Today’s plays › All plays › Prospecting & research
Before a product launch, have AI sort every account by whether they build what you sell
Picked and checked by Olga Pearson, SalesUpLevel
Get tomorrow’s play in your inbox. Free ↓
Reported in Clay's case study: a ranked list of about 13,000 accounts was in Salesforce within two days. Clay-built lists helped identify over 55% of Agent Analytics revenue in its first quarter and drove two-thirds of its pipeline, and the launch beat its Q1 sales target by more than 200%.
Case study: Pendo (customer) and Clay (vendor). For a new product (Agent Analytics), Pendo's GTM engineer had AI agents read job postings, company websites and 10-K filings to find which companies were building their own AI agents and what they were building. Each account got a label (AI-native, AI-forward and so on) and a fit rank, and the list went into Salesforce as the one source of truth, so BDRs walked into calls knowing what the account was building.
By Zach Baumgarten (Pendo), in a Clay case study, GTM Engineer, Pendo · Case study (Clay customer story) ↗ · Shared by Clay on LinkedIn Jun 18, 2026 (page has no date)
If you sell: open every launch call with what the account is already building
- Write down the one thing a perfect buyer of the new product is already doing (for Pendo: building AI agents).
- Have an AI agent check the public evidence for each account: open job posts for that work, the website, and the annual report.
- Ask for a label and a one-line reason per account, not just yes or no.
- Rank by fit and push the list and labels into your CRM as custom fields.
- Give each rep the 'what they are building' line before the first call.
- After the first quarter, compare revenue and pipeline from the AI list to everything else.
You are researching [Company] for our new product, [product], which helps teams that [what the buyer is already doing]. Check their open job postings, their website and their latest annual report or 10-K. Answer: 1. Are they doing [that work] today? Label: building / exploring / no sign. 2. What exactly are they building? One sentence, with the source link. 3. Fit score 1 to 5 and the one reason. If you find no evidence, say 'no sign' and do not guess.
If you lead a team: roll it out
- Have ops build the labeled list before launch day, not after.
- Track launch revenue from the AI list separately from the rest.
Watch out for
- Vendor case study; Clay is selling its research agents.
- The revenue share is for one product launch in one quarter, with no baseline from a past launch.
- Job posts and filings show intent, not budget. Reps still need to qualify.
Source: Clay customer story 'Pendo' (shared by Clay on LinkedIn, Jun 18, 2026). Case study: customer Pendo, vendor Clay. Results are as reported in the vendor case study; we did not measure them.
Get tomorrow’s play before your first call
Free. One sourced play a day.