In partnership with
Beyond the Newsletter with Fractional ABM
Beyond the newsletter, I occasionally work with B2B SaaS teams, typically Series A through C, who are building or rebuilding their ABM motion with or without a full-time person in seat. If you're in that situation, or heading toward it, reply to this email.
Happy to talk through what you're working on.
One of the most common traps with running a closed lost ABM strategy is when you treat the entire list like they have the same problem.
A deal lost to a budget freeze in Q3 is not the same deal as one lost to a product gap. A prospect who went dark is not the same prospect as one who chose a competitor on price. When you send the same message to all of them, you're not running re-engagement, you're running targeted demand gen.
Last week, I walked through how to audit your closed lost deals and pull the real reason you lost each one (not the CRM dropdown, the actual story from AE notes and call recordings). If you ran that process, you now have at least 20–40 accounts with specific, usable lost reasons.
This week is what you do next.
The move: cluster before you build
Before you write a single email or set up a single LinkedIn campaign, you need to know which accounts share the same underlying concern. Because the messaging for "we lost on price" is completely different from "we lost because the integration wasn't built yet" and both are different from "we lost because they froze the budget."
When you cluster first, you can end up with 3–5 distinct groups, each with its own blocker, its own evidence requirement, and its own re-engagement timing. You also stop wasting SDR time on accounts whose blocker hasn't actually changed.
Three prompts, one motion per cluster
I built a three-prompt stack in Claude that produces a complete multi-channel play for each cluster. You run them in order, once per cluster. Here's what each one does:
Prompt 1 takes your account list with real lost reasons and returns 3–5 clusters, each with a name, the underlying buyer concern, a messaging angle, and the single best piece of evidence to bring into re-engagement.
Prompt 2 takes a single cluster's output from Prompt 1 and drafts a 3-touch outbound sequence: two emails and a LinkedIn DM. Tone is peer-level, not sales-y. Each touch is written around the specific blocker and what's changed since the deal closed.
Prompt 3 takes that same cluster context and produces the messaging outlines for LinkedIn ad creative, a custom landing page brief, and an event or roundtable concept. Every channel anchors to the same underlying concern and the same evidence.
Run Prompts 2 and 3 in parallel (separate conversations for each). By the time you're done, you have a complete play for that cluster: direct outbound, paid social, web experience, and an in-person activation, all saying the same thing to the same people.
Total time to go from audit output to a briefed multi-channel play: under two hours.
Access the full prompt stack here → Copy the prompts and instructions
The doc includes the exact copy for all three prompts, sample output so you know what you're getting, and notes on when to run each one.
If you already ran last week's audit, you can start Prompt 1 today.
