What is a TAM analysis, and how do you build one?

A TAM analysis lists every company you could sell to, then scores each one on fit and timing so the team knows where to focus. TAM stands for total addressable market. Bought lists cover the obvious firmographics, while a useful TAM analysis scores each company against the signals that showed up in your past wins.

What a good TAM analysis includes

A TAM list your team can actually use has a few parts:

• The whole market, not a self-selected slice of it
• A unique ID for each company, such as a license number, NPI or EIN, so you can join data sources without guessing
• A score for each company, based on the conditions behind your past wins
• Evidence for each score, so a rep can see why an account ranks where it does
• Contacts, found once you know which accounts matter

Why bought lists fall short

Jordan Crawford calls most bought prospect lists a photocopy of a photocopy. Someone scraped a source and resold it, and by the time it reached you the names were stale and you couldn't see where the data came from.

He ranks data sources by how close they sit to the real event. At the top are government registries with a unique ID, such as ATF firearms licenses, the NPI provider file or a state license roster. Next come legally required disclosures like SEC filings and state business registrations. Scraped aggregators and B2B databases sit at the bottom. Much of the best data for a TAM is public. See data sources for prospecting for more.

How to build a TAM list, step by step

1. Start from your wins and losses. Pull everything you know about recent closed deals, including the ones you lost. Ask what public data would have shown each customer was about to buy.
2. Pick the source that covers the whole market. For a licensed trade, that's often a state or federal registry. For software buyers, it might be technology or job posting data.
3. Join sources on clean keys. Jordan once built a healthcare list that joined three federal files on the NPI number with no fuzzy matching. A different build with no shared key produced a 196,000-row over-merged list he had to split back apart.
4. Score and disqualify. Score every company against the signals from step one, and drop the ones that don't match.
5. Find contacts and keep it fresh. Add people once you know which accounts qualify, and refresh the sources on a schedule.

Examples of TAM builds

The Blueprint homepage shows real projects from the last 12 months:

• 47,228 third-party logistics warehouses scored against 20 signals in about 48 hours. 93% were disqualified, and 3,326 survived.
• 240,094 licensed HVAC contractors across all 50 states, built in about 24 hours from bulk downloads, web scrapers, FOIA requests and municipal records.
• 1.17 million companies checked against nine signals to find which healthcare practices ran a specific business model, each enriched for less than two cents.

The scoring is where a TAM analysis becomes useful. A list of every warehouse tells you little. A list of the 3,326 that match your past wins tells the team where to start.

TAM, segments and signals

A scored TAM is the base for a Pain-Qualified Segment, a group of companies that share a specific, data-detectable problem. Generic timing signals can help you decide when to reach out, but Jordan builds the segment on pain. See buying signals vs pain signals.

How Jordan builds TAM lists

Jordan builds bespoke TAM and lead lists from public web data, finding and scoring 100,000 to 500,000 companies with AI. In an advisory engagement, the scored TAM lives in a shared context repo next to campaigns, customer context and playbooks for your vertical, so your team can keep running it after the engagement ends.

If you lead a revenue team and want help building yours, see fractional GTM engineering advisory. To see what the finished output looks like, browse the Blueprint Showcase.