The TAM Mapping Playbook

Turn closed-won patterns into a scoring model.

• Base the model on three data sets: firmographics, technographics, and account-fit signals.

• Backtest the model against closed-won. 95%+ should qualify if the model is right.

Source companies from multiple sources so the TAM isn't biased by one filter.

• Pull broad lists from databases like Apollo, AI Ark, Ocean.io, DiscoLike, Sales Navigator, and StoreLeads.

• Pull scraped lists via Apify, Octoparse, Serper, or Python.

• Avoid over-relying on native filters (many software companies aren't tagged under "software development").

Turn multiple raw lists into one deduped source of truth.

• Dedupe across sources.

• Merge fields to consolidate columns.

• Find missing domains and validate that every domain is operational.

Filter out non-fit companies before running expensive enrichment.

• Design a research agent that does binary qualification (e.g. "is this actually an eCommerce brand?").

• Use lower-cost models like 4o-mini or Haiku given the volume.

Layer the data every qualified account needs.

• Run enrichments in Clay, starting with the ones most likely to disqualify a company.

• For qualified accounts, add custom research relevant to your GTM.

Turn enriched data into a priority score.

• Score every surviving account into Tier 1, Tier 2, or Tier 3.

• Score on account fit only. Never mix in intent signals at this stage.

Find the right titles at every qualified account.

• Analyze titles typically involved in your closed-won deals.

• Split them into decision makers, champions, and influencers.

• Source those titles at every qualified account.

Load the TAM into the CRM so reps build their segments from it.

• Upload companies, contacts, and custom properties into HubSpot or Salesforce.

• For large TAMs, upload only Tier 1 to start.

• Turn the TAM Map workflow into an automated CRM enrichment flow so every new record runs through the same research.