Five ABM Intent Data Mistakes That Waste a Good Account List

Most account-based marketing programs do not fail because the target list was wrong. They fail because the list stops changing the moment it is built. Intent data is supposed to fix that, and in most programs it does something much smaller: it sorts the accounts you already picked into a new order.

Sorting is an improvement over alphabetical. It is not what the data can do, and it is not where the pipeline is. The five mistakes below show up most often in programs that have intent data wired in and still cannot point to a single meeting that came from it.

What Is ABM Intent Data?

ABM intent data is evidence that a target account's behavior has changed in a way that suggests it may be entering a buying window. In an account-based program it should inform three things: which accounts get attention, which play runs, and which members of the buying group are worth engaging.

That last part holds most of the value, and it is the part most programs skip. A score tells you an account got warmer. It does not tell you who to talk to or what to say, and those two decisions determine whether the play works.

It is worth separating from two neighbors. Firmographic fit says an account could buy. The distinction between first-party and third-party intent data describes where the evidence came from rather than what it proves. If the category itself is new, the four types of intent data and their limits cover the ground first.

Mistake 1: Reading an Intent Score as an Event

An intent score is a deviation from an account's own baseline, not a purchase decision. Topic-level providers build it that way deliberately. Bombora describes an Intent score near 50 as normal activity for that account, and a Topic Surge Score of 60 or higher as a sudden spike worth acting on.

A spike means research intensity rose above what is usual for that company. It does not mean budget exists, a project has been approved, or anyone has been assigned to the problem. Teams that read the score as an event write messages that assume a decision is already underway, and those messages land badly when the account was simply reading more than usual.

The fix is to pair the score with a signal that carries a date. A job posting, a funding round, a leadership change, or a procurement notice each has a timestamp and an implied owner. The score tells you which account deserves a look. The dated event tells you what to reference and when that reference expires.

Mistake 2: Letting Intent Only Re-Rank the List You Already Had

If a signal can only reorder your existing target list, it can never surface the account you failed to qualify. This is the most common wiring in ABM. The list gets built from firmographics, an intent score is applied as a sort key, and the set of accounts never changes.

The best conversation available in a given quarter is often at a company that did not match the filter last quarter. A business that just hired its first revenue operations lead, or opened a second market, can look wrong on paper and right in practice.

The fix is to let signals nominate accounts, with a cap. Reserve a defined share of the program, a tier of its own, for accounts that entered on signal alone and were never on the list. After a quarter, compare what that tier produced against the main list. If it produced nothing, the signals are wrong. If it produced meetings, the firmographic filter is too tight.

Mistake 3: Personalizing to One Champion Instead of the Buying Group

This mistake has the clearest evidence against it, and most ABM playbooks still make it. Gartner surveyed 632 B2B buyers between August and September 2024 and found that tailoring content to the buying group raised consensus by 20 percent while content focused on individuals reduced it by 59 percent. The same research found that 74 percent of buyer teams show unhealthy conflict during the decision process, and that groups reaching consensus are 2.5 times more likely to describe the deal as high quality.

Read together, the implication is uncomfortable. A play built around one enthusiastic champion's individual hook does not merely underperform a group-level play. It can reduce the internal agreement you need in order to win.

So the signal's job is not to find one person to impress. It is to show which roles the business priority behind the signal actually touches. A hiring cluster in lifecycle marketing and revenue operations touches whoever owns retention, whoever owns the data, and whoever will be asked to approve the spend. Each of them needs the same underlying priority framed in their own terms, which is a different exercise from writing three variations of one hook. Letting the signal pick the contacts covers that selection step in detail.

Mistake 4: Letting the Signal Expire Inside Your Own Routing

Most intent signals lose their value faster than a standard ABM program can respond. The sequence is familiar. A signal fires, an overnight job scores it, the account enters a nurture track, a weekly review flags it, and a seller reaches out eleven days later referencing something the company has already moved past.

No single step in that chain is broken. The problem is that total latency exceeds the useful life of the signal. A late message referencing a stale event is worse than a message referencing nothing, because it advertises that the research was automated and unattended.

The fix is a launch window per signal type rather than one program cadence. Funding rounds and leadership changes stay useful for weeks. A review complaint or a single job posting is close to spent within days. How long each signal stays worth acting on sets that schedule, and auditing the steps between a signal firing and a reply landing is how you find where your own latency accumulates.

Mistake 5: Assuming the Signal Tells You the Channel

A signal indicates timing and priority. It does not indicate whether the next move should be a seller's message or a self-serve asset, and ABM programs guess wrong in both directions.

Gartner research published in May 2026, based on 645 B2B buyers surveyed between August and September 2025, found that 69 percent of buyers turn to sales reps to validate AI-generated insights even though 67 percent say they would prefer a rep-free experience. Buyers used an average of seven information sources, and reps remained the most important source at the moments when buyers defined the problem, selected a supplier, and built internal support.

The practical reading is that early research activity is usually a content moment, while a signal implying a decision is being formed is a seller moment. A topic surge with no dated event behind it rarely justifies a rep reaching out. A new executive with a mandate, an active procurement notice, or a specific competitor complaint usually does.

What to Wire Instead

Before anyone writes a message, a signal should move through four decisions.

• Does this account belong in the program at all, including accounts the firmographic filter missed?
• What does the signal prove: a problem, an owner, or a deadline?
• Which roles does the business priority behind it touch, and what does each of them care about?
• Is this a content moment or a seller moment, and how long is the launch window before the signal goes stale?

Sorting signals by what each one proves makes the second decision fast, and it is the one most teams skip. If a signal does not change at least one of those four answers, it is being recorded rather than used. That is the difference between a program that reports intent coverage and one that reports meetings. The same logic applied to outbound rather than ABM is covered in signal-based selling as an operating model.

Where Alsona Fits

Alsona handles the execution half of that sequence rather than the advertising half. A team could monitor job boards, funding announcements, review sites, technology changes, and executive posts by hand, then research each account, decide which roles the signal touches, and write per-role messaging for LinkedIn and email.

Alsona is built to do that continuously. It monitors 32 intent signals across hiring, advertising, technology, funding, reviews, social activity, and company changes, scores accounts on those unstructured signals alongside ICP fit rather than firmographics alone, researches the account, and turns that context into individualized LinkedIn and email messaging for each contact. Follow-ups run automatically, and replies land in one shared inbox so a signal-triggered conversation does not get lost between channels.

It is not an ad platform and will not run a display budget. It covers the part of an ABM program where a detected signal has to become a relevant conversation with several people at the same account.

The Takeaway

Intent data earns its place in an ABM program when it changes who is on the list, who gets contacted, what each person hears, and how fast the play goes out. Most deployments change only the sort order, which is why they are hard to defend at renewal.

Pick one of the five mistakes and test for it this quarter. The re-ranking problem is the easiest to prove. Look at every meeting the program produced last quarter and count how many came from an account that was not on the original list. If the answer is zero, the signals are decorating a static list.

Build outbound around evidence you can point to. See what a signal-based account briefing looks like before you commit a play to it.