ABM Intent Data: How to Tier, Trigger and Convert Accounts

ABM intent data shows which accounts on your target list are researching your category right now. It scores that research so you can tell a warm account from a quiet one.

The trouble starts at the budget meeting. Your target list is already agreed with sales, so nothing new gets discovered. You end up with a dashboard of surging accounts and the same flat budget spread across all of them. Reps work a spreadsheet of company names with no context attached. The signal fires. Nothing changes.

In this guide I'll show you how ABM intent data works as an allocation tool rather than a discovery tool. How to tier accounts and move them as signals shift. How to map a company-level signal to real people. How to hand it to sales, and how to diagnose a program that looks busy but builds no pipeline.

What ABM intent data is

ABM intent data is buying research activity, measured at the account level, applied to a defined list of target accounts.

The mechanics of how signals get collected and scored work the same way they do anywhere else. How B2B intent data works doesn't change because you point it at an account list.

What changes is the population. Broad intent programs watch the open market and ask who's out there. ABM programs watch a bounded list, often 50 to 500 companies, agreed between marketing and sales before the quarter starts.

So the question shifts. Not "who is in market?" but "which of my accounts moved, and what do I do about it?"

Plenty of that movement happens in the B2B dark funnel, where nobody fills in a form.

Why ABM changes the job of intent data

Demand generation uses intent to widen the funnel. More accounts, more reach, more names. Account based marketing works the opposite way.

ABM can't do that. The list doesn't grow. That's by design, and the cost per account runs high. A single one-to-one play can absorb custom research, bespoke content, executive time and a dedicated marketer working alongside an account executive. That cost climbs further in enterprise account based marketing. That doesn't scale past a couple of dozen accounts.

Here's the math that decides everything.

You have 200 target accounts. You have budget and people for 20 deep plays this quarter. Which 20?

Firmographics can't answer that, because every account on the list already passed ICP screening. Revenue potential can't answer it alone, because the biggest logo isn't always the one paying attention. Intent answers it, because it's the only input that tells you which accounts changed behaviour recently.

That's the whole value. Intent ranks a list you already committed to.

How to tier target accounts with intent data

Three tiers is the working model. The tier decides how much money and human attention each account gets.

Tier 1: one-to-one

Five to 25 accounts. Custom research, tailored content, executive involvement, a marketer paired with the account executive.

Entry rules:

• High ICP fit, no exceptions.
• Sustained signal, meaning spikes in at least two of the last four weeks.
• More than one persona researching inside the account.
• Deal size large enough to justify the cost of the play.

The sustained-signal rule matters more than the score itself. One person reading three articles on a Tuesday creates a spike. It doesn't create a buying cycle. A documented intent data scoring model keeps that judgement consistent between reps.

Tier 2: one-to-few

Roughly 25 to 100 accounts, clustered by industry, problem or buying scenario. Semi-custom messaging built on reusable assets, so one campaign serves eight to twelve accounts at once.

Entry rules:

• Good ICP fit.
• Moderate or newly emerging signal.
• Enough shared context with other accounts to justify a cluster.

Tier 2 is where accounts wait. Some climb into Tier 1. Others sit here for two quarters and convert anyway. Each cluster runs as its own ABM campaign with shared creative.

Tier 3: one-to-many

Everything else on the list. Programmatic ads, automated nurture, light or no SDR involvement.

Entry rules:

• ICP fit.
• Low signal, no signal, or signal too new to act on.

Tier 3's job is monitoring and brand presence. You're keeping accounts warm and waiting for a signal worth promoting.

What to do: set a hard cap on Tier 1 based on how many accounts your team can serve properly. If 40 accounts qualify and you can serve 20, take the top 20 and leave the rest in Tier 2. Half-serving 40 accounts produces worse results than fully serving 20.

Moving accounts between tiers as signals change

Static tiers waste the data. An account sitting in Tier 3 in January might reach active evaluation by March. Nobody notices, because someone set the tier once and moved on.

Promote an account when:

1. Signal sustains across two or more weeks instead of spiking once.
2. A second or third persona starts researching the same topic.
3. First-party activity appears, like a pricing page visit or a demo form.
4. The topic shifts from broad category terms to comparison or pricing terms.
5. Champion tracking shows a known advocate moving into the account.

Demote an account when:

1. Signal decays with no response after 30 to 60 days.
2. A full outreach cycle produces no buying group engagement.
3. An opportunity closes lost and the account enters a cooling period.
4. Firmographics change and the account no longer fits your ICP.

Review cadence: monthly. Weekly review creates thrash, with accounts bouncing between tiers faster than any campaign can run. Quarterly review is too slow, since a buying cycle can open and close inside 90 days.

One caution from the demotion side. Losing a deal doesn't mean losing the account. Set a cooling period of two quarters, then let the account re-enter Tier 2 if signals return.

When intent should change your target account list

Tiering reorders the list. Sometimes the list itself is wrong.

Adding accounts. A company outside your list spikes repeatedly on your narrow topics and matches your ideal customer profile on paper. That's a candidate, not an addition. Take it to sales, get agreement, then add it. Accounts added unilaterally by marketing get ignored by reps, every time.

Removing accounts. An account sitting on the list for four quarters with no signal, no engagement and no sales conversation is holding a slot. Cut it. Your total addressable market has better candidates waiting.

Keep the list stable enough to measure. Swapping 40% of your target accounts every quarter makes it impossible to tell whether the program works or the list just changed. I'd cap quarterly churn at around 10 to 15% of the list.

Run this review with sales in the room. A target account list that marketing edits alone stops being a shared commitment.

From account signal to buying group

An account-level signal tells you a company is researching. It doesn't tell you who.

This is where ABM programs quietly turn into ordinary demand generation. The account gets flagged, a sequence fires at one contact, and the play gets called account-based because the list was account-based. One thread into an account is not ABM.

Map the topic to the persona

The topic usually points at the role behind it.

• Security, risk and compliance topics point to security and risk leadership.
• Pricing, contracts and procurement topics point to finance.
• Implementation, integration and migration topics point to technical operations.
• Category and definitional topics point to a researcher building a business case.

That mapping gives your first message context. A note referencing the exact problem an account is working on lands differently from a generic introduction.

Set a coverage gate

Before an account enters a Tier 1 play, check whether you can reach the buying group at all. My rule: three or more reachable contacts across at least two functions.

If you can't clear that bar, the account isn't ready for expensive treatment regardless of how strong the signal looks. You'd be funding a campaign that reaches one person.

Contact data quality decides the outcome here. A perfect signal attached to a bounced email and a departed champion produces nothing at all. B2B data decay breaks the play, not the signal.

The handoff: what marketing passes to sales

The handoff breaks more ABM intent programs than the data ever does.

Marketing flags an account. Someone exports a list of company names. A rep calls without knowing what the account was researching, how strong the signal was, or who inside the company is paying attention. The intent window closes during the discovery questions.

Pass these fields with every routed account:

• The specific topics that triggered the flag.
• Signal strength and how long it has been running.
• The date the signal first appeared.
• Which personas or contacts sit behind the research.
• Any prior touches, campaigns or content the account engaged with.
• A suggested opening line tied to the topic.

Set an SLA by tier:

Account-based routing replaces the usual SQL vs MQL handoff, so agree the definitions before launch. Then track follow-up completion rate, meaning the share of routed accounts that received the agreed action inside the SLA. If that number sits below 80%, stop tuning topics. The handoff is broken, and better data won't fix it.

How long an ABM intent signal stays useful

Signals rot. Treating a 30-day-old spike the same way you treat a fresh one produces awkward outreach about a decision that already happened.

Days 1 to 5. Research intensity peaks. Direct outreach and a named-persona message do their best work here.

Days 6 to 14. The account is narrowing options. Comparison content, proof and specific differentiation carry more weight than an introduction.

Days 15 to 30. A shortlist may already exist. Shift to ads and nurture rather than cold outreach that pretends the research just started.

Beyond 30 days. Treat it as context for the account record, not a trigger for a play. Pair it with other buying signals before acting on it again.

Two operational rules follow from this. Don't re-trigger the same play on the same topic inside 30 days, or you'll batter the same buying group with repeat messaging. And check your alert cadence, because a signal that reaches a rep in a weekly digest has already lost half its value.

Accounts that should never trigger a play

Suppression is dull work. It protects your budget better than any scoring change.

Build an exclusion list covering:

• Existing customers, unless you're running a deliberate expansion play with different messaging.
• Open opportunities in late stage, where a rep is already engaged and ads add noise.
• Accounts in active renewal, which need customer success attention rather than acquisition messaging.
• Non-ICP domains that fit no segment you sell to, including public sector bodies and consumer brands outside your market.
• Your own employees, vendors and agencies, who research your category constantly.
• Competitors, who research you more than anyone.

Review the list quarterly and add whatever surfaced in your engagement reports. These groups often show high engagement. They browse a lot and buy nothing.

How to diagnose an ABM intent program that isn't working

Numbers read in isolation hide the problem. Read them in pairs and the diagnosis usually falls out.

Work one row at a time. Changing topics, thresholds and messaging in the same month leaves you unable to tell which change helped.

For the pipeline and revenue side of this, account based marketing metrics covers the KPIs worth reporting to leadership.

Benefits and limits of ABM intent data

What it does well:

• Allocates finite budget across a fixed list with something better than gut feel.
• Times outreach so a first call lands during active research rather than months after.
• Gives sales context that turns a cold opening into a relevant one.
• Surfaces expansion and churn risk inside existing accounts researching alternatives, especially when paired with the ability to track job changes.
• Aligns sales and marketing around one ranked list instead of two competing views.

Where it falls short:

• Category interest is not product interest. An account researching your category may be building a business case, evaluating a rival, or doing a student project. Weighing first party vs third party intent data helps separate the two.
• Account-level signals name companies, not people, so contact resolution stays your job.
• Coverage varies by market. Signal density thins outside North America and outside technology-adjacent categories.
• Noise scales with topic breadth. Broad topics generate constant flags that train reps to ignore alerts.
• It rewards teams with operational discipline and punishes teams without it. The data doesn't supply the strategy.

What an ABM intent program really costs

The data subscription is rarely the largest line.

1. Content production. Tier 1 plays need custom assets. Tier 2 needs cluster-level variants. Budget content by tier, not by campaign.
2. Headcount. One-to-one plays need a marketer's time per account. This is the constraint that caps Tier 1, not your score threshold.
3. Media budget. Account-targeted advertising costs more per impression than broad targeting because the audience is deliberately small.
4. Operations time. Someone owns topic configuration, suppression lists, routing rules and the monthly tier review. Unowned programs drift within a quarter.
5. The cost of a weak list. Every play aimed at an account that was never going to buy spends the same money as one aimed at a real buyer.

Pricing models for the data itself vary by topic count, account volume, platform tier and seat count. Intent data providers breaks down how those models compare.

SMARTe intent data for ABM

SMARTe builds buying signals and contact data into the same platform, which matters for the coverage gate described above. A flagged account is only workable if you can reach the people inside it.

Intent runs on Bombora's co-op network, where consent gets captured at the publisher rather than inferred from an ad auction. Alongside topic surges, SMARTe tracks funding events, leadership changes and headcount growth. Those carry a real event behind them, which makes them useful promotion triggers for tier changes.

The contact layer covers 289M+ verified profiles and 66M+ company profiles across 200+ countries, with 75%+ US mobile coverage and 50%+ global direct dial. Verified email reaches 86% of US decision-makers, and CRM enrichment matches at 90%+. Coverage runs deep in LATAM and APAC, which helps if your target list isn't North America only.

A few things support an account-based motion specifically.

• AI Agents map the buying group automatically once an account gets flagged, so coverage checks don't sit with a marketer.
• Technographic filters across 64K+ tracked products help build Tier 2 clusters around shared tech stacks.
• Native Salesforce and Outreach integrations push signals and context into the tools reps already work in.
• SMARTe Prospector, a browser extension for pulling verified emails and mobiles while researching accounts.
• Published pricing, starting free at 10 credits a month, with Pro at $25 a month and no per-seat charge.

SMARTe holds SOC 2 Type II certification, aligns with GDPR and complies with CCPA.

Where it stops: there's no predictive buying-stage model, and intent runs on Bombora rather than a proprietary collection engine. If stage prediction decides your evaluation, weigh that carefully.

Book a demo