What the State of ABM report covers

Account-based marketing in 2026 does not look like the ABM of even two years ago. The tooling got cheaper, the AI got real, and the old lead-first playbook quietly stopped working. This report is our field view of what actually changed, what is vendor theatre, and what a program that wins the accounts that matter looks like now.

It is written for revenue leaders who are past the "should we do ABM" question and are trying to build something that compounds, not another dashboard nobody actions.

The lead is dead. Long live the account.

The central shift is a change in the unit of measurement. Counting leads rewards volume and hides whether the right companies are moving. Counting account progression rewards the thing that actually pays: the named accounts you decided were worth pursuing, moving stage by stage. Everything downstream, scoring, routing, reporting, follows from that one decision. We walk through making that switch in the 90-day ABM roadmap, and it starts with an ICP you can turn into a ranked account list rather than aspirational prose.

The three engines of AI-era ABM

The report breaks the modern motion into three engines that have to work together:

  • Targeting. Who to contact, and when. A scoring model trained on your closed-won data, refreshed continuously, fed by intent and trigger signals. This is where most programs are weakest and where the payoff is largest.
  • Reach. Coordinated plays across email, LinkedIn and ads, fired by the signal rather than by a campaign calendar. Getting this right is a question of how you set the ABM strategy up before launch, not which tool you buy.
  • Measurement. Account-level attribution that tells you which accounts advanced and why, so the model gets sharper every quarter.

RFME: the scoring shift that changes the ABM game

Classic customer scoring borrows RFM from retail: Recency, Frequency and Monetary value. It is good at one question, which customers are worth keeping. ABM asks a different one: which accounts are worth pursuing, right now. So we extend RFM to RFME, adding a fourth axis:

  • Recency. How recently the account showed a real buying signal.
  • Frequency. How often those signals repeat, because one visit is noise and a pattern is intent.
  • Monetary. What the account is worth if it closes: deal size and expansion potential, so effort follows value.
  • Engagement. The intent and engagement signals, site visits, content, webinars, replies, that say the buying committee is actually paying attention.

RFM tells you who already bought. RFME tells you who is about to. That fourth letter is what turns a static account list into a live, ranked queue your team works in priority order, and it is the scoring model underneath the targeting engine above.

Intent, honestly

Intent data is the most oversold layer in the category. It is a genuinely useful signal and a genuinely bad master. The report is direct about what it can and cannot tell you, and about the gap between what a platform's predictive model does and what you can replicate yourself. If you are weighing whether to buy that layer at all, we costed it out separately in do you need ABM software.

Who should read it

If you run growth, demand gen or revenue at a B2B company and ABM is either on the roadmap or already stalled, this is the honest brief we wish existed when we started. It pairs with our account-based marketing service, where we build the engine described here inside your own stack, so you own every part of it.