Before you greenlight another agentic GTM pilot, it's worth asking whether your data, response times, and channel mix can actually keep up.
This week's edition covers:
- Kevin's Take: agentic GTM falls apart when your data lives in twelve places, here's what to fix first
- The Signal: Signal: Clay's own outbound agents got ignored by two-thirds of leads, plus fresh benchmarks putting LinkedIn ahead of Google and Meta for B2B
- Tools & Tactics: OpenAI's new conversational ads and Shopify app open a fresh paid surface, and Claude Opus 5.5 is worth a look before your competitors test it
- Quick Links: Marketo's co-founder launches an AI-native automation platform, a practitioner's Grok bot stack for GTM work, and IDC data showing 68% of buyers now use AI to compare vendors
Kevin's Take
Agentic GTM won't work if your data still lives in twelve places
I think the vision George Schildge lays out in this Matrix Labs piece on Agentic GTM is directionally right, that we're moving from a chain of fragmented point tools with manual handoffs between SDR, marketing, ops, and sales into a governed system of agents that owns the whole revenue workflow end to end, but I also think most companies are about to spend a lot of money learning that you can't just drop agents on top of the mess they've built over the last decade and call it a new operating model.
The reason handoffs are broken today is not because we lacked an agent to run the handoff, it's because the data on either side of the handoff doesn't agree. Marketing's account definition isn't sales' account definition, the CRM says one thing and the product usage database says another, the enrichment tool overwrites the field the SDR just corrected, and the intent signal fires three weeks after the deal already closed.
If you put an agent on top of that, all you've done is automate the confusion at a much higher speed. The agent will confidently qualify, route, and email its way through bad data, and now instead of one rep being wrong on a Tuesday you have a system being wrong at scale on a Tuesday.
What we've been finding in our own work rebuilding internal tooling is that the agent layer is actually the easy part now, the hard part is the substrate underneath it, one source of truth for accounts, contacts, activity, and stage definitions that every agent reads from and writes to the same way. That's unglamorous plumbing work and it doesn't demo well, but it's the thing that decides whether agentic GTM becomes a real operating model or just a more expensive version of the fragmented stack we already have.
Before anyone at our shop pitches leadership on agentic GTM, I'd want us to answer one question first, which is whether we could get two agents to agree on what a qualified account even is today. If the answer is no, is that where the work actually starts?
That's it for this week. Talk soon.
— Kevin Kerner, CEO, Mighty & True
The Signal
Clay's own outbound agents got ignored by two-thirds of leads (3 min read)
Clay ran AI agents to request demos from 6,346 companies and found two-thirds never responded, exposing how poor speed-to-lead and outbound follow-up still is even as AI floods the top of funnel. The post also covers Opus 5.5 pricing and open GTM-engineer roles.
Why it matters: If your team is investing in AI-driven outbound, this data shows your response process is probably the actual bottleneck, not lead volume.
New benchmarks show LinkedIn beating Google and Meta on B2B ROI (1 min read)
Dreamdata's new 2026 LinkedIn Ads Benchmarks Report shows LinkedIn outperforming Google and Meta on ROI for B2B campaigns, with specific ABM tactics ranked by effectiveness.
Why it matters: If you're allocating paid spend across channels, this gives you fresh benchmark data to justify shifting budget toward LinkedIn and away from underperforming ABM tactics.
Tools & Tactics
OpenAI launches conversational ads and a Shopify ChatGPT ads app (1 min read)
OpenAI is testing Sponsored Agents, letting users click an ad in ChatGPT and start a conversation with a business-sponsored agent, alongside a new ChatGPT Ads app in the Shopify App Store for US merchants.
Why it matters: If you run paid media, this is a new ad surface and conversational ad format you'll need to test before competitors lock in early placements.
Anthropic ships Claude Opus 5.5 (1 min read)
Anthropic has released Claude Opus 5.5, the latest version of its flagship model, though specific capability improvements aren't detailed in this announcement.
Why it matters: If your team relies on Claude for content, campaign, or agent workflows, you'll want to test what's new before your competitors do.
Quick Links
News & Trends
- New study maps how 189 B2B buyers actually compare vendors — Carbon's quantitative study of 189 B2B decision-makers examines how buyers compare providers, reconsider partners, and decide what warrants a closer look
- IDC data: most B2B buyers now use AI to evaluate vendors — IDC's 2026 buyer survey finds 68% of B2B buyers use AI always or often to compare vendor products, 70% for industry research, and 46% turn to tools like ChatGPT during evaluation.
Resources & Tools
- Marketo's co-founder launches AI-native marketing automation platform — Marketo co-founder Jon Miller has launched Phave, a new marketing automation platform built to reason about complex B2B buying groups rather than follow static rule-based workflows (via Mike Pastore)
- A practitioner's Grok bot stack for automating GTM work — Maja Voje shares her first-person breakdown of building a Grok-based automation bot for GTM tasks, including the specific stack and templates she used over a month. (via Maja Voje)
Read Inference — long-form essays
Inference is our long-form essay series on the future of B2B marketing. Where Drag and Drop is the weekly signal, Inference is the deeper analysis. Subscribe to get the new pieces in your inbox.
Want a second set of eyes?
Running a marketing team right now feels like rebuilding the plane mid-flight. If you want a second set of eyes on what is actually working (and what you can safely cut), we do free 30-minute marketing audits. No pitch deck, just a honest conversation. Reply to this email and we will find a time.