Account-Based Marketing
Choose the accounts first, then market to the committee inside each
ABM reverses the usual order. Instead of generating enquiries and hoping some of them fit, you decide which companies are worth winning and build marketing for the several people inside each one who will actually decide. It costs a great deal per account, which is exactly why the selection matters more than anything that follows it.
The inversion
One method starts with a message, the other starts with a list
These are not competing philosophies and most B2B businesses need both. They are different orders of operation, and treating one like the other is where the money goes.
| Dimension | Demand generation | Account-based marketing |
|---|---|---|
| Where it starts | With a message published broadly, to find the people who respond to it | With a list of named companies chosen before any message exists |
| Who you are addressing | An individual who completed a form and identified themselves | Several people inside one company, each with a different reason to care |
| What gets measured | Leads, cost per lead, conversion rate through the funnel | Account engagement, committee coverage, opportunities opened, pipeline and wins |
| What early progress looks like | Volume rising while cost per lead holds steady | More roles from the same account engaging, and sales securing a first conversation |
| Cost per account | Low, because one asset serves everybody who finds it | High, because research, tiering and personalisation are done per account |
| Who owns it | Marketing, working to a lead or pipeline target | Marketing and sales jointly, working the same list with the same definition of progress |
| When it is the wrong choice | When your market is a few hundred companies and you can name most of them | When contract value cannot pay for the work spent on the accounts you do not win |
Before anything gets built
Four decisions that determine whether this works
Which accounts, and on what evidence
Built from your own closed-won and closed-lost history rather than from a firmographic filter. What did the companies that bought have in common, what did the ones that stalled have in common, and how many companies of that description actually exist? This analysis returns more than anything else on the engagement.
How many tiers, and what each one receives
A handful of accounts can justify genuinely bespoke work. A larger group can be served by programmes built around a situation several accounts share. A wider set gets targeted advertising and content with light personalisation. Deciding the tiers is how the budget stops being spread evenly across accounts of very unequal value.
Who owns the account, and at which moment
ABM fails on ownership more often than on creative. Marketing warming an account that sales is already in conversation with, two teams contacting the same person in a week, a hand-off with no defined trigger. This gets agreed in writing before the first campaign, including what happens when the two teams disagree about an account.
What counts as progress this quarter
Engagement across roles, meetings held, opportunities opened, pipeline created and won. Agreed and put on the dashboard before the programme starts, because a reporting change made after a disappointing quarter always looks like an excuse, however correct it is.
What we see go wrong
Where account programmes come apart
- The list was never really agreed.
- Sales has a target list, marketing has a segment, and the two overlap by about half. Programmes run against accounts sales has already written off, while accounts sales is actively working receive nothing. Nobody notices for a quarter because the two teams are reading different reports.
- Only the champion was ever spoken to.
- One enthusiastic contact carries the whole case internally, translating it for a finance director and a security team they do not report to. The deal reaches the final stage and stalls on an objection from somebody marketing never addressed. This is the most expensive failure in ABM because it happens late, after most of the cost has been spent.
- Personalisation went as far as the company name.
- A generic asset with a logo dropped into the header and a first line naming the industry. Senior buyers recognise this instantly, and it reads as less effort than sending nothing. Real account work means showing that you understand something specific about their situation, which requires research time that has to be budgeted rather than assumed.
- The list never changes.
- Accounts get added and none are ever removed, so the programme spreads thinner every quarter while the cost rises. An account that has shown no engagement across several quarters and has no sales activity is consuming budget that a live account could use. Removal has to be a scheduled decision, not an admission of failure.
- Lead count stayed on the dashboard.
- The programme is reported alongside demand generation using the same measure, and it loses every month because it was never designed to produce that number. Within two quarters somebody senior concludes ABM does not work, when what actually happened is that it was assessed with an instrument built for a different method.
The part usually left out
This is expensive per account, and that is the method working as intended
The only question worth settling first is whether one won account is worth enough to pay for all the effort spent on the accounts you do not win. If it is not, no amount of execution quality will rescue the arithmetic.
What the threshold depends on is specific to you: average contract value, gross margin, how long a deal takes, how many companies of the right description exist in your market, and how much time your sales team can genuinely give to each one. We would rather work that out with you before an engagement than publish a number that would be wrong for most businesses reading it.
There is a version of this conversation that ends with us recommending you do not run ABM. It ends there reasonably often. A business selling a modest contract to a very large market is better served by a broad demand programme, better tracking and a faster sales process, and we will say so rather than sell a method that will not repay its cost.
Running it
How an account programme is actually built
Analyse what has already been won and lost
Your own history first. Which companies bought, what they had in common, which stalled and at what stage, what the losses shared. This is where the ideal customer profile comes from, and a profile built any other way is a description of who you would like to sell to.
You get: An evidence-based ideal customer profile
Build and tier the account list, jointly
Score the addressable set against the profile, put it in front of sales, and argue it out until both teams will defend the same list. Then tier it by what each account justifies, and write down the criteria so the tiering can be repeated next quarter without the same debate.
You get: A tiered target account list with agreed selection criteria
Map the committee inside each account
Which roles are involved in a decision of this kind, what each is measured on, what would make each of them say no, and who is currently reachable. This is the step that prevents a deal stalling on an objection from somebody nobody ever spoke to.
You get: Committee maps with role-level messaging per tier
Agree the rules of engagement in writing
Who contacts whom, at what point marketing hands over, what triggers the hand-back, and how a disagreement about an account gets settled. Unglamorous, and it prevents more waste than any creative decision on the programme.
You get: A written service agreement between sales and marketing
Run, measure by account, and prune quarterly
Programmes go live by tier, reporting runs on account engagement and pipeline, and every quarter accounts come off the list as well as going on. The pruning is the discipline that keeps the cost per active account inside the range the arithmetic allowed for.
You get: Account-level reporting and a standing quarterly review
Questions
What B2B teams ask about running ABM
How do we know whether ABM is worth it for us?
Work out roughly what it costs to run a programme against one account for a year, then compare that with the gross margin on one won deal and the realistic probability of winning it. If the programme cost is a small share of a single win, the method makes sense. If it approaches a meaningful share, a broader demand programme will serve you better.
Two other conditions matter as much as the arithmetic. There has to be a defined market of companies you can name, and there has to be a sales team with the time and the appetite to work them. ABM without sales participation is expensive advertising to a narrow audience.
Who chooses the accounts, sales or marketing?
Both, from the same evidence, in the same room. A list drawn up by sales alone tends to reflect existing relationships and recent conversations. A list built by marketing alone tends to reflect firmographic tidiness and misses the accounts where a real opening exists.
What we bring to that room is the analysis: which characteristics your won deals share, which your lost deals share, where deals stall, and how large the addressable set genuinely is. The decision is then made jointly and written down, which matters later when somebody wants to add an account mid-quarter.
What does the buying committee actually mean in practice?
It means that in most considered B2B purchases, several people have to agree, and they are not measured on the same things. The person who will use the product cares about whether it makes their week easier. The person who owns the budget cares about what it displaces. Security, legal or procurement care about whether it creates a problem they will be blamed for.
A single message aimed at the champion leaves the other roles to be persuaded second-hand by somebody who does not speak their language. Committee mapping is simply deciding, per role, what that person needs to see, and making sure it exists before the deal reaches them.
How do we measure this if we are not counting leads?
By account rather than by person. The measures that matter are how many roles inside a target account have engaged, whether engagement is deepening or fading, whether a first meeting happened, whether an opportunity opened, and what happened to it. Those move slowly and they are the honest picture.
The change has to be made before the programme starts. If lead count remains on the dashboard, it will be the number people react to, and the first quarter of an account programme always looks poor by that measure.
How many accounts should be on the list?
Fewer than instinct suggests, and the constraint is usually sales capacity rather than marketing capacity. The list has to be small enough that each account genuinely receives attention, because an account that is nominally targeted and practically ignored costs money and returns nothing.
We usually work in tiers instead of one flat number: a small group receiving genuinely bespoke work, a larger group receiving clustered programmes built around a shared situation, and a wider set receiving targeted advertising and content with light personalisation.
Work out whether the arithmetic supports an account programme
Bring your contract values, your win rates and a sense of how many companies you could realistically sell to. We will tell you whether ABM makes commercial sense for you, and what we would recommend instead if it does not.
If we don't deliver the work we agreed to deliver for reasons within our control, you don't pay for the undelivered work. Read our guarantee
Last updated · Reviewed by Zubair Afzal