Most account tiering models sort the list by what is easy to query. Pull annual revenue and employee count from the data provider, draw two lines, and call the top slice Tier 1. The spreadsheet looks rigorous. It also tells you almost nothing about which of those accounts has a reason to buy your product this quarter, because revenue and headcount describe how big a company is, not what it needs.
You have this model because it is the one the tools make cheap. Firmographics come pre-filled in every CRM and every enrichment export. Fit against a specific product line does not: someone has to read the account and decide whether the problem you solve shows up there. So the tiering model gets built from the columns that already exist, and the hard judgment gets deferred to the rep, one account at a time, usually never.
This article lays out a tiering model that ranks accounts on fit against what you sell and on dated evidence of change, then shows what that does to coverage. You can run it by hand on a list of fifty accounts this afternoon.
What a standard account tiering model looks like
Read the guides that rank for this phrase and the structure is consistent. Three tiers, sized as a pyramid, each with its own level of investment. One widely read ABM guide sets Tier 1 at deals of $500k+ and 10 to 30 accounts, Tier 2 at $100k to $500k and 100 to 200 accounts, and Tier 3 at 5,000 or more, with a reassessment every quarter. Other guides use a scoring sheet: weight revenue, industry, headcount, technology and intent, sum the score, and cut at the top ten percent.
Nothing in that structure is wrong. The pyramid is a sound way to match effort to value, and quarterly review is better than none. The problem sits one level down, in the inputs. Look at what actually goes into the score at most companies:
- Revenue band. A proxy for how much the account could spend in total, across every vendor.
- Employee count. A proxy for seats, which matters only if you price per seat.
- Industry code. A proxy for fit, assigned by a data vendor at the parent company level.
- Technology installed. A useful signal when the integration is the point, a weak one otherwise.
- Intent score. A third-party estimate of topic research, rarely tied to a specific product line.
Every one of those is a property of the company. None is a property of the company in relation to the thing you sell. Two accounts with identical revenue, headcount and industry code can sit in completely different places for your product: one just consolidated three regional teams into a single function that now owns the problem you solve, the other outsourced that function last year. The firmographic model tiers them the same.
Why size is the wrong first axis
Size tells you the ceiling of a deal. It does not tell you the odds of one. And the odds, at any given moment, are low for most accounts on your list. Research by Professor John Dawes of the Ehrenberg-Bass Institute for the LinkedIn B2B Institute put a number on this: up to 95% of businesses are not in the market for most goods and services at any one time. The reasoning is simple. If companies change providers of a service roughly every five years, about a fifth are in the market in a given year and about one in twenty in a given quarter.
Apply that to a tiering model built on size. Your Tier 1 holds the thirty largest accounts. If the five percent rule holds roughly for your category, one or two of them are in a buying window this quarter, and the model cannot tell you which. The account team then spreads its most expensive effort (executive outreach, custom content, field events) evenly across thirty accounts, twenty-eight of which have no reason to respond.
Meanwhile, somewhere in Tier 3, a mid-sized account just hired a new leader for the function you sell into and posted three roles that describe your exact use case. It gets the automated nurture track, because it is mid-sized.
There is a second cost. Buying groups have grown. Forrester's 2026 buyer research reports that the typical buying decision now includes 13 internal stakeholders and nine external influencers. Working an account properly means mapping and reaching a large group, which makes each Tier 1 slot more expensive to fill. The more each slot costs, the more it matters that the account in it is actually moving.
A tiering model built on fit and evidence
The alternative keeps the pyramid and changes what decides the rows. It scores every account on two axes, both measured against a specific product line rather than against the company in general.
Axis one: product fit
Product fit asks one question: does the problem this product solves exist at this account, and how badly? You answer it by writing down the two or three problems your product line solves, then looking for evidence of each one at the account. Not industry codes. Evidence. A support team growing faster than its tooling, a compliance requirement that just came into force in their market, a migration their engineering blog says is half done.
Score it on three levels:
- Strong fit: you found public evidence of the problem at this account, with a source you could show the buyer.
- Plausible fit: the account looks like your best customers, but you found nothing specific.
- Weak fit: the evidence points away (they solved the problem another way, or the function does not exist there).
If you sell several products, score each account once per product line. An account can be strong fit for one line and weak for another, and a single blended score hides exactly the information the rep needs to open the conversation. Our ICP definition framework covers how to write the problem statements this axis depends on.
Axis two: dated change
The second axis is the one the five percent rule demands. Has something changed at this account recently that makes the problem more urgent? A new executive in the buying function, a funding round, an acquisition, a hiring push, a reorganization, a product launch, a public commitment that your product helps them keep. The trigger events that matter depend on what you sell, which is why this axis also has to be scored per product line.
The test for counting a change is strict, and it is the same test a rep should run before any call. Call it the reason test. You can name:
- What changed. A specific event, not a trend.
- Where you saw it. A source someone else can open.
- When it happened. A date recent enough to matter, usually inside the last 90 days.
Miss any of the three and the account does not score on this axis. "They are probably growing" fails. "They announced a new EMEA headquarters in their August 14 press release and are hiring a regional operations lead" passes.
How the two axes set the tiers
Cross the two axes and the tiers fall out of the grid. Size comes back in only as a tiebreaker inside a tier, which is where it belongs.
| Product fit | Dated change found | Tier | What the account gets |
|---|---|---|---|
| Strong | Yes | Tier 1 | Named owner, multithreaded outreach this month, opener built on the change |
| Strong | No | Tier 2 | Watched for change, light touch, moved up the day a reason appears |
| Plausible | Yes | Tier 2 | One research pass to confirm the problem exists, then promote or drop |
| Plausible | No | Tier 3 | Programmatic coverage, rescored next cycle |
| Weak | Either | Off list for this product | Scored again for your other product lines, not worked for this one |
Three things change when you tier this way.
Tier 1 gets smaller and moves faster. It holds only accounts with both a confirmed problem and a dated reason. That is usually fewer accounts than the old size-based list, and each one comes with the first line of the conversation already written: the change you found. The account team stops spending executive time on accounts that are big but asleep.
Tiers stop being annual. A size-based model barely moves between reviews, because revenue bands do not change much in a quarter. A change-based model moves weekly, because the evidence does. The quarterly review becomes a check that the scoring rules are right, not the only moment accounts can move. For how that interacts with territory ownership, see sales territory planning.
The weak-fit row becomes useful. Under firmographic tiering, a large account with weak fit sits in Tier 1 and burns effort. Here it drops out for that product line and gets rescored for the others. For teams that sell more than one product, that per-product scoring is what turns tiering into an expansion tool, and it is the same logic behind account whitespace analysis: the empty cell tells you where you could sell, the dated change tells you why to call this quarter.
