You've invested in a CRM and a marketing automation platform. You're capturing leads and running campaigns. Now your leadership wants more: a real account-based marketing (ABM) program. Targeted outreach to the right people at the right accounts, with messaging that maps to their role in the buying process.
So you stand up your ABM platform, build your target account list, and start pushing campaigns. And then the results disappoint. Coverage is thin. Contacts at key accounts are missing or misclassified. Your "VP of Marketing" segment pulls in coordinators and assistants. Your enterprise tier includes companies that are nowhere near enterprise size.
The ABM strategy isn't the problem. The segmentation powering it is.
Account-based marketing lives or dies on the quality of the data underneath it. ABM requires knowing not just who is at a target account, but what role they play, what function they sit in, and whether the account actually fits your ICP. Without accurate, consistently classified data across all four of those dimensions — job level, job function, company size, and industry — your ABM program is firing at a blurry target.
This article covers the four segmentation dimensions that matter most for ABM, what it takes to make each one work, and what leading B2B companies have achieved when they get the data right.
Why ABM segmentation is a data quality problem
The fundamentals of B2B segmentation haven't changed much since they were first formalized. You still need to know who you're talking to, what role they play in a buying decision, and whether their company fits your ICP. For ABM specifically, accurate segmentation enables four critical activities:
- Building target account lists based on data rather than intuition, anchored to firmographic attributes that match your best existing customers
- Identifying gaps in account coverage by role and seniority, so you know which buying committee members you're missing at each target account
- Triggering personalized, role-specific content and outreach for each stakeholder in the buying group — the economic buyer, the technical evaluator, the end user, the champion
- Managing suppression lists so that contacts outside your ICP don't dilute engagement metrics or trigger sequences intended for high-priority accounts
What has changed is the expectation of how fast and accurately segmentation needs to happen. In a batch-processing world, stale segmentation data produced suboptimal campaigns. In a real-time ABM environment, inaccurate segmentation data breaks the entire program: wrong contacts get added to account plays, buying committee maps are incomplete, and personalization logic fires against the wrong persona.
1. Job level: know where every contact sits in the buying committee
For ABM, job level isn't just a campaign filter — it's the primary input for buying committee mapping. You can't build a complete picture of stakeholder coverage at a target account if you don't know which contacts are decision-makers, which are influencers, and which are end users. Job level is what tells you that.
Job level can be inferred from job title, but the inference logic is less straightforward than it looks. "VP" typically indicates an executive — except in financial services, where VP is often equivalent to a senior individual contributor. "Director" usually means a manager, but "Director of Architecture" at a large enterprise can carry more decision-making weight than a VP of IT at a smaller company. "Assistant to the VP of Marketing" contains the word "VP" but refers to a coordinator. Rule-based filter stacks in a MAP can't handle this kind of contextual variation reliably.
For ABM, the cost of getting this wrong is high. If your platform classifies a coordinator as an executive, you waste a high-touch sequence on the wrong person. If it misses a decision-maker because their title doesn't match your keyword list, that contact never enters the right account play. Either way, your buying committee map is wrong and your ABM program underperforms against an account it should be closing.
Clari's experience illustrates how much buying committee accuracy matters. After deploying Openprise to replace fragile Salesforce APEX code and clean up their attribution data, their team built three attribution models in just weeks. One of the immediate discoveries: the finance team was a key role in the buying committee that had previously been invisible to their segmentation and targeting. Identifying that gap directly improved their conversion rates. That kind of insight only surfaces when job level and function data are clean and consistently classified.
The more scalable approach to job level classification uses a classification engine trained on large datasets of job titles across industries, with the ability to customize outputs for your specific business. Openprise's job title segmentation uses a catalog of thousands of job title keywords combined with contextual logic that handles edge cases like the coordinator-with-VP-in-title problem natively — delivering accurate classification up to three times more often than third-party enrichment vendors, which typically achieve job level match rates of around 30%.
2. Job function: know what each contact does and who they buy for
Job function is the segmentation dimension that tells you which persona a contact maps to and which part of your ABM playbook applies to them. For multi-stakeholder enterprise deals, you're rarely talking to one buyer type. The economic buyer, the technical evaluator, the practitioner, and the IT decision-maker each need different messaging, different content, and different sequences — and function is what routes them to the right one.
A practical ABM-oriented job function segmentation scheme uses two levels:
- A coarse-grain tier that sorts contacts into major business functions (Finance, Sales, IT, Marketing, Operations, HR, Legal) for use cases like suppression, exclusion from irrelevant account plays, and high-level buying committee coverage analysis
- A fine-grain tier that creates sub-function segments within the categories most relevant to your product — for example, within IT: security, networking, infrastructure, architecture — enabling persona-level targeting within the same account
The fine-grain tier is where ABM personalization lives. It's what lets you deliver a security-specific case study to the security architects at a target account while serving a different message to the CIO and a third to the VP of Finance — even when all three are in your database under the same parent company domain.
Nutanix offers a concrete example of what happens when job function data is poor at scale. When their ops team examined job title and function match rates in the APAC region, they found that only 15% of records had accurate classification. With that level of data quality, a meaningful ABM program in the region was effectively impossible — you can't orchestrate buying committee outreach when 85% of your contacts are in the wrong function bucket. After deploying automated classification with Openprise, that figure rose to 75%, enabling reliable persona-level segmentation and scoring across a region that had previously been unworkable.
3. Company size: know whether the account fits your ICP and your motion
For ABM, company size is the gate. It's the primary filter that determines whether an account belongs on your target list at all, and if it does, which product tier, which sales motion, and which ABM play applies to it.
Most B2B companies run different programs for different size tiers. Enterprise accounts get high-touch, multi-threaded account plays with executive engagement and custom content. Mid-market accounts get a more templated approach with less custom investment. SMB accounts might not be in ABM at all. Misclassifying an SMB as enterprise doesn't just waste ABM budget — it also directs a high-touch sales motion at an account that will never close at enterprise deal sizes.
Two complications make company size harder to maintain accurately than it looks. First, the data decays. Companies grow, get acquired, spin off divisions, and contract. A record classified as mid-market 18 months ago may belong in a different tier today, and without continuous enrichment feeding updated firmographic data into your CRM, your tier assignments drift quietly out of accuracy.
Second, single-vendor enrichment misses too many accounts. Enrichment vendors typically achieve match rates of around 50% on their own, which means up to half of your target accounts could have missing or stale company size data — a significant gap for any ABM program trying to maintain accurate tier assignments across a large account list.
JumpCloud ran directly into this problem when building out their segmentation infrastructure to support a shift from product-led growth to a sales-led ABM motion. With no ICP defined and incomplete firmographic data, they had no reliable way to identify which accounts to prioritize. By implementing a multi-vendor enrichment waterfall through Openprise — sequencing providers to maximize match rate and fill rate — they increased their contact match rate by 48% and expanded their total addressable market by 3x within 90 days. That TAM expansion wasn't new accounts magically appearing; it was existing accounts that now had enough accurate firmographic data to be correctly identified as fitting the ICP.
4. Industry: know which vertical plays apply to each account
For ABM, industry segmentation is what enables vertical-specific plays — the industry-specific case studies, the vertical-focused event invitations, the regulatory-context messaging that makes outreach feel relevant rather than generic. Account-based programs that send the same content to a financial services firm and a manufacturing company are leaving a significant conversion advantage on the table.
The data challenge with industry segmentation is translation. Standard industry classifications — NAICS codes, SIC codes — are highly granular, running to several thousand distinct categories. Most B2B marketing teams have ten to fifteen industries that are actually relevant to their business. The work is mapping the granular standard taxonomy to your internal list and doing it consistently across every data source feeding your database.
This becomes more complex in an ABM context, where the industry classification at the account level needs to match across your CRM, your MAP, your ABM platform, and any intent data or technographic data you're layering on top. If your CRM classifies an account as "Technology - Software" and your intent data provider uses "Computer Software," and your ABM platform can't reconcile them, your account-level segmentation breaks down even if each individual field value is technically correct.
Adobe's ops team encountered a version of this problem at significant scale, managing 2.5 million account records in their corporate master data management system. Their marketing team was manually uploading and enriching 800 lead lists per month, with no automated process for cross-referencing those lists against account records or maintaining consistent industry and firmographic classification. After deploying Openprise to automate the list loading and matching process, they improved SMB account data match rates from 50% to 88% — the kind of match rate improvement that makes account-level segmentation reliable enough to actually build ABM plays on.
What's new: AI-assisted ABM segmentation and the real-time imperative
The four segmentation dimensions above have been the foundation of B2B marketing ops for over a decade. What's changed is the infrastructure required to make them work reliably in an ABM context.
Modern ABM programs don't tolerate segmentation latency. When a new contact at a target account fills out a form or is added via a list import, the account play should be able to start immediately — which means job level, function, company size, and industry need to be classified at the point of record creation, not during a downstream campaign setup step. If a contact arrives without accurate segmentation fields, they either sit unworked or get routed to the wrong play.
AI is also changing how ABM segmentation logic gets built and maintained. Rather than constructing and maintaining filter trees in a MAP that break down as job title conventions evolve, teams are increasingly using classification models that handle novel title patterns without explicit rule updates — a significant operational advantage for global ABM programs where title conventions vary by region and industry.
The common thread across the highest-performing ABM teams is that they've stopped treating segmentation as a pre-campaign task and started treating it as a data infrastructure problem. Nutanix, JumpCloud, Clari, and Adobe all arrived at the same conclusion: clean, classified, continuously updated data is the prerequisite for any account-based program to perform. Zendesk, which automated its data quality processes including segmentation, normalization, and enrichment with Openprise, saw a 25% improvement in data cleansing efficiency and a 25%+ increase in marketing and sales team efficiency — with $500,000+ in productivity gains that freed the ops team to run more proactive, higher-impact programs.
What it takes to get ABM segmentation right
To summarize the key requirements for reliable ABM segmentation in a modern ops environment:
- Job level and function need to be derived from raw job title data using classification logic that handles industry variation and contextual edge cases, so buying committee maps are accurate and complete
- Company size needs to come from a multi-vendor enrichment waterfall that maximizes match rate and fill rate, mapped to internal tiers that reflect how your business actually segments the market
- Industry needs to be translated from standard classifications into your internal taxonomy through a consistent normalization process applied across all data sources and platforms
- All four dimensions need to be populated at the point of record creation — not during campaign setup — for real-time ABM routing, AI scoring, and account play enrollment to work correctly
The teams that get this right don't just run better ABM programs. They build the data foundation that makes every account-based motion — from buying committee identification to personalized sequencing to pipeline attribution — more accurate and more reliable.
Want to go deeper? Download The modern B2B segmentation handbook or schedule a demo to see how Openprise handles job title segmentation, firmographic enrichment, and real-time lead classification in practice.
.jpg)
