Account-Based Marketing: How to Tell If It Fits

What account-based marketing changes

Published definitions of account-based marketing largely converge. Researchers have measured the same practice under another name since 1990, across 104 papers reviewed in the Journal of Business Research in 2023. Underneath the agreement sits one change that decides everything else: you stop counting leads and start counting companies.

Account-based marketing treats a named list of companies as the market. The practice targets and engages specific high-value accounts by name. Most published definitions are versions of that sentence.

The change underneath the definition is a measurement change. Ordinary B2B lead generation counts leads: individual people who raised a hand. Account-based marketing counts accounts, so five people from the same company are one number, not five.

An ad account audience list, account levelWhere the change of unit becomes visible

The first two rows target a list of companies somebody chose by name. The rest target people who match a description. Reporting the first two with the counting rules of the last three is how a programme keeps its old numbers and its new label.

That distinction has a practical edge. A lead-counting team celebrates volume from companies it will never sell to, because a lead from a nine-person startup and a lead from a bank both add one.

An account-counting team cannot do that. The list is fixed before the campaign starts, so anything outside it does not register as progress. That makes the difference between an MQL and an SQL a smaller argument than it usually is.

Say you sell warehouse management software and forty companies in your region run warehouses big enough to need it. Under lead counting, a good month is 200 form fills. Under account counting, a good month is movement inside eleven of those forty.

That older name is key account management, and it carries its own literature. So the question of whether concentrating on a few named customers pays has been asked for three decades.

The answers there are more specific than the category guides suggest. Two of them are further down this page.

How many of your accounts are buying now?

One number decides the shape of your programme: the share of target accounts buying now. A target account list is not a pipeline. Multiply it by the share of buyers who are in the market right now, and you get the size of this quarter.

Three tiers

Start with the tiers, because that part is documented. A common framework splits the list three ways, with example account counts attached to each tier.

The three tiers, with account counts attached

1. 1Strategic work treats one company as its own market. It suits enterprise deals with long sales cycles and high contract values, which is the fit condition stated in words.
2. 2Lite covers a cluster that shares a problem. One worked example covers 5 to 15 companies in one industry.
3. 3Programmatic is the rest of the list, run through advertising and email at account level.

Buyers in market now

Now the missing number. Put a hundred companies on a list and it feels like a hundred opportunities, and it is not. Most of them will not buy anything in your category this year, whatever the B2B buyer journey looks like once they do.

John Dawes at the Ehrenberg-Bass Institute put a figure on that for the B2B Institute in 2021.

He started from a simple observation. Businesses replace a major supplier roughly every five years, so about 5 percent of the category is in the market in any given quarter.

The same arithmetic works across annual and quarterly windows. A five-year cycle gives you about 20 percent in a year and something like 5 percent in a quarter.

Same buyers. Two windows.

That distinction matters because the best-known challenge to the figure runs into it. Forrester analyst John Arnold published a post in May 2025 arguing that 5 percent is too low, and he has a measurement behind the argument.

Annual B2B martech buying rates against the same rule

1. 1Twenty percent of organisations plan to change provider, and 78 percent of those inside twelve months. That is where 15.6 percent a year comes from.
2. 2The category spread is the useful part: 23 percent for ABM platforms, 21 for B2B data providers, 18 for customer data platforms.
3. 3The closing advice is a range, not a constant, 15 to 30 percent, and only for marketing technology.

Read the two windows together and the disagreement gets smaller. The 95:5 rule implies about 20 percent in a year, while a separate measurement put B2B martech at 15.6 percent.

Comparing an annual number with the quarterly 5 percent makes it read as a correction. That is why it reads as a correction. On the same window the two sit close, and his category figures are the better input because they belong to a named market.

Neither figure is your number. Your category has its own replacement cycle. Ask your last twenty customers how long they had their previous supplier, and the median gives you the share.

For example, say that median comes back at three years. A third of your category is in the market this year, not a fifth. Your quarter holds about eight accounts in every hundred, not five.

Work out whether it fits

Four inputs, and you can get all four from records you already have. We built the arithmetic below, and every input in it is something you can look up instead of estimating.

Run it before you buy anything. Six steps, in order.

1. Count the realistic buyers

   How many companies could plausibly sign, ever. Not the total addressable market from a slide. The list you could write out by name.

2. Find your replacement cycle

   Ask your last twenty customers how long they had their previous supplier. The median is your cycle. Divide one by it to get the share in market per year.

3. Multiply

   Realistic buyers times annual in-market share gives the accounts in play this year. Quarter it for this quarter.

4. Divide your budget by that number

   That is what you can spend on each account in play. If the figure will not cover a piece of research and the B2B content marketing that goes with it, your tier one list is too long.

5. Check the deal against the cost

   One-to-one work is a person's time. If a won deal is worth less than a few weeks of that time, run the programmatic tier and skip tier one entirely.

6. Name the accounts you cannot afford to lose

   If three accounts would be more than a third of revenue, you are buying a risk as well as a strategy. The next section prices it.

The last row is where most plans break. Twenty tier-one accounts sounds modest until you divide the budget by it and see what is left for the 220 companies that are also real.

Which twenty is a lead scoring question, applied to companies instead of people.

A small realistic universe with long cycles and large deals is the case account-based marketing was built for. A large universe of small deals is the case where ordinary demand generation wins, and demand generation vs lead generation is the split to read next.

What it costs and what it concentrates

Two costs, and only one of them is a budget line. The first is what the programme takes as a share of marketing spend. The second is what happens to the business when your revenue sits in fewer hands.

Budget share

Start with the budget. Momentum ITSMA runs an annual benchmark of roughly 300 ABM leaders, and the share it reports moves with how far along the programme is.

Notice what the 66 percent implies. At full scale this stops being a tactic you add and becomes the thing your marketing is. That is a decision about the company, not about the quarter.

Say you spend $180,000 a year on marketing. At the early-stage share, $47,000 goes to named accounts. At full scale it is $119,000, and the rest of your B2B demand generation lives on what is left.

That is the budget line. The second cost is not one.

Concentration risk

Now the second cost. Concentrating revenue in a few accounts changes the business, and two peer-reviewed measurements say what it changes.

Take the two findings in order. Patatoukas compiled three decades of supply chain relationships and found that suppliers with concentrated customer bases earn higher accounting returns, with less operating expense per dollar of sales.

That is the efficiency case for account-based work, measured outside marketing. Serving twelve buyers well costs less per dollar than serving twelve hundred.

Dhaliwal and colleagues measured the other side. Suppliers with concentrated customer bases carry a higher cost of equity capital. The effect is strongest where the supplier is likely to lose a major customer, or would be hurt badly by losing one.

One detail in that paper is the condition you need. Suppliers concentrated on safer government customers had a lower cost of equity. So the penalty tracks how replaceable the customer is, not concentration by itself.

Say you name your own concentrated accounts and price the programme against them: that is our B2B PPC service.

Avoid these five failures

Each of these is cheap to avoid at the start and expensive to fix in month nine, and the first two are versions of the same mistake.

Five, and the first two are the same mistake.

The list is too long. A hundred tier-one accounts is a segment with a new label, and it produces generic work at high cost.

The measurement never changes. Sales and marketing agree on the account list and keep reporting MQLs, so nothing about the reporting tells you whether the account moved.

The whole budget goes to accounts in market now. Dawes's point cuts both ways. The 95 percent who are not buying this quarter still choose from what they already recognise, and lead nurturing is what keeps you in that set.

The evidence comes from one side of the table. Momentum ITSMA reports revenue growth across ABM programmes. That finding is useful, and it comes from a survey of the people running the programmes.

That is not a reason to discard the finding. Check it against your own closed-won records before you put it in a plan.

And the quiet one. The account list is never revisited. For example, a company on your tier one list that was acquired in March is still getting the plan you wrote for the company it used to be.

Which number and which window

Three measures, and the window matters as much as the number. Read a quarterly figure against an annual expectation and you will kill a programme that was working.

Measure these three

Measure account coverage first. Take the share of the buying group at each target account that has engaged with anything, because one contact is one person and an account is a group.

A marketing qualified lead is still a person, so it is the wrong unit here.

Then measure movement between stages at the account level, not the lead level. The stages come from the B2B marketing funnel, and one account moving from unaware to a first meeting is the unit.

Then measure win rate and deal size against your own pre-programme baseline. Not against a published benchmark, because the published ones come from surveys of programme owners.

That last row is where the arithmetic from the fit test comes back. Say your quarter holds twelve accounts in play.

With twelve accounts, the difference between three wins and two is too noisy to read. A quarterly review will still read them as 25 percent against 17 percent, a gap of eight points.

Print the denominator.

Report the account count beside the rate every time you show it. A win rate with the denominator printed next to it is a number someone can argue with. That is the point.