Account selection: how to choose the accounts your ABM programme can actually win
Fit first, signals second, capacity always
Account selection is the single decision that determines whether your ABM programme works. Pick the right accounts and average execution still produces results. Pick the wrong ones and flawless execution produces expensive noise. This guide covers how I select and tier target accounts in practice: fit first, signals second, capacity always.
Why selection beats execution
Most ABM programmes that fail were lost before the first campaign went out. The list was built on gut feeling, on whichever logos sales wanted to chase that quarter, or on a raw firmographic export nobody sanity-checked. Everything downstream, the content, the plays, the ad spend, then works on accounts that were never going to buy.
The uncomfortable rule: your ABM programme is only as good as the worst account on your list. Every weak account eats budget and, worse, sales attention that a better account deserved.
Start with evidence, not ambition
Your ideal customer profile (ICP) should come out of your own won and lost deals, not out of a workshop about who you would like to sell to. Pull your last two or three years of closed business and look for the pattern behind the deals that closed fast, expanded later and renewed:
- Firmographics. Segment, size, geography. Necessary, but the weakest predictor on its own.
- Situational fit. What was true at the moment they bought? A legacy system at end of life, a new leadership team, a regulatory deadline, a merger. These conditions repeat across your market and are findable.
- Relationship reality. Did you already know people there? Warm accounts close at a different rate than cold ones, and your tiering should reflect it.
When I ran ABM in healthcare IT, the pattern that mattered most was situational: hospitals mid-way through an EPD consolidation behaved completely differently from those that had just finished one. No firmographic filter would have caught that.
Tier before you target
Not every account deserves the same investment. The classic three tiers still work because they map investment to expected value:
- Tier 1 (one-to-one): a handful of accounts, each treated as a market of one. Dedicated research, personalised content, named owners in both sales and marketing. If you cannot describe each account's situation in two sentences, it does not belong here.
- Tier 2 (one-to-few): clusters of 5 to 15 accounts that share a situation, for example "regional care groups facing the same compliance deadline". Same play, lightly personalised per account.
- Tier 3 (one-to-many): the wider ICP universe. Programmatic, mostly automated, and the pool you promote accounts from when signals appear.
A scoring model you can defend
Keep scoring simple enough that sales can challenge it. Two axes are enough: fit (how closely the account matches your evidence-based ICP) and signal (is anything happening right now: intent data, hiring patterns, technology changes, funding, leadership moves). High fit plus live signal is Tier 1 or 2. High fit without signal stays in Tier 3 until something moves. Signal without fit is a distraction, however tempting the logo. I cover the signal side in depth in the intent data guide.
Let capacity set your list size
The most common mistake I see is a target list sized to ambition instead of capacity. The maths is simple: a serious Tier 1 account costs several hours of combined sales and marketing attention per week. If your team can genuinely work 8 accounts at that level, a 25-account Tier 1 list means every account gets a third of what it needs, which in ABM rounds down to nothing.
Better a list you can flood than a list you can sprinkle.
Select together or fail separately
Account selection is a joint decision with sales, full stop. Marketing brings data and pattern analysis, sales brings territory knowledge and relationship reality. If either side owns the list alone, the other side quietly ignores it, and the programme is dead within a quarter. How to organise that is its own discipline: see the sales and marketing alignment guide.
Revisit quarterly, ruthlessly
A target account list is a living document. Every quarter: demote accounts that showed no engagement despite real effort, promote Tier 3 accounts that lit up with signals, and re-check whether your ICP evidence still holds. The discipline to remove accounts is what keeps the list honest.
Frequently asked questions
How many accounts should an ABM programme target?
As many as you can genuinely work, and no more. A dedicated team can typically handle 5 to 15 one-to-one accounts and a few hundred programmatic accounts. Size the list to your capacity for sustained attention, not to your ambition.
What is the difference between an ICP and a target account list?
The ICP describes the characteristics of your best customers, based on evidence from won and lost deals. The target account list contains the specific named companies that match that profile and show enough fit and signal to justify investment now.
Should sales or marketing own account selection?
Neither alone. Marketing brings data and pattern analysis, sales brings territory and relationship knowledge. Select together in a working session and review the list quarterly together, otherwise one side will quietly ignore it.
How often should I update my target account list?
Review quarterly. Demote accounts that show no engagement despite genuine effort, promote accounts from your wider ICP universe when buying signals appear, and validate that your ICP evidence still holds.
Account-based marketing strategist and full-stack marketer with 20+ years of B2B experience, including ABM programmes in healthcare IT. Writes about account-based marketing and sales, and builds his own marketing and AI tooling. Find me on LinkedIn.