ABM targeting: a contact-level point of view

Nearly 80% of B2B organizations are actively running an ABM program. Most of them target accounts instead of buyers. The gap between those two things is where budget disappears. The account looks right, the spend goes out, and no individual buyer ever sees a relevant message. Adoption did not solve the precision problem. It made that problem larger.

TL;DR

• ABM targeting fails when you activate the campaign, not when you select the accounts.
• Targeting an account is a probability bet; targeting a specific contact is a decision.
• Match rate (not CPM or CTR) is the metric that shows whether your spend actually lands.
• Buying committees have 13+ internal stakeholders; account-level targeting cannot address them by role.
• Only 21% of marketers use intent data, making signal-driven contact targeting a structural advantage.

The targeting failure mode hiding in every ABM program

A working ABM program tells you Acme Corp is showing intent. It does not tell you which of Acme's 800 employees is researching your category. Or whether your ads are reaching the economic buyer at all.

The gap is structural. When you target at the account level, you produce account-level signal: impressions, clicks, and engagement attributed to a domain, not a person. IDC's research on ABM advertising shows this directly: account-level approaches overlook individual personas within buying groups, diluting both relevance and precision.

Tiering accounts, refining your ICP, and stacking intent scores improve which accounts are on the list. They don't answer whether you're reaching the right person inside them.

Account-level targeting is a probability statement. Contact-level is a person.

Targeting Acme Corp is a bet. Your message needs to find the relevant buyer inside a company of hundreds or thousands of employees, via a platform matching by domain or IP. That bet gets worse as the account gets larger.

Targeting the VP of Engineering at Acme Corp who has visited your pricing page three times this week is a different act. You're addressing a person with demonstrated behavior.

The case for account-level awareness spend is real. Early in a buying cycle, you don't always know which contacts will become active, so building organizational familiarity has value. But that argument breaks when you examine where the impressions land. Brand-building spend aimed at "Acme Corp" reaches the right functions only by chance. If your message hits IT procurement while your economic buyer sits in the CFO's office, you're generating impressions on a domain — paying for logos, not people.

ABM is moving from orchestration toward discovery, according to IDC. Orchestration manages outreach to known accounts. Discovery uncovers buying groups before they self-identify. That shift only happens at the contact level. You cannot discover an unknown buyer by targeting their employer.

What it actually takes to target ABM buyers, not ABM accounts

Start with de-anonymization: knowing who is actually engaging, not just which company's IP hit your site. An account visiting your pricing page is a signal. A named VP of Operations visiting it twice in one week is a targeting chance.

From there, build a contact profile from behavioral signals, not just firmographic fit. Title and company size tell you who might buy. On-site behavior, topics researched off-site, and competitive intent show who is in an active buying motion right now.

Finally, audiences must refresh in real time. A CSV export uploaded to LinkedIn on the first of the month is not ABM targeting. It is a list of people who were relevant when you exported it.

Airbyte put this into practice by building contact-level audiences on two signal types: off-site research intent (contacts consuming data integration content before reaching Airbyte's site) and competitive intent (contacts searching competitor brand terms). The results from the Airbyte ABM playbook: LinkedIn CPC between $0.50 and $2.00 and measurable growth in enterprise inbound pipeline. Same channels, sharper signals.

The signals that should define who you target

Only 21% of marketers currently use intent data in their ABM targeting, a structural gap most competitors haven't closed. The 2026 ABM Benchmark Survey puts AI's average effectiveness score for ABM at 7.3 out of 10 for account selection and personalization. That number only holds if the base signals are right.

Four signal categories should define your contact targeting:

• On-site behavior records page visits, pricing views, and repeat sessions by a named contact at a target account. Returns to pricing or technical documentation signal a later-stage buying motion.
• Off-site research intent tracks the topic clusters a contact consumes before they reach your site. A contact reading about your category on third-party publications is already in a buying motion. You can reach them before competitors do.
• ICP fit at the contact level uses title, function, and seniority within a target account. Firmographic fit gets you to the right company; contact ICP fit gets you to the right person inside it.
• Custom intent topics cover search behavior and content consumption tied to your category, competitors, or the specific business problem your product solves. They narrow the audience to buyers already inside the decision, not just browsing the category.

Get all four right and your audience is a person with a demonstrated need, a defined role, and active research behavior. Not a domain on a list.

Why match rate is the targeting metric most reviews miss

Build the best contact list you can. Now ask: what percentage of those contacts can your ad platform actually resolve to a targetable identity? That percentage is the match rate, and for most B2B lists it is low. A list of 5,000 contacts with a 15% match rate means 4,250 of those buyers never see your ad. Your impression count and CPM look fine. That gap rarely appears in standard performance dashboards — which is exactly where pipeline anxiety comes from: the spend clears, the pipeline doesn't, and nothing in the report says why.

Most ABM program reviews track impressions, click-through rate, and pipeline influence. None of those confirm whether the spend reached the intended contacts. Match rate is the prior metric: reach must be confirmed before efficiency questions matter.

Workbar hit this same wall. Their campaigns on Google, Meta, and Reddit produced clicks with no way to know which real contacts were behind them. By de-anonymizing ad clickers rather than only site visitors, they identified verified contacts and integrated them into HubSpot as dynamic lists. Previously paused channels came back online. The channels hadn't changed. Who they were reaching had.

The buying committee problem, and why contact-level is how you solve it

The scale of the problem

A typical enterprise B2B deal in 2026 involves 13 internal stakeholders plus 9 external influencers. Account-level targeting treats all 22 of them as one unit labeled with a domain name.

The economic buyer, the technical evaluator, the champion, and the likely blocker all have different questions and different content needs. A message calibrated for a CFO won't move a Head of Engineering. Targeting the account produces one message aimed at everyone, which lands with no one in particular.

Why the ROI case is strongest here

99% of B2B organizations with a dedicated ABM team report higher ROI than traditional marketing programs, according to Forrester. That reflects what happens when you concentrate spend on accounts that fit. The programs with the highest returns go further: they address buying groups by role, with sequences that map to each stakeholder's position in the decision.

Buying groups are dynamic and often represent only a fraction of a large account (IDC). Treating the account as the targeting unit means you have no way to map who holds each role in the decision.

The sequence that account-level cannot produce

The champion needs proof-of-concept material. The economic buyer needs a business case. The technical evaluator needs integration documentation. You can't deliver any of that if your targeting unit is the company. Identifying individual contacts is the only way to reach each person with the right message.

Activating contact-level data across the funnel

Closing the identity gap with Reveal

Vector Reveal de-anonymizes who is engaging, down to the specific person, not just the company. That includes ad clickers who never converted. A contact can click your LinkedIn ad, skip the form, and vanish into your analytics as a ghost — an anonymous session with no name attached. Ad Reveal surfaces their name, title, and company.

Your target list is built from observed buying behavior. You can see which contacts at Acme Corp have already demonstrated interest.

Closing the activation gap with Target

With Vector Target, you turn those identified contacts into audiences that push directly to LinkedIn, Google, and Meta, refreshing in real time as signals change. When a new contact at a target account starts consuming competitor research, they enter the audience. When they go quiet, they cycle out.

Airbyte's contact-based audiences drove LinkedIn CPC to $0.50-$2.00 by reaching people in active buying motions rather than domain-matched profiles. Workbar turned anonymous ad clickers into identified contacts in HubSpot and reactivated previously paused channels on the strength of that identity data.

Read more about building a contact-based marketing strategy for the full workflow.

The step ABM was always missing

Account selection is sound practice. The precision problem was never which accounts to pick. It was whether you ever reached the people inside them.

If your program can tell you which account is active but not which person inside it is researching you right now, you are working from a map with no address. Contact-level targeting is not a refinement of ABM. It is the step ABM skipped.