ABM campaigns: a contact-level point of view

99 percent of B2B organizations with dedicated ABM teams report higher ROI than traditional marketing, according to Forrester. Yet the buyers those programs are built to win keep getting harder to pin down: a single B2B purchase now pulls in 13 internal stakeholders and nine external influencers, per Forrester, and that number climbs on complex deals. That gap lives in one place: the audience layer. Teams are targeting company logos. The 3-5 people inside each logo who are actually reading, clicking, and forming a buying opinion remain invisible. Static account lists cause this failure, regardless of the strategy, creative, or channel mix.

TL;DR

• Most ABM teams track email opens while pipeline attribution goes unmeasured.
• Buying decisions are made by a 3-5 person committee, and most ABM campaigns can't reach them by name.
• Signal-driven audiences replace stale CSVs with live contact data.
• A single contact-level segment activates across LinkedIn, Google, Meta, and Reddit simultaneously.
• Unknown buyers in your best accounts means your ABM has an audience problem.

Why account-level campaigns leak budget by default

When you target "Acme Corp," you pay for impressions served to every employee who loosely matches a job-title filter. Most aren't on the buying committee. Most never will be. They burn impression budget without the power to sign anything. That's where pipeline anxiety starts: spend goes out, no names come back.

Most teams still lean on activity metrics — email opens, content downloads, page views — that say little about pipeline. Forrester has spent years pushing B2B marketers to ditch the sourcing and activity metrics that reward easy-to-collect data over revenue, noting that buying has changed but measurement hasn't kept up. Teams measure what's easy to collect. The easy metrics don't connect to pipeline, so the feedback loop reinforces waste rather than correcting it.

Why firmographic filters don't fix it

The budget leakage isn't a targeting mistake you can fix with a tighter firmographic filter. Firmographics describe companies, not the individuals inside them who hold budget, sign contracts, and run evaluations.

The unit of an ABM campaign isn't an account. It's a buyer.

Buying decisions don't happen at the account level. They happen inside a committee of specific individuals who evaluate your category, compare options, and eventually recommend a vendor. ABM has evolved into a media-first strategy because those individuals need to be reached before they self-identify, according to IDC. Waiting for a form fill means you've missed most of the buying process.

The ABM leaders who can't drive engagement from key accounts aren't struggling with creative or messaging. The audience is undefined. Targeting a logo doesn't tell an ad platform who to reach. The platform applies its own probabilistic matching against a company domain and surfaces whoever happens to be active. The result is a random group of employees who have no part in the evaluation. You're paying to reach logos when the buyers are people.

Account-level targeting does have one defensible use case: a high-frequency awareness push where reach across a logo matters more than precision. For a 30-day awareness campaign, reaching broadly across an organization is intentional. That logic breaks over a 6-month sales cycle. Every week of account-level spend burns budget on HR directors and junior analysts who will never touch your evaluation. The committee is 3-5 people. Everyone else is noise.

Build campaign audiences from signals, not CSVs

The typical ABM workflow: export a target account list from your CRM, upload a CSV to LinkedIn Ads, wait for match rates to resolve. The problem is that the CSV is already stale the moment it uploads. People change roles. New stakeholders join an account mid-cycle. A VP who just joined starts researching your category and never appears in your segment because they weren't in the export.

Signal-driven audiences work differently. They build from behavior: who visited your pricing page this week, which contacts opened every sequence email and went quiet. The audience reflects what's happening now, not a CRM snapshot from three weeks ago.

Why identifying the right buyers is the hard part

Identifying the specific people inside an account and reaching them at the right moment is where most programs stall. Gartner describes the modern buying group as numerous members who each run their own research and bring different goals, looping through six distinct "buying jobs" rather than moving in a straight line. These aren't strategy problems. They're data access problems. You can't identify the right buyers if your audience input is a company name and a loose title filter.

What Airbyte found

Airbyte, a data movement platform, shifted from account-level uploads to contact-based audiences built from both on-site visitor behavior and off-site research signals. The result: LinkedIn CPCs between $0.50 and $2.00 and a measurable increase in enterprise inbound pipeline from paid media. Airbyte's Tanmay Sarkar described the shift plainly: "Vector gives us onsite and offsite contact-based audiences we can target across every channel." The multi-channel ABM playbook details how they built the system. The audience wasn't a CSV export. It was a live feed of identified buyers.

Run the same audience across LinkedIn, Google, Meta, and Reddit

Channel coverage is a different question than channel budget. The question isn't whether the marketing team can afford to be on Reddit. The question is whether the buying committee uses Reddit. If they do, the same segment should be there.

Reaching fragmented B2B buying groups now requires going beyond LinkedIn and Meta, according to IDC. Programmatic display, connected TV, and niche platforms like Reddit are where a committee's attention lives outside working hours. A single buyer might check LinkedIn at 9 a.m., search on Google at 2 p.m., and scroll Reddit at 8 p.m.

Account-level audiences can't follow that buyer across channels without rebuilding the segment for each platform. Contact-level audiences can. The same named segment activates on LinkedIn, Google, Meta, Reddit, TikTok, and X from a single definition via Vector's Target tool. The buyer sees consistent messaging regardless of where they are. Your campaign doesn't fragment into disconnected impressions that never accumulate into a coherent point of view.

Keep audiences live as buyers move (and committees shift)

A contact who visited your pricing page three months ago and went dark is not the same signal as one who returned last Tuesday. A static segment treats them identically because it was built once and never updated. You're running ads at a buyer who may have already purchased a competitor, changed roles, or simply moved on.

Live audiences update as behavior changes. Contacts who enter the buying window get added. Contacts who go cold get removed. A new VP who joins a target account and reads your blog appears in the segment within hours, not at the next quarterly review.

Workbar, a coworking brand, built signal-driven segments integrated with HubSpot to track demand for new locations before those locations opened. When specific contacts hit threshold intent behavior, sales got automated Slack alerts, not a monthly spreadsheet refresh. The Workbar case study describes this as "reactivating paused channels to target only the real people who matter." Live audiences reflect where buyers are today, not where your CRM was last month.

Closing the loop with identification and activation

Vector's Reveal de-anonymizes 15 to 30 percent of US site traffic into identified contacts, per an independent assessment by Docket.io. Anonymous page views on your pricing page and competitor comparisons become a qualified list of active researchers.

From identified contacts to activated audiences

Those contacts flow directly into Target, which syncs them to ad platforms across LinkedIn, Google, Meta, Reddit, and more. Vector's LinkedIn match rates run well above the roughly 13 percent average for account-level uploads. A higher match rate means more of the contacts you've identified actually receive the impression. Account-level uploads lose the majority of the segment to failed matches before a single ad is served.

The closed-loop result

Tanmay Sarkar, Airbyte's head of growth, put it plainly: Vector provided "onsite and offsite contact-based audiences we can target across every channel." The audience wasn't assembled from a static account list. It was built from behavioral signals, resolved to identified buyers, and activated simultaneously across channels. No manual CSV touched the workflow. The same approach applies to measuring ABM intent data ROI: once the audience is live, attribution gets sharper too.

Solving the ABM audience problem

The budget, the creative, and the channel mix all rest on one assumption: that the audience is real. For most ABM programs, it isn't. A company domain and a job-title filter produce a guess — an audience of ghosts. That guess costs real money to run.

The teams that report high ROI from ABM know who is in the buying committee. They can name the specific people. The creative, the channel, the cadence: all of it follows from that.

If you can't name the 3-5 people inside your best target accounts who are actively evaluating your category right now, the rest of the campaign is built on speculation. Fixing the creative won't change that outcome. Fixing the audience will.