What is ABM data, actually?
Account-based marketing treats individual companies as markets of one. Instead of casting a wide net and qualifying inbound leads, you choose a defined set of target accounts and orchestrate outreach across every relevant contact within each one. The data question follows directly from that: you need to know which companies to target, who inside them to reach, what they care about, and how to contact them.
Those four requirements map to four data types:
- Account list data: the specific companies you have decided to pursue, with firmographic detail (industry via UK SIC 2007 code, employee count, turnover band, geography, Companies House registration status).
- Contact data: named individuals within each account, with job title, job function, seniority, direct telephone, business email, and LinkedIn URL.
- Intent data: signals that a company or individual is actively researching topics relevant to your product, typically sourced from B2B publisher content networks.
- Tech-stack data: which software platforms an account runs (CRM, ERP, marketing automation, cloud infrastructure), relevant when your product integrates with or displaces a specific tool.
The critical point is that these four types are not interchangeable. Buying intent data does not tell you who the Finance Director is. Buying firmographic data does not tell you whether the company is actively evaluating your category right now. Each layer answers a different question, and the contact layer is the one most buyers underinvest in.
The two-layer model: what to build vs what to buy
A well-structured ABM programme separates data responsibilities cleanly. Layer one, the account list, is yours to own. Layer two, the contact data, is where a quality data provider adds genuine value.
| Data element | Build or buy? | Rationale |
|---|---|---|
| Named account list (which companies) | Build internally | Must reflect your ICP, pipeline gaps, strategic priorities, and existing customer lookalikes. No vendor knows your commercial strategy. |
| Firmographic data (headcount, SIC, turnover) | Buy or append | Appending Companies House and public sources to your account list gives you the segmentation detail to prioritise tiers. |
| Named contacts and buying committee | Buy | Direct dials, business email, LinkedIn URL for specific job functions and seniority levels per account. This is the core purchased data layer. |
| Intent data overlay | Buy (as overlay only) | Useful for sequencing and prioritisation, not for list building. Signals decay within weeks. |
| Tech-stack data | Buy (as overlay only) | Relevant for messaging personalisation when your product integrates with or competes against a specific platform. Verify UK coverage before purchasing. |
| Pre-packaged ABM target account list from a vendor | Skip | These are generic prospect lists dressed in ABM language. Account selection must be strategy-driven, not vendor-driven. |
Why bundled ABM lists fail
Several data providers and sales intelligence platforms sell products marketed as "ABM lists" or "ABM-ready accounts." The pitch is seductive: they have pre-identified companies showing buying signals, scored by fit and intent, ready to import into your CRM. In practice, the model breaks down for three reasons.
First, account selection is a strategic act. The companies on your ABM target list should reflect where your business can genuinely win: existing customer lookalikes, whitespace in your territory plan, companies where you have a referral connection, or deals you lost two years ago that are now due a re-evaluation. None of that context lives with a data vendor.
Second, you do not control the inclusion criteria. When a vendor builds a "fintech ABM list" or a "manufacturing ABM list," they are applying their own scoring model to their own data. You have no visibility into what signals drove the selection, whether those signals are current, or whether half the accounts are already customers of your closest competitor.
Third, every other company buying from the same vendor gets the same list. In markets with concentrated buying committees, that means multiple vendors simultaneously reaching the same Finance Directors at the same 300 accounts, all claiming their ABM programme is personalised.
The correct sequence is: define your ideal customer profile internally, build your named account list from your own data and commercial reasoning, then go to a provider to purchase contact data for those specific accounts.
How to build a named account list yourself
Before you buy a single contact record, you need a documented account list. The inputs are almost entirely internal.
Step 1: define your ideal customer profile
Your ideal customer profile (ICP) describes the company characteristics that correlate with your best customers: those who close fastest, renew longest, expand most, and refer others. Pull your top 20 customers and look for patterns. Industry by UK SIC 2007 code is the most reliable firmographic signal. Headcount and turnover band tell you which tier of the market you serve best. Geography (region, postcode district) matters for field sales and for services with local delivery constraints.
Step 2: score accounts against fit
Once you have ICP criteria, apply them to a universe of companies. Companies House data via the free public API gives you registered company name, SIC code, registered address, and filing status for every active UK company. That is enough to build an initial pool of fit-matching accounts. You can append employee count and estimated turnover from a data provider as a second pass.
Step 3: layer in intent signals
Intent data from B2B publisher networks shows which accounts in your pool are actively consuming content on topics you own. A software firm targeting logistics companies might overlay intent signals on "warehouse management" or "last-mile delivery" research activity. This does not replace fit scoring; it re-ranks accounts within your already-qualified pool so your sales team calls the warmest names first.
Step 4: apply commercial context
The final filter is the one only you can apply. Accounts where a relationship exists, even a cold one, go higher. Accounts in your CRM marked as "closed lost" 18 to 36 months ago often deserve re-engagement. Accounts with known incumbent competitors you displace regularly are worth prioritising. That context never appears in a purchased dataset.
Buying contact data per account: what to specify
Once you have your named account list, you can approach a B2B data provider with a specific brief. The quality of that brief determines the quality of what you receive.
Specify contacts by company (by Companies House number if possible, not just company name, since duplicate trading names cause matching errors), job function, and seniority level. For a mid-market SaaS product targeting Operations teams, you might request: Operations Director, Head of Operations, VP Operations, and Operations Manager at seniority levels Head of and above, for each of your 150 named accounts.
The fields you should always request for each contact are:
- Full name
- Job title (verbatim, not normalised)
- Job function (standardised category, e.g. Operations, Finance, IT)
- Seniority level
- Direct telephone number
- Business email address (including personal-name format at corporate domain)
- LinkedIn URL
- Company name and Companies House number
- Office postcode
Do not request role-based generic emails (info@, sales@, enquiries@) as your primary contact field. In a true ABM programme, you are reaching named individuals, not hoping a generic inbox gets forwarded to the right person. Generic inboxes also carry higher spam-complaint risk and weaker deliverability on email channels.
On choosing a B2B data provider, ask specifically about their UK coverage for mid-market companies. Many data platforms skew towards FTSE-listed and large enterprise records where public information is abundant; coverage thins out for companies with 50 to 500 employees, which is where most ABM programmes sit. Request a sample match rate against five or ten of your target accounts before committing.
Multi-contact targeting within an account
Single-threaded ABM is a common failure mode. Most B2B deals involve 6 to 10 people touching the decision according to research from Gartner's B2B buying group studies, and that figure is higher in regulated industries or where IT security review is mandatory. If your entire ABM strategy runs through one contact per account, a job move, a maternity leave, or a disengaged champion kills the deal entirely.
For one-to-one ABM targeting 10 to 30 strategic accounts, aim for 5 to 10 named contacts per account: the economic buyer (typically CFO or CEO for large purchases), the technical evaluator (IT Director, Head of IT, CTO), the business-line champion (the job function most directly affected), a procurement or legal contact for deals over a certain threshold, and at least one end user who will influence the internal recommendation. The exact shape depends on your deal size and your product category.
For one-to-many ABM across 50 to 200 accounts, 3 to 5 contacts per account is a workable starting point. That gives you enough coverage to survive personnel changes without making the data spend prohibitive. In our experience, campaigns running to 3 or more contacts per account show materially higher pipeline conversion than single-contact approaches, particularly in the 90 to 180-day sales cycles typical of mid-market B2B.
The job function and seniority targeting logic that applies to broad B2B campaigns applies here too, just applied at account granularity. The article on job function and seniority targeting for B2B campaigns covers the matrix of function and level in detail; the principle is identical for ABM, you are just restricting the universe to your named account list rather than a broad sector or geography.
Tech-stack and intent overlay data: useful but not foundational
Technology intelligence data (which platforms a company runs) is sold by several providers and appended via company domain or Companies House number. It is genuinely useful for ABM messaging personalisation. If your product replaces a specific piece of software, knowing which accounts run that software lets you reference the incumbent by name in your outreach, which lifts reply rates measurably.
Two caveats apply. UK coverage is uneven. Technology intelligence is derived largely from publicly visible signals (job postings mentioning specific tools, website scripts, API integrations that surface in public documentation). Large enterprises with rich digital footprints are well-covered. A 120-person manufacturing firm in Coventry with an outdated website and no public API presence may have no tech-stack record at all, even though it runs exactly the ERP system you want to displace.
Intent data has a shelf life of two to four weeks before the signal fades. Buying intent data, sitting on it, and actioning it six weeks later produces noise, not pipeline. If you are going to use intent data in your ABM programme, you need a process to act on it quickly: the intent feed should trigger outreach within five to ten business days, not sit in a spreadsheet waiting for a quarterly review.
Lawful basis under legitimate interests for ABM contact data
Under UK GDPR Article 6(1)(f), legitimate interests is the correct lawful basis for B2B outreach to purchased contact data, including ABM campaigns. The three-part test (purpose, necessity, balancing) applies in the same way whether you are mailing 50,000 cold prospects or 300 carefully selected named contacts.
If anything, the ABM context makes the legitimate interests case stronger. You are contacting individuals at companies whose profile closely matches your existing customers, the relevance of your message to their professional role is high, and the volume of intrusion is lower than broad-reach campaigns. The Information Commissioner's Office (ICO) has long held that relevant, proportionate B2B outreach to professional contacts is a reasonable use of legitimate interests, provided you complete and document the assessment.
The practical requirements remain the same. Complete a Legitimate Interests Assessment (LIA) before the campaign launches. Ensure your data provider can evidence that the contact data was compiled under legitimate interests from publicly available sources. Include a clear opt-out mechanism in every communication. Honour opt-outs promptly and record them in a suppression file. For telemarketing, wash your dial list against the Telephone Preference Service (TPS) every 28 days.
ABM does not create a special compliance regime, but it does create a natural audit trail. Because you have documented account selection criteria and a defined buying committee per account, you can show exactly why a given individual received your outreach and demonstrate the proportionality argument clearly. That documentation is worth keeping alongside your LIA.