable | 57 |
| 6 | Airtable | 48 |
| 7 | Apify | 53 |
| 8 | Apollo.io | 54 |
| 9 | Attention | 32 |
| 10 | Augment Code | 17 |
Go-to-market teams designed around AI from the start, with different roles, costs and speed.
Attention peaked in May and has drifted down since, off about 24% in the last 30 days. The content changed shape more than it shrank: less stack mapping, like the Battery Ventures AI-first graphic, and more operating detail, like Zapier's draft list of skill and context ownership needs and Caitlin Fischetti's pilot team ripping into what they shipped.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Clay | 90 |
| 3 | Salesforce | 87 |
| 4 | Apify | 53 |
| 5 | HubSpot | 83 |
| 6 | Lovable | 57 |
| 7 | Rox | 30 |
| 8 | n8n | 59 |
| 9 | AI Ark | 43 |
| 10 | Airtable | 48 |
Teams building internal software with AI instead of buying another SaaS product.
This sat between 0.36 and 0.55 every month from February through August, and the L30 is down 39% off that band. The plateau is the story. No launch drives it. It runs on cost math, from Jake Dunlap's plan to replace Gong, Salesforce and Clay to Bill Binch's CRO dinners where build versus buy splits the room.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | Salesforce | 87 |
| 3 | Sumble | 32 |
| 4 | 11x | 25 |
| 5 | 1mind | 37 |
| 6 | AiSDR | 18 |
| 7 | Attention | 32 |
| 8 | Augment Code | 17 |
| 9 | Clarify | 29 |
| 10 | Claude Code | 99 |
How sales teams use AI and how it changes the work of selling.
Share of the conversation has been sliding since February and fell again over the last 30 days, down 58 percent. The talk that is left is not about tools. It is about the seller: pair sales skill with technical skill or fall behind, specialized SDR and AE roles collapsing back to full cycle, and the counter case that selling gets safer as AI spreads.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Apollo.io | 54 |
| 2 | Attention | 32 |
| 3 | Clari | 34 |
| 4 | Explorium | 27 |
| 5 | FullEnrich | 42 |
| 6 | HubSpot | 83 |
| 7 | Instantly | 61 |
| 8 | Lemlist | 50 |
| 9 | Lovable | 57 |
| 10 | PredictLeads | 38 |
The sameness that appears when AI gives every company similar messaging, claims and websites.
Share peaked in February at 0.66, dipped in March, then came close again in April. Every month since May has been lower, and the last 30 days are down 45 percent. The complaint has not changed, it has thinned. Fewer people are still saying out loud that AI makes everyone sound alike.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Codex | 79 |
| 2 | Salesforce | 87 |
| 3 | Semrush | 49 |
| 4 | Ahrefs | 49 |
| 5 | AirOps | 36 |
| 6 | Clay | 90 |
| 7 | PhantomBuster | 31 |
| 8 | Substack | 51 |
| 9 |  Userpilot | 23 |
| 10 |  Viktor | 31 |
Tools that estimate whether text was written by AI by analyzing its style and statistical patterns.
This did not exist in February and is up 287% across ninety days, with July at 0.28 the high month. Two things built it: a Pangram Labs study of over a million LinkedIn posts putting 40% at AI slop, and platforms shipping detection themselves. Substack counts LinkedIn, Spotify and Snapchat following with anti-slop features. The L30 is down 47%.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Claude Code | 99 |
| 2 | Substack | 51 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Grok Bot | 57 |
| 5 | Codex | 79 |
| 6 | ElevenLabs | 49 |
| 7 |  Grammarly | 28 |
| 8 | Lindy | 24 |
| 9 | Lovable | 57 |
| 10 |  Netlify | 25 |
The online channel mix used to reach, convert and retain customers.
Small and choppy, with no sustained base. Share has bounced between 0.10% and 0.39% since February with no direction, and the L30 is down 19.7%. The recurring frame is naming what got replaced: Rand Fishkin mapping what zero-click made useless, Tom Wentworth counting the earlier waves that each minted a new job title.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | HubSpot | 83 |
| 2 | Claude Code | 99 |
| 3 | Frontier | 21 |
| 4 | Grok Bot | 57 |
| 5 | HG Insights | 27 |
| 6 | Midjourney | 40 |
| 7 | Replit | 54 |
| 8 | Salesforce | 87 |
| 9 | TechTarget | 23 |
| 10 |  The Trade Desk | 21 |
Pricing tied to a delivered result, such as a resolution, meeting or completed task.
Share peaked in May at 0.387 after HubSpot put per-outcome prices on two of its agents in April. It fell every month after that, down 42.5 percent over 90 days to 0.092 in August. The last 30 days are close to flat at minus 7 percent, with the argument now about what outcome pricing does to sales teams.
Posts to understand this · as of Sep 12
inPrice changePriced two agents per outcome: 50 cents a resolution, one dollar a lead.VE
inInvestor framingVista calls the seat-to-outcome shift a market expansion story for enterprise software.JV
inSales falloutArgues outcome pricing breaks the close, forecasting and qualification in orthodox sales orgs.| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | Cursor (Anysphere) | 78 |
| 4 | Intercom | 46 |
| 5 | ZoomInfo | 52 |
| 6 | Apify | 53 |
| 7 | Bland | 20 |
| 8 | HubSpot | 83 |
| 9 | Lovable | 57 |
| 10 | OpenRouter | 51 |
The belief that spending more on model calls and agent work can create an advantage.
The word arrived in April as marketer slang and got its argument in June, when Wade Foster tied it to Uber reportedly burning a year of AI budget in four months and pushed deterministic workflows for the repeat work. July was the high point. Since then it has become someone else's framing, like Altimetrik's move from tokenmaxxing to FDE-maxxing, and the last 30 days are down 66%.
Posts to understand this · as of Sep 12
The plan, act, inspect and retry cycle an agent follows until it finishes.
Joe Rhew called it in June: context is solved, loops are next, evals come back. The share quadrupled that month and held through July, when Palantir shipped Agent Engine and Agent SDK at Devcon 6. Then the pushback arrived, and the line is down 52% over 30 days. Still up 176% across 90.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | AssemblyAI | 34 |
| 2 | beehiiv | 47 |
| 3 | Claude Code | 99 |
| 4 | Clay | 90 |
| 5 | Cursor (Anysphere) | 78 |
| 6 | Devin | 36 |
| 7 | Grok Bot | 57 |
| 8 | Lovable | 57 |
| 9 | OpenClaw | 50 |
| 10 | Replit | 54 |
AI-generated creators, avatars and user-style content used in marketing.
Share has sat in a low band since April. The May distribution deals, HeyGen going native inside Canva and shipping a Lovable connector, did not lift the monthly line, which is the useful signal here. The last 30 days are up about 16%, and the posts now come from people testing their own avatars rather than platforms announcing reach.
Posts to understand this · as of Sep 12
Interfaces that let agents understand and operate software built for human users.
This moves on vendor announcements, not a debate. The May high of 0.32% faded, the L30 is down 63%, and L90 is flat, so the swings are noise around a thin base. The evidence is all supply side: Optimizely calling agents a new kind of customer, LangChain rebuilding internal data access around agents, Tealium shipping configuration MCP.
Posts to understand this
| Brand | GTM Index | |
|---|---|---|
| 1 | Salesforce | 87 |
| 2 | Clay | 90 |
| 3 | ZoomInfo | 52 |
| 4 | Apify | 53 |
| 5 | Apollo.io | 54 |
| 6 |  Ciro | 20 |
| 7 | Clarify | 29 |
| 8 | Claude Code | 99 |
| 9 | Codex | 79 |
| 10 | Common Room | 44 |
Brands use human authorship and craft as a reason to trust or value their work.
Small and choppy from a near-zero base. The May bump was Dave Gerhardt arguing people will place more value on the real and the imperfect. Since then the phrase has turned into sales copy: 200+ human-crafted skills in Amos Bar-Joseph's GTM marketplace, agencies promising content that does not scream AI slop. Up a third over 90 days, down 39% in the last 30.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Rox | 30 |
| 2 | Attention | 32 |
| 3 | Clarify | 29 |
| 4 | Claude Code | 99 |
| 5 | Clay | 90 |
| 6 | Clearskies | 21 |
| 7 | Grok Bot | 57 |
| 8 | Nooks | 46 |
| 9 | Sumble | 32 |
| 10 | Salesforce | 87 |
Marketing that builds demand through credible proof, useful work and long-term audience relationships.
No single event carries this one. June was the high point and the line has fallen since, down 34% over 30 days. What is left is argument rather than news: Tyrona Heath saying volume is not a trust signal, Ross Simmonds calling AI visibility the compounded return on years of distribution decisions.
Posts to understand this · as of Sep 12
The argument over requiring a form before someone can access useful content.
Share peaked near 0.32 percent in February and March, then slid through spring and summer to under 0.10 percent, with the last 30 days down 22 percent. The argument itself has not settled. Amanda Natividad draws a line on what earns a form, Ashley Faus says ungate all of it.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | HubSpot | 83 |
| 3 | Salesforce | 87 |
| 4 | ActiveCampaign | 29 |
| 5 | BetterContact | 42 |
| 6 | Calendly | 37 |
| 7 | Chili Piper | 32 |
| 8 | Clari | 34 |
| 9 | Claude Code | 99 |
| 10 |  Crossbeam | 25 |
Using product activity to identify, time and support sales opportunities.
This line sits near zero and only moves when something concrete happens. Brendan Short's March post closed the category: $150M raised, 14 startups, none still operating independently after Pocus went to Apollo. What remains is build talk, like Dan Rosenthal's HubSpot and Clay stack for a high-volume PLG client. Up 29% over 30 days from a tiny base.
Posts to understand this · as of Sep 12
| Brand | GTM Index | |
|---|---|---|
| 1 | Clay | 90 |
| 2 | 6sense | 39 |
| 3 | Attention | 32 |
| 4 | Clarify | 29 |
| 5 | Claude Code | 99 |
| 6 | Clearskies | 21 |
| 7 | Codex | 79 |
| 8 | Demandbase | 31 |
| 9 |  Hightouch | 33 |
| 10 | HubSpot | 83 |
Software designed for agents to operate as the primary user.
New in the last 90 days and entirely vendor-driven. Netlify, MotherDuck and Retool each published a version of the same claim: agents sign up, read docs and call APIs, so build the product for them. No practitioner thread picked it up, and the share is down 82% over 30 days.
Posts to understand this · as of Sep 12
How model costs, services and agent work put pressure on software gross margins.
Claude Code is Anthropic's coding agent for writing, testing and shipping code from a terminal or IDE.
Share peaked in March at 11.8%, then fell in every completed month through August, ending at 5.8%. The last 30 days are down another 17%, with September only 8 days in. The posts still traveling are narrow and practical: a gripe about the terminal, a twelve-step setup guide, and a security find passed along.
Posts to understand this · as of Sep 8
Cursor is an AI code editor with agents that write, edit and review code.
One event, one bump. Cursor closed its acquisition by SpaceX on August 14 and joined the SpaceXAI team, and the follow-on talk carried it another week: Martin Casado's board farewell, the reshared $60 billion price. Outside that, Cursor holds a narrow two to three percent band all year. L30 is up about 26%.
Posts to understand this · as of Sep 12
Clay is a flexible data and workflow platform for enrichment, research and automated go-to-market work.
Clay peaked in May and gave back 30% over the following 90 days. The last 30 are up 13%, and the fresh item is the September 9 Series D: $115M at $7.1B, plus a $1M scholarship fund for people becoming GTM engineers. Underneath the funding news, Clay keeps showing up as assumed infrastructure in other people's play lists and stacks.
Posts to understand this · as of Sep 12
n8n is an open-source workflow automation tool that combines visual steps, code and AI agents.
n8n peaked in June after the May multi-year partnership that embedded it natively inside SAP's Joule Studio, then gave most of that back over the summer. The last 30 days are up 8.6%. The talk has shifted off n8n's own news and onto whether Clay workflows or Anthropic's Routines replace it.
Posts to understand this · as of Sep 12
Lovable is an AI app builder that turns plain-language prompts into working software.
Lovable hit 3.0% in March behind the free build day it announced with Anthropic, then slid to 0.7% by July. L90 is down 47%. L30 is up 135%, but the recent lift comes from other companies' launches, like Slack listing Lovable among ten partners for Add to Slack. Its own big moments are the older ones.
Posts to understand this · as of Sep 8
Replit is an AI development platform that builds, deploys and runs applications from a prompt.
Replit is down a quarter over 90 days and up 10.6% in the last 30. August carried the weight: OpenAI and Replit both announced free mode powered by GPT-5.6 Luna on the same day. The other thing keeping the name in circulation is the old deleted production database story, now retold as an AI governance lesson.
Posts to understand this · as of Sep 12
OpenClaw is a self-hosted agent that remembers context and works across chat apps and local tools.
Share fell from 3.6% in February to 0.6% in July, then stopped moving. The last 30 days are flat at -1.3%. The recent posts sit inside that flat stretch, not the decline. In them OpenClaw is the comparison point for Grok Bot and one named example of a harness, not the thing people are picking.
Posts to understand this · as of Sep 12
Codex is OpenAI's coding agent for planning, writing, testing and shipping software changes.
Codex has held near 3% since May, the steadiest line in this set and up 12.4% over 90 days. It now appears as an assumption rather than a subject. Exa ships a plugin for it, Retool builds a governed runtime for what it produces, and Lenny Rachitsky reports 95% of OpenAI engineers use it daily.
Posts to understand this · as of Sep 8
Hermes Agent is an open-source personal agent that keeps memory and learns reusable skills across tasks.
Hermes peaked in spring, when builders like Cody Schneider were wiring it into live business data as a GTM engineering agent, and has faded since. Down about 42% over both 30 and 90 days. Shipping has kept it on the board rather than grown it: Stripe skills in June so an agent can pay, bot mode in August.
Posts to understand this · as of Sep 8
Salesforce is the enterprise CRM for managing sales, service, marketing and customer data.
Salesforce has sat in a tight four to six percent band all year, and the flatness is the story. It keeps showing up as the counterparty in other companies' news: the roughly $3.6B Fin deal in June, the Claudeforce partnership with Anthropic in August, Informatica's Dreamforce debut in September. L30 is up about 28% with Dreamforce in frame.
Posts to understand this · as of Sep 12
HubSpot is a CRM and go-to-market platform for marketing, sales, service and content.
HubSpot's share slid through the summer, down 30 percent over 90 days and sitting near 3.4 percent in August. The last 30 days are up 6.7 percent, so the slide has stalled rather than reversed. September is only eight days in. HubSpot still shows up in other people's stacks, but as one part, not the story.
Posts to understand this · as of Sep 8
Gong records and analyzes sales conversations to improve coaching, deal execution and forecasting.
Down 21% across 90 days, and the reason is not silence. Gong published its own numbers in May: half a billion in ARR, 55% year-over-year growth, half the Fortune 10. Since then it appears mostly inside other people's posts, as a data source in Clay audiences and as the target of a demo-response test.
Posts to understand this · as of Sep 12
Vercel is a platform for deploying web applications and the infrastructure used by AI agents.
Vercel's L30 is up 38%, and almost none of it is Vercel talking. Benioff launched Slack Code with Vercel among the agent partners, Cognition announced Jared Palmer, who built v0 and the AI SDK there, and Notion's Ivan Zhao called his company's moment a re-founding onstage at Vercel Ship. Other people carry the name.
Posts to understand this · as of Sep 8
Claude Cowork is Anthropic's general-purpose agent for research, documents and other knowledge work.
A long fade. March was the high, every month since has been lower, and the line is down about 58 percent over the quarter plus another third in the last 30 days. The talk moved off the product and onto what people do with it: editing a newsletter, wiring up the Zapier MCP, arguing over whether Google absorbs the category.
Posts to understand this · as of Sep 8
Instantly is a cold-email platform for finding leads, managing inboxes and running outbound campaigns.
Instantly mostly rides other people's stacks. It climbed through spring as builder posts named it the send layer, like Jan Rasmussen's 32-hook Claude Code pack in April. Its own AI Lead Finder launch in July landed quietly, and share has stayed well under the June level since. The last 30 days are up 37%.
Posts to understand this · as of Sep 12
OpenRouter is one API for accessing, routing and paying for models from many providers.
One event did this. OpenRouter announced on August 19 that it is joining Stripe, and share jumped from 0.39% in July to 1.16% in August. Peter Walker's token-spend breakdown the next day kept it going, putting agents and code generation at 60% of every AI dollar. The L30 move is still up 104%.
Posts to understand this
Substack is a publishing platform for newsletters, podcasts and paid creator subscriptions.
The line drifts down all year, from 1.16% in February to under 0.8% since June, off 19.5% over 90 days and 12.9% over 30. No crash, just fewer reasons to bring it up. What talk remains is about leaving it or policing it, not building on it.
Posts to understand this
Semrush is an SEO and content platform for search research, competitive analysis and AI visibility.
Semrush peaked in June and has fallen since, with the last 30 days down 40 percent. Its own post on Google's goto redirects, arguing rank tracking was unaffected, is the freshest thing in the set. The rest uses Semrush as a reference point: Austin Armstrong's "RIP Semrush and Ahrefs," Brendan Hufford on the Cannes villa. The brand is mostly carried by its own publishing.
Posts to understand this · as of Sep 12
Canva is a visual design platform for creating graphics, presentations, video and other marketing assets.
April was the peak, and the fall since has cut share roughly in half over 90 days, with the last 30 flat. Canva's own posts still land, like the 100 visual suite upgrades. The rest is other people's context: a Klaviyo email integration, and a Canva leader talking about agentic prospecting and MCP rather than design.
Posts to understand this · as of Sep 12
Ahrefs is an SEO platform for keyword research, backlinks, rank tracking and AI visibility.
Ahrefs spiked in May, faded through summer, and is up about 21% over the last 30 days. The recent lift is company-driven: the statement on Google routing results through /goto redirects and breaking data, plus the Evolve speaker lineup. Pulling the other way are posts that use Ahrefs as the thing AI citation is supposed to replace.
Posts to understand this · as of Sep 12
Zapier connects applications and automates work through triggers, actions and AI agents.
Zapier's share peaked in February near 2.2 percent, then halved by June and sat flat through August. The last 30 days are up 35 percent, but off that low base, and September is only eight days in. The posts run from the February high to a Zapier event push in early September.
Posts to understand this · as of Sep 8
ZoomInfo provides B2B contact, company and intent data for sales and marketing teams.
Down about 40 percent across the quarter, then up about 39 percent in the last 30 days. The recent lift is product and founder-led rather than category: the ZoomInfo MCP going live in the Slack marketplace, and a founder DM getting written up. The pressure underneath is price, with cheap contact data aimed straight at ZoomInfo and Apollo.
Posts to understand this · as of Sep 8
Profound measures and improves how brands appear in AI answers.
ElevenLabs provides voice models and agents for speech, dubbing and conversational AI.
ElevenLabs share slid through the summer, from its spring peak down to an August low. The last 30 days turned that around, up about 30 percent, and the posts driving it are all recent: an owned product launch, a GTM podcast, and a rival benchmarking against them. September is only twelve days in, so the month is not done.
Posts to understand this · as of Sep 12
Attio is a flexible CRM with a database-style model, automation and agent access.
Share doubled into a May peak alongside the GTM Atlas launch, then gave most of it back and sits down roughly a quarter over 90 days. What kept Attio in the conversation after that was argument about the company rather than shipping news: Anthony Pierri picking apart the tagline as a paying customer, and reviews of the flexible data model.
Posts to understand this · as of Sep 12
Apollo.io combines a B2B contact database, sales engagement tools and AI-assisted prospecting.
Apollo started the year at zero share, climbed through spring to 1.1% in May, and has held a narrow band since. The 90-day read is flat and the last 30 days are up 11.5%. Placement holds it there more than product news: the Earthquakes sponsorship, a slot in Replit's growth kit, and practitioners stripping stacks back to it.
Posts to understand this · as of Sep 12
Apify provides ready-made web scrapers and cloud infrastructure for collecting web data.
Apify was barely present in March and then got pulled into everyone else's stack posts: Claude MCP connector roundups, a six-workflow GTM recipe, a podcast clip on cutting scraping costs. That is the pattern here. No Apify launch drives this curve, other people's tooling lists do. L90 is up 6.5 percent and L30 is down 4.9.
Posts to understand this · as of Sep 12
Granola is an AI meeting notepad that turns rough notes and transcripts into useful summaries.
Granola peaked at 1.03% in May and has run near half that since, down 41.4% over 90 days and 15.2% over 30. The Apple Watch launch in late July did not restore it. What surrounds the decline is boundary talk: where the tool fits, and what would make it unnecessary.
Posts to understand this
HeyReach automates LinkedIn outreach across multiple sender accounts and a shared inbox.
HeyReach goes from zero in February to visible in March, when brendan-short listed it alongside Apollo, lemlist and Amplemarket as banned from LinkedIn. Attention held near 0.9 percent through June on stack roundups rather than launches, then settled lower. The L30 move is up 3.5 percent, so the current level is holding rather than recovering.
Posts to understand this · as of Sep 12
RB2B identifies the people and companies visiting a website so sales can follow up.
RB2B climbed for five straight months into a June high, then gave most of it back by August, and the last 30 days are down 28 percent. None of the recent talk is about the product. Adam Robinson writing about VC-backed founders, Sam Dunning naming RB2B as an SEO client, Bill Stathopoulos filing it in a tool list. Founder voice and stack mentions, not launches.
Posts to understand this · as of Sep 12
Intercom is a customer messaging platform built around Fin, its AI support agent.
June carried the whole story: Salesforce signing a definitive agreement to acquire Fin, the customer agent platform used by more than 30,000 companies. Share fell to about 0.3% across July and August once the deal news passed, with Jason Lemkin's $3b+ exit reference as the main echo. Up 77% over 30 days from that low base.
Posts to understand this · as of Sep 12
LangChain provides open-source frameworks and managed infrastructure for building stateful AI agents.
LangChain's share swung hard early in the year, from 0.39% in March down to 0.08% in April. It then held near 0.4% for three straight months, June through August. September is only 13 days in, and the last 30 days are down 16%, so the line is easing off that plateau rather than climbing.
Posts to understand this
Baseten deploys and runs custom AI models in production.
Baseten has risen steadily since March with no single spike. What sustains it is being named inside other companies' launches: day-0 partner for Kimi K3, then Harvey's post-trained agents for M&A diligence. The free inference engineering book is the one piece of its own that traveled. L30 is up 45.2%.
Posts to understand this · as of Sep 12
Nooks is a sales workspace for parallel dialing, prospecting, coaching and AI-assisted outreach.
Nooks is up about 7% over the last 30 days, and the shape is a bounce rather than a climb: 0.66% in July, 0.34% in August, back near the July level in early September. Product proof carries it, from a 25-award G2 haul to Parallel's case study on cutting web grounding spend 70.5%.
Posts to understand this · as of Sep 12
Common Room turns buyer signals from communities and the web into prioritized sales opportunities.
Share climbed from 0.27% in February to 0.66% in July. Zoom's acquisition surfaced in early July, near the top of that run rather than before it, and both outside posts here are about the deal. August fell to 0.24%, and the last 30 days are down another 43%.
Posts to understand this
Exa.ai is a web search API designed for models and AI agents.
Exa has gone from 0.05% in February to near 0.45% on a small base. The $250m Series C at a $2.2b valuation in May is the one big moment; everything since is operators listing Exa as the signals layer in custom sales stacks, beside Sumble, Apify and Parallel. Up 51% over 30 days.
Posts to understand this · as of Sep 12
Warmly identifies website visitors and uses AI agents to start sales conversations.
Warmly climbed through the first half of the year and topped out in July around the HubSpot acquisition, then dropped in August once the news cycle closed. That is the shape of an exit: the MCP and API launch in June, founder tributes at month end, the read on consolidation, then quiet. Down 27% in the last 30 days.
Posts to understand this · as of Sep 8
OpenCode is an open-source coding agent that works with cloud and local models.
OpenCode is up 212% over 90 days, from near nothing in spring to 0.45% in August, with the last 30 flat. Distribution did the work rather than any single launch. The Grok account pitched 4.6 straight to OpenCode users, Supabase named it alongside Claude Code and Codex in its open-sourced benchmark, and Together AI put it in the harness layer.
Posts to understand this · as of Sep 12
Gamma turns prompts into presentations, documents, websites and images.
Gamma peaked in March and gave most of it back by July. No single launch holds the curve; it runs on operators poking at the product's edges, from the n8n integration that made a deck a workflow trigger to Casey Hill's read of the pricing page. The last 30 days are up 53%, but the three-month line is still down 44%.
Posts to understand this · as of Sep 8
AirOps helps content teams measure AI visibility and build workflows to improve it.
AirOps spikes when it publishes data and sinks between. May's peak came with the AirOps Next event and a 5.1 million response study showing only 10% of AI citations point to brand-owned domains. Three months later the share is down 65%. The newest pickup is its 3.5 billion citation report and YouTube's 158% growth in AI citations.
Posts to understand this · as of Sep 8
1mind provides AI sales agents for websites, demos, live calls and customer success.
1mind's high was February, when Nathan Latka reported the $30M round and $400k contracts and Jacco van der Kooij published a year of running his cloned agent. Nothing since has matched that pairing of money and proof. Share has fallen in most months since and the last 30 days are down 47 percent, with the recent post being 1mind's own webinar result.
Posts to understand this · as of Sep 12
Firecrawl turns websites into clean, structured data for models and agents.
Firecrawl went from no share in February to 0.329 in May, dropped by more than half in June, then peaked at 0.388 in July. The July peak lands in the same month as Corey Haines's post on Firecrawl's free-credit onboarding. Share settled near 0.24 in August and the last 30 days are down 4.7 percent.
Posts to understand this · as of Sep 12
Higgsfield generates video and images for ads and social content.
Attention bottomed in April, then climbed through July as Higgsfield opened agent access and creators built on its models. August fell back, and the last 30 days are up about 16 percent, so the current direction is up again. The August hosting note describes a company already scaling after its Series B, not a round landing that week.
Posts to understand this · as of Sep 12
Grok Bot is an always-on agent that learns routines and completes work across existing software.
New to the tracker and up 836% over the last 30 days. SpaceXAI opened the early beta on August 11 and the line starts there. Two things sustained it: users posting what the bot did on its own, like cancelling a service without being asked how, and Stripe giving bots a card to buy with.
Posts to understand this
Claudeforce brings Salesforce customer context and actions into Claude through Salesforce and Anthropic's partnership.
Claudeforce has one full month of data. August opened at 0.16% share with nothing before it, and September is only 12 days in, so there is no second full month to compare against. L30 and L90 both read NEW, so no change figure supports a direction yet. All three posts land inside a two-week window after the Salesforce and Anthropic partnership.
Posts to understand this · as of Sep 12
GPT-6 is OpenAI's frontier model line, tracked as the conversation about its releases, its evals and early use.
Nothing at all until August. A reshared claim that an internal version of Astra had solved ten open problems in math and quantum computing put the name in circulation, and launch partners followed: Legora published evals against GPT-5.6 Sol, fal shipped a build demo. This is days old, so read the L30 move rather than the month.
Posts to understand this · as of Sep 12
Claude Fable is Anthropic's frontier Claude model line, tracked as the conversation about its releases and what people build on it.
Fable went from nothing to 3.73% in June on the Fable 5 launch, ran hot in July, then cooled to 1.35% in August. Up 12.2% over 30 days. September's partial bar is high again, with Cursor shipping Fable 5.1 and calling it the most capable model it has benchmarked.
Posts to understand this
GPT-5 is OpenAI's flagship closed model family, offered as a hosted API across tiers rather than an open, self-hosted release.
GPT-5 is fading as a topic, down about 72 percent over 30 days and about 54 percent over 90. It never had a spike month. Both posts use it the same way, as the number a smaller or specialized system beats: You.com's finance API on FinSearchComp, PolyAI's Raven on customer service benchmarks.
Posts to understand this · as of Sep 8
Claude Opus is Anthropic's top-tier model, built for buyers weighing closed frontier power against cheaper open alternatives.
Opus moves in release spikes and little else. April, May and July each clear 1%, with the Opus 5 launch carrying July, then the line falls to 0.31% in August. Down 64.2% over 30 days. Between launches it appears as a line item in someone else's comparison rather than a subject of its own.
Posts to understand this
Grok 4 is xAI's frontier language model powering the Grok assistant and developer API.
Grok sat near zero for five months, then jumped roughly tenfold in July when Grok 4.5 started topping benchmarks, and held that level through August. Leaderboard posts do the work: AutomationBench, browser use, then CursorBench for 4.6. L30 is up about 101 percent, so the July step is holding rather than fading.
Posts to understand this · as of Sep 8
Claude Sonnet is Anthropic's mid-tier AI model, built for coding, agents and everyday work at lower cost than its top-tier sibling.
Sonnet 5 landed at the end of June and the launch did the work. Wade Foster's AutomationBench numbers and Legora's testing notes both arrived June 30. The curve has come down every month since, and the last 30 days are off 30%. Anthropic making the introductory price permanent in August did not rebuild the June level.
Posts to understand this · as of Sep 12
Muse Spark is Meta's AI model for running coding and other agents.
Muse Spark sat near zero until July, and the last 30 days are up about 50 percent. Meta shipped Muse Spark 1.3 and a personal agent called Muse, Genspark put the model into its chat and code agent, and Surge benchmarked it against Fable 5.1, Gemini 3.8 Flash and GPT-6 Astra. Release week, not a slow build.
Posts to understand this · as of Sep 12
Gemini Flash is Google's fast, low-cost model line, built for high-volume API work and coding.
This is a release sawtooth. Gemini 3.5 Flash from the Google I/O stage drove May, 3.6 Flash drove July with Harvey publishing its own benchmark read, and the early September bar sits high after Databricks offered 3.8 Flash on day zero. The pattern worth noting is that partners, not Google, carry most of the volume. Up about 12% over 30 days.
Posts to understand this · as of Sep 12
Kimi is Moonshot AI's open-weight model line, used for long-context work and coding.
July was the whole story. Kimi K3 took first on the frontend code arena, which David Sacks called concerning, and cost-per-point results kept it in circulation. August halved and L30 is down 77.9%. Harvey post-training Tenet on a K3 base is what remains, a build story rather than a launch.
Posts to understand this · as of Sep 12
Qwen is Alibaba's open-weight model family, widely used for local and self-hosted deployments.
Qwen went from a rounding error in February to the loudest open-weight name by August. Hugging Face's summer report put it first in local inference, and practitioners argued in public about whether the 27B is good enough to ship. The last 30 days are down 20.4% off that high, while the 90-day trend is still up 59.6%.
Posts to understand this
GLM is Z.ai's open-weight model line, used for coding and vision work at low cost.
GLM sat under 0.1% share through May, then stepped up to 0.67% in June when coding tools began shipping it. July and August held that level, helped by a benchmark result that put an open-weight model at the top of a category. The last 30 days are up 54%, to 0.82%, as a newer GLM version lands in more tools.
Posts to understand this · as of Sep 12
Mythos is Anthropic's frontier model line, built for advanced reasoning and agentic work.
Mythos spiked in April on Project Glasswing, the Anthropic model held back to 40 partners for defensive security work, and again in June. Since July it has run near 0.3%, down 57.6% over 90 days. The talk moved from access to arithmetic: Mythos 5.1 now prices the same as Fable 5.1 and GPT-6 Astra.
Posts to understand this · as of Sep 12
Midjourney is an AI image generator known for stylized, painterly visuals used across creative and marketing work.
Share ran high in February, faded through spring, then jumped again in June. The two strongest posts both land on June 18, so that lift looks concentrated in a moment rather than broad. It came back down after: July and August sat near a quarter of a percent, and the last 30 days are down another 10%.
Posts to understand this · as of Sep 12
Join 19,213+ GTM leaders and practitioners. Free, unsub anytime.
Hi, I'm Adam. I'm an independent builder & industry analyst. My goal is to help GTM leaders make sharp decisions in uncertain markets.
This is the methodology for my GTM Market Tracker — the score, the rankings, and the category maps.
Topics measure what the GTM conversation is actually about, at two grains.
Clusters are built from topics, so both appear in one ranked list.
The volume and change measures are both variations on Share of Voice (SOV). The volume is a count of posts fitting the topic or theme, engagement weighted. The denominator is every panel post that matched at least one maintained topic.
Shares do not add up to 100%. A post about coding agents inside GTM engineering is about both, so it counts for both — the number is prevalence, not a slice of a pie.
Beta. These figures move as the panel grows and the definitions sharpen. Every published number carries the registry version it was measured under, and when a change reprices the board I say so in that issue's notes.
I cover products and companies with a central application for B2B go-to-market in the AI era. Most are sales- or marketing-focused (e.g. Gong, Clay), but I also cover broader solutions that are clearly GTM-relevant (e.g. Claude Code, Vercel, Exa).
On the edges my judgement determines what fits (and what doesn't). Like all things, this evolves as I learn and as the market changes.
Brand's share of panel conversation, normalized 0–99 against the field.
My scores are a share-of-voice (SOV) measure, not a rating of the product capabilities. I am not scoring whether a product is good or best. I'm measuring how much it shows up in the GTM conversation. This a useful concept to understand market trends, but isn't a buyer score alone.
Here's how it works:
| Window | What it answers | |
|---|---|---|
| SOV Score | Last 90 days | Where does this brand stand right now? |
| 60-day move | Last 60 days vs the 60 before them | Which direction is it heading? |
The move is a percentage change in share of panel conversation — the brand's share over the last 60 days against its share over the 60 days before that. Both are recalculated nightly, against the panel as it stands that night. The move window ends two days back, because the most recent day or two of collection is still filling in. Scores you see today are as of 2026-09-13.
A voice is a source on the curated panel. It's a person or a publication whose GTM commentary I've chosen to track, across LinkedIn, X, newsletters, blogs and podcasts. It is a carefully curated list or experts and operators, not a crawl of the internet. I have three rules for reading the panel.
Brands sit inside Markets. These are my framework for grouping companies by the AI era "races" and don't necessarily align to traditional industry categories. Two different things can be true of a brand at once:
My highest level organizing framework has a strong bend toward AI. I believe the traditional categorization is breaking down quickly. This is how I see the space at a high level:
| Type | Definition | Examples |
|---|---|---|
| Coding Agents & Experience Builders | AI coding assistants and app builders. | Claude Code, Cursor, Lovable |
| Orchestration | The workflow layer that wires data, agents, and apps into GTM systems. | Clay, Zapier, n8n |
| Agent-Native GTM | Software where agents are core and do the work — purpose-built for GTM jobs. | 11x, Artisan, Attio |
| GTM Developer Tools | The API- and MCP-first primitives used to build and run agent-based GTM systems. | Exa, Vercel, ElevenLabs |
| SaaS for GTM | The GTM application layer — multi-feature software you operate for sales, marketing and CS. Many are going headless and shipping agents. | Salesforce, HubSpot, Gong |
| Creative & Content | AI generation of media assets — images, video, decks, avatars, copy. | Canva, Runway, HeyGen |
| AEO | Platforms that help brands measure & optimize visibility inside AI answers. | Profound, Scrunch AI, AthenaHQ |
| GTM Adjacent | Broad, general-purpose platforms a company runs on (comms, docs, cloud, data) — used by GTM but orthogonal to sales and marketing execution. | Slack, Notion, Snowflake |
Categories are generally organized around use cases, go-to-market motions, or the intersection of the two. That said, there is a balance of many factors in this framework and it ultimately comes down to my judgement.
| Category | Definition | Examples |
|---|---|---|
| GTM Developer Tools | Broad dev primitives GTM teams run through agents and APIs. | Exa, Vercel, Twilio |
| Data & Intelligence | Enrichment, intent, and signal data that feeds the pipeline. | Clay, ZoomInfo, Apollo |
| Selling | Active deal execution — conversation intelligence, coaching, proposals, demos. | Gong, Salesforce, Granola |
| Marketing Execution | Campaigns, content, ABM, and marketing ops. | HubSpot, Klaviyo, Customer.io |
| Advertising | Paid-media execution and measurement — DSPs, paid social and search, attribution. | The Trade Desk, LiveRamp, Metadata |
| Creative & Content | AI generation of GTM media — video, images, decks, avatars, copy. | Canva, Runway, HeyGen |
| GTM Infrastructure | GTM-specific pipes — email deliverability, CDP and reverse-ETL, billing. | Hightouch, Metronome, Paddle |
| Website & Inbound | Websites, chat, and inbound capture that convert traffic into pipeline. | Webflow, Chili Piper, Framer |
| AEO | Visibility inside AI-generated answers. | Profound, Scrunch AI, AthenaHQ |
| Prospecting | Outbound — finding, qualifying, and reaching buyers. | Outreach, Salesloft, Nooks |
| Customer Success | Post-sale onboarding, retention, and expansion. | Intercom (Fin), Gainsight, Zendesk |
| Autonomous CRM | AI-native CRMs that run pipeline, deals, and relationships themselves. | Attio, Day.ai, Clarify |
Products can fit types and multiple categories, but there must be a strong case.
I score Agent Readiness as a secondary factor, but it matters: in an agent era, whether your systems can actually build on a product matters more than its UI. It's an analyst-judged 0–99 score (with an A–F grade) across four surfaces — API, MCP server, SDK, and CLI — plus the quality of the docs behind them. Multiple agents, each with a different persona, assess those surfaces and docs; the Surfaces column shows exactly which a product ships.
Send any suggestions or issues to adam at adamgtm.com. I verify them, write the fix back to the system and publish.
If you're building something interesting in this space, but it's not covered yet, please suggest a product here.
Building or running a product that belongs in the GTM Index? Tell us about it. We review every submission and add the ones that fit.
join 19,213+ gtm leaders, unsub anytime.
