Named Accounts as the Unit of Marketing
Account-based marketing treats a defined list of named companies as the market, rather than treating individuals as leads to be collected. The account is the unit of targeting, of measurement and of the plan; it may generate no leads for a year and still be progressing.
The term was introduced and named by ITSMA, the Information Technology Services Marketing Association, in 2003. Bev Burgess, then head of ITSMA's European division, is the person most commonly credited with coining it. The software vendors that popularised the category from around 2014 did not invent the term; treating a few large accounts as markets predates the label by decades.
Two confusions to clear early. Abm marketing is not a synonym for firmographically targeted advertising, although that is what most programmes sold under the name consist of. Nor is it the opposite of demand generation: the two coexist, with account-based effort where deal value justifies the labour.
The Three Tiers, and Which One You Are Actually Running
ITSMA documented three types of ABM in a 2016 study, and its taxonomy remains canonical:
| ITSMA name | Common name | What it involves |
|---|---|---|
| Strategic ABM | One-to-one | Bespoke programmes for individually named accounts: custom research, custom content, account plans, executive sponsorship. Single-digit to low-double-digit account counts per marketer. |
| ABM Lite | One-to-few | Clusters of accounts sharing an industry, use case or trigger. Content is tailored to the cluster, not the company. Typically tens of accounts per cluster. |
| Programmatic ABM | One-to-many | Technology-driven personalisation across hundreds or thousands of accounts. In practice, account-list-targeted advertising plus website personalisation. |
Stated plainly: most programmes marketed as b2b abm are programmatic. One-to-one ABM is labour-intensive and does not scale, which is why few organisations run much of it, and why the tier distinction matters when someone proposes a budget. A one-to-many programme is a media buy with better targeting and should be judged as one; name the tier before committing budget, because the three cannot be measured alike.
Why the Account Frame Fits Business Buying
Buying is done by a group. Gartner's survey of 632 B2B buyers, fielded in August and September 2024, described teams ranging from five to sixteen people across as many as four functions, found that around 74% of buying groups exhibited unhealthy conflict during the decision, and reported that groups reaching consensus were 2.5 times more likely to call the deal high quality.
If the deciding unit is a group, counting individual leads inside a company tells you little: three downloads from three departments may be one project or three idle readers.
One finding deserves attention from anyone buying personalisation software: Gartner reported that group-level relevance improved consensus by 20%, while individual-level relevance reduced it by 59%. That argues for content addressed to the buying group's shared problem rather than each individual's inferred interests. High deal values and small addressable markets complete the case: where a hundred companies hold most of the available revenue, spending evenly across a hundred thousand is irrational.
Building a Target Account List That Survives Contact With Sales
A target account list is a finite, named list of companies marketing and sales have jointly agreed to pursue. In practice it is keyed on domains, because that is what ad platforms match against; company names break matching immediately, and subsidiaries, acquired brands and country entities break it soon after.
- Define the ideal customer profile from firmographics and technographics, then validate it against your closed-won, high-retention customers rather than against ambition.
- Size the addressable set from a business database such as ZoomInfo, Apollo or Cognism.
- Score and tier: Tier 1 to one-to-one, Tier 2 to one-to-few, Tier 3 to one-to-many, layering in intent signals.
- Get written agreement from each rep who owns the accounts. A list sales did not agree to is the commonest reason ABM produces no pipeline.
- Fix the match keys, cap the list at what sales can work, and review it quarterly.
One mechanical warning. On LinkedIn, company-list targeting resolves your upload against LinkedIn's member database, and the enforced minimum applies to matched members, not rows in your file; location is a required facet and reduces the count further. LinkedIn recommends audiences above 50,000 for Sponsored Content, which a 200-company list will never reach.
What ABM Platforms Genuinely Do
The category has real capabilities and a layer of marketing on top. What is verifiable:
- Account identity resolution — mapping web traffic, ad impressions and CRM records to a company entity, usually through reverse-IP lookup or pixel identity graphs.
- Firmographic and technographic enrichment.
- Account-list media buying across programmatic display, LinkedIn and sometimes connected TV.
- Website personalisation by matched account or segment.
- Intent data: first-party from your own properties, third-party from publisher co-ops such as Bombora.
- Account engagement dashboards and journey reporting, with write-back to the CRM.
Treat four claims sceptically. That the platform identifies the specific buying committee members researching you: third-party intent data is generally account-level, aggregated and modelled. That artificial intelligence predicts which accounts will buy: predictive scoring is real, but its accuracy is vendor-specific and rarely validated against a published holdout. Any headline uplift figure, which will be a self-reported customer survey with no control group. And funnel replacement, which is framing rather than a capability.
The Prerequisite That Cannot Be Bought
ABM fails without sales alignment, and no software substitutes for it. The alignment is specific and testable rather than a matter of goodwill:
- The account list is agreed rep by rep, in writing, with a named owner per account.
- Both functions use the same definition of an engaged account and the same source of truth for it.
- There is an agreed response commitment when an account shows activity, and someone reports compliance against it.
- Tier 1 accounts have a plan marketing and sales wrote together, with sales contributing what it knows about the account's politics.
- Marketing's target is pipeline and progression inside the named accounts, not lead volume from anywhere.
Forrester, which absorbed SiriusDecisions, has published a piece on ABM's myths, among them that sales alone should choose the accounts and that ABM cannot be measured. The test before spending anything: can you produce a list of accounts with a named rep against each, and would that rep recognise it?
Measuring Account Engagement Rather Than Leads
ABM measurement replaces lead counts with account progression, which is more honest and considerably harder. Report coverage — how many target accounts show any contact; depth — how many functions and people inside each are engaged; progression between account stages; opportunities, pipeline and closed revenue from named accounts; and cost per engaged account.
Four cautions are structural. Engagement scores are platform-defined, have no external benchmark, and are produced by the vendor whose spend you are evaluating. Small lists make incrementality testing very difficult: with fifty or two hundred accounts you usually cannot construct a properly powered holdout. Reverse-IP account identification degraded materially with remote and hybrid work, so accounts-on-site figures are noisier than a dashboard suggests. And a list drawn from accounts sales already knows excludes accounts that have not surfaced yet.
There is, as far as can be established, no independent controlled evaluation of ABM against broad-based demand generation. That is an absence of evidence rather than evidence of absence, but anyone presenting ABM as proven is overstating the record.
When ABM Is the Wrong Model
ABM is the wrong model for a company with a low average deal value or a large addressable market. The per-account cost of research, tailored content, coordinated advertising and sales attention is high and largely fixed, so it is only recoverable from deals big enough to absorb it. If your typical contract is small, or tens of thousands of companies could plausibly buy, the money belongs in reach and efficiency.
There is also a theoretical tension. John Dawes of the Ehrenberg-Bass Institute proposed the 95-5 heuristic in May 2021, published with the LinkedIn B2B Institute: because purchase cycles are long, most category buyers are out of market at any given moment. It is derived from interpurchase intervals, and Dawes explicitly called 95% a heuristic rather than a precise rule, with the true share depending on the category's cycle. On that evidence, spending everything on a small list of currently known accounts biases a programme towards short-term activation and away from future demand. Both bodies of work are credible; this is a real tension, not a settled question.
Other disqualifiers: no sales capacity to work named accounts, CRM data too poor to identify accounts, or nobody willing to own the list.
Frequently Asked Questions
How many accounts should be on a target account list?
As many as sales can genuinely work, which is usually far fewer than marketing proposes. For one-to-one programmes the practical ceiling is single figures to low double figures per marketer, because each account needs its own research. One-to-few clusters run to tens of accounts; one-to-many programmes cover hundreds or thousands, but at that point you are running targeted advertising and should judge it as such.
The binding constraint is follow-up capacity, not media budget.
Do we need an ABM platform to run account-based marketing?
No. The minimum viable version is an agreed account list keyed on domains, company-list targeting on LinkedIn or a comparable channel, account-level reporting in the CRM, and a sales team that follows up when a named account shows activity.
Platforms earn their place at scale, where matching traffic to accounts and orchestrating media across hundreds of accounts becomes impractical by hand. Buy one after the account list and the sales agreement exist, never before.
Is ABM worth it for a company with a low average deal value?
Generally not. The cost per account of custom research, tailored content and coordinated sales attention is high and does not fall much with deal size, so a small contract value cannot recover it. Payback is also long, which compounds the problem for a business that needs revenue this quarter.
The same applies where the addressable market is very large: narrowing spend to a few hundred companies forfeits reach you need.
How do we measure ABM if we are not counting leads?
Measure coverage, depth, progression and outcome within the named list: the share of target accounts showing engagement, how many people and functions inside each are engaged, movement between account stages, and the pipeline and revenue those accounts produce.
Two warnings. Vendor engagement scores are platform-defined with no external benchmark, so use them as internal trend lines. And with fifty or two hundred accounts you generally cannot run a powered holdout, so treat any lift claim sceptically.