ABM

How to build a target account list from your ICP

An ICP is a description. A target account list is a set of company names you can act on. This is the method for getting from one to the other.

ICP TO NAMED ACCOUNTS fit filters TIER 1 TIER 2 TIER 3 EXCLUDED Wide pool in, short named list out

"Mid-market SaaS companies with a sales team" reads well in a deck and is useless in a spreadsheet, because nobody can open it and start working. A target account list makes the same idea concrete: named companies, ranked, with the right people attached. Knowing how to build a target account list is mostly a matter of order, and the order below works.

This is the method we use. You can run it yourself or hand it to us.

How to build a target account list in eight steps

Step 1. Turn the ICP into filters you can test

Write every ICP trait as something you can check on a company's public pages. Industry, size band, country, business model, the tools they use, the roles they hire. A trait you cannot verify from public pages cannot work as a filter.

Split your traits into two groups: must-haves that remove a company outright, and nice-to-haves that raise its rank. Keep must-haves short. Every extra one cuts the pool.

Step 2. Name the buyers, not just the companies

An account needs named buyers before anyone can act on it. Decide which roles matter: who signs, who uses it, who can block it. You will need titles to search for later, and the list of titles shapes the ads you run against the accounts.

Step 3. Pull a wide candidate pool

Start bigger than you need. Use open-source data and several publicly available sources: company websites, business registries, industry directories, event exhibitor and speaker lists, job boards, press. Each source has blind spots, so combine several. Data sourcing describes the kinds of sources we draw on.

At this stage you only want company name and website. The website domain is your unique key for everything after.

Step 4. Enrich and apply the must-haves

For each candidate, fill in the fields your filters need: size band, country, industry, what they sell. Then cut everything that fails a must-have. This step is tedious and does most of the work for quality. A company that fails a must-have is removed, not kept as a maybe.

Check a sample by hand. Open twenty or so companies from the survivors and read their sites. If several are wrong, your filter is loose. Fix the filter, not the rows.

Step 5. Score and tier

Use the nice-to-haves to rank. Keep the scheme simple enough that a teammate can explain why a company sits where it does.

TierWhat it meansTypical treatment
Tier 1Meets every must-have and most nice-to-havesPersonal outreach plus ads
Tier 2Meets every must-have, a few nice-to-havesAds and light sequences
Tier 3Meets every must-have, little elseAds only, review later

Step 6. Remove what you should not touch

Take out current customers, open deals, competitors, partners, and anyone sales has already marked as a no. Deduplicate by domain, because the same company often appears under two spellings or a parent and a subsidiary. This step prevents many awkward overlaps later.

Step 7. Attach the people

Now find the roles from step 2 at each account, tier 1 first. Capture name, title and a work email where one can be found in publicly available sources. Then verify the email addresses before anyone uses them. Our email verification service does this as a separate step if you already have contacts.

Step 8. Use it, then keep it alive

Account lists decay. Companies are acquired, shrink, or pivot. Review tier 1 on a schedule and refresh the rest less often. Also record the outcome per account so the next pass can use it: which tier replied, which filter turned out to be useless.

If you plan to run LinkedIn ads against the list, note that a company-level upload and a person-level upload behave differently, and small lists can run into audience minimums. We cover that in the matched audience size post.

Illustrative clip with schematic bar widths, not real counts.

An illustrative scenario

A company sells onboarding software to HR teams. Its ICP says "growing companies with remote staff". That is too soft to filter on, so the team rewrites it. Must-haves: a certain size band, offices or hiring in more than one country, an HR function that exists on paper. Nice-to-haves: open HR roles, a recent funding or expansion announcement, an HR tool already in use.

Now each trait maps to a publicly available source. Size from the company site and registries, hiring from job boards, expansion from press. The pool becomes checkable. No numbers are given because the size depends on your market.

How big should a target account list be?

Keep it smaller than you want. A list you can review, tier and personally understand beats a giant one nobody has read. If your sales team can name ten accounts from the list without looking, it is probably the right shape. If tier 1 is too large to treat differently from tier 2, your must-haves are too loose or your tiers are meaningless.

For LinkedIn ads there is a second constraint: the audience has to be large enough once matched. If your list is very small, think about adding a second ring of look-alike accounts, or running the campaign at company level.

Which public sources help most

  • Company websites for what they sell and who they sell to.
  • Job boards for growth signals and the roles they are hiring.
  • Event exhibitor and speaker lists for companies actively spending on your category.
  • Industry directories and registries for size, location and legal basics.
  • Press and announcements for expansion, funding, new products.

Common mistakes

  • An ICP with no must-haves, so the list is just "everyone".
  • Skipping the sample check, so a loose filter produces a thousand wrong rows.
  • Building contacts before tiers, then researching people at companies that never mattered.
  • Treating the list as finished when it is a working document.

Doing it with us

We run this process end to end and deliver the named accounts, tiers and contacts as files you own, built from open-source data. Start from ABM Target Accounts, or see the company lists service if you only need the accounts for now.

Frequently asked questions

What is the difference between an ICP and a target account list?

An ICP describes the kind of company and buyer you want. A target account list names the actual companies that match it, ranks them into tiers and attaches the people to contact.

How many accounts should a target account list have?

Keep it small enough that you can review and understand it. If your sales team can name ten accounts from the list without looking, it is probably the right shape. If tier 1 is too big to treat differently from tier 2, tighten your must-haves.

How often should I update a target account list?

Review tier 1 on a schedule and refresh the other tiers less often. Companies get acquired, shrink or change direction, so treat the list as a working document and record which tier and which filter performed.

Have an ICP and no list yet?

Send us the ICP. We will tell you how we would build the account list and what a sample would look like.