The B2B Stack is free. I write one practical piece a week on B2B advertising, ABM and measurement. Subscribe free and the full archive unlocks.
What is an ABM platform?
An ABM (account-based marketing) platform is software that helps you choose the companies you want to win, spot which of them are showing buying behaviour, and put ads and outreach in front of the people inside those companies. 6sense, Demandbase, RollWorks (now unified with AdRoll) and StackAdapt are the names most B2B teams shortlist. They overlap more than their sales decks admit. In practice, what decides whether ABM works is rarely the logo you choose. It’s the target account list, the data underneath, how you define the buying committee, and whether anyone owns turning signals into pipeline.
This page brings together everything I’ve written on ABM tools and account-based advertising, in a sensible reading order. I’ve worked in adtech since 2011, and I run a B2B programmatic business that competes with some of these tools for budget, so I’ll flag that where it matters.
Where to start
If you only read three or four things, read them in this order:
6sense vs Demandbase vs StackAdapt: Do ABM Tools Suck, or Do You Need Managed Service?. It covers what each tool actually is and why most ABM failures aren’t vendor failures.
The CIO You’re Targeting Might Not Exist. This is why job-title targeting quietly breaks account-based programmes.
The Account Graph Is Not an Identity Graph. It covers the data question you should ask every vendor before you sign.
The ABM Measurement Glossary: the terms you need when the CFO asks what ABM delivered.
The platforms at a glance
This summary is drawn from my ABM tools comparison and the mid-market vs enterprise stack piece. It’s a starting point for a shortlist, not a substitute for a trial on your own data.
6sense: A predictive GTM brain: intent, account scoring and AI-driven workflows. Usually strongest for: Teams with clean CRM data and RevOps who will act on predictions. Where it tends to break: Brilliant with the right data, poor when misconfigured or fed a messy CRM.
Demandbase: A unified identity and account data spine. Usually strongest for: Enterprise teams that need account data joined up across systems. Where it tends to break: Powerful but heavy for mid-market. Cost and complexity can outrun the team.
RollWorks / AdRoll: ABM as a product line inside a broader ad platform. Usually strongest for: Mid-market teams that want simple, fast activation. Where it tends to break: Lighter on enterprise orchestration and governance.
StackAdapt: A mid-market DSP with ABM-style features. Usually strongest for: Programmatic activation at mid-market scale. Where it tends to break: Great for activation, but it isn’t a GTM operating system. Readers flag the CPMs.
Expert-managed B2B programmatic (what FunnelFuel does): A service plus an account graph and curated supply. Usually strongest for: Enterprise teams without the in-house time or skills to run ABM media weekly. Where it tends to break: You give up some hands-on control, and it’s a service cost, not a licence.
In November 2025, Demandbase was named a Leader in Gartner’s Magic Quadrant for ABM platforms, and 6sense launched an agentic GTM product. I covered both in the RollWorks vs Demandbase vs 6sense news round-up. Treat analyst badges as one input. The Reddit threads I quote in the comparison post are often more honest about the day-to-day.
Mike’s take: I’ve sat in a lot of “should we switch ABM platforms?” meetings. Almost every time, the platform wasn’t the thing that was broken. The target account list hadn’t been touched in a year, Salesforce was a mess, and nobody owned what happened after an account “surged”. One reader comment I quoted sums it up: “We bought an ABM platform, and then realised we didn’t even have a TAL.” Fix that first. Then judge the tool.
Mid-market and enterprise need different stacks
The same tool can be right for one company and wrong for its neighbour, because mid-market and enterprise buying behave differently. In The Hidden Faultline in B2B I set out the split:
Typical customer size: Mid-market: About 100–1,000 employees. Enterprise: About 1,000–100,000+ employees, global buying groups.
Sales cycle: Mid-market: Roughly 30–120 days. Enterprise: Roughly 6–18 months, often non-linear.
Stakeholders: Mid-market: Often 1–3 drive the deal. Enterprise: 10–20+ across roles, regions, procurement and InfoSec.
Core systems: Mid-market: 5–10 tools, easy to buy and replace. Enterprise: 20–40 interconnected systems.
Programmatic: Mid-market: Minimal (StackAdapt, RollWorks or built-in ABM bidders). Enterprise: Specialised, curated supply partners.
What buyers want: Mid-market: “Show me it works fast”. Enterprise: “Integrate with our world, don’t break anything”.
If your segment is shifting upmarket, your stack needs to shift with it. Bolting enterprise architecture onto mid-market maturity is one of the most common reasons teams decide “ABM tools suck”.
The B2B Stack is free. If this is useful, subscribe free for a new teardown every week, plus the full archive.
Target the buying committee, not a job title
Every account has a buying committee. Not every account has a CIO. Most ABM programmes still do a two-step lookup: build the account list, then find the CIO or CTO inside each account and point the media at them. It feels precise. But org charts differ from company to company and decay fast. In The CIO You’re Targeting Might Not Exist I note that average tenure in a US role is now around 18 months, and B2B email addresses decay at roughly 25–30% a year.
The committee is also far bigger than most media plans assume:
Forrester’s State of Business Buying research, which I cover in There Is No B2B Funnel, puts the average complex purchase at 13 internal stakeholders plus 9 external participants: analysts, consultants, peers, references.
Gartner puts the typical committee at 6–10 stakeholders for deals above $100,000, up from 4–6 as recently as 2017. Forrester’s research on deals above $1 million puts it at 14–23. I set out both in The Account Graph Is Not an Identity Graph.
In our own enterprise tech work we often see committees of 20+, 60+ touchpoints and buying cycles of 11+ months (B2B Is Not Just ABM).
The better approach is to infer the buying group from signal: who is actually researching, from which offices, across which functions. A pre-canned title list misses the people who matter. That’s especially true of the “shadow” members, such as legal, security, procurement and outside advisers, who never appear on a TAL export. LinkedIn has the same blind spot, which I go into in Why LinkedIn Ads Miss Most of the Buying Committee.
Mike’s take: A job title tells you what someone’s badge says. It doesn’t tell you who’s in the room. In B2B the maths favours casting the net a bit wider across the committee: one missed influencer on a six-figure deal costs far more than a few wasted impressions.
The data underneath: account graphs, firmographics and TALs
An account graph is not an identity graph
An identity graph stitches identifiers together (IP addresses, cookies, device IDs, hashed emails) so a fact about one person can be reached across their devices. An account graph maps how real companies are structured (parents, subsidiaries, offices, roles) and then resolves identity onto that structure. Most of the market has the first. Very few have properly built the second, and a lot of what gets pitched as an “account graph” is an identity graph with a company name bolted on. When you talk to a vendor, ask whether it’s a reach graph (how many identifiers it can address) or a proof graph (whether it can show which accounts, and which people inside them, actually engaged). Read The Account Graph Is Not an Identity Graph and the earlier product perspective on the account graph.
Firmographic codes are older than you think
A lot of TAL building still leans on SIC and NAICS codes. SIC got its first real list of industries in 1938 and was last substantially revised in 1987. Microsoft’s SEC filings still carry SIC 7372, “Services, Prepackaged Software”, and Dell’s classification is 3571, “Electronic Computers”. If that’s the filter deciding who gets onto your target account list, everything downstream inherits its blind spots. The fix is to index what companies actually publish about themselves, and keep re-indexing. That’s in Your Firmographic Targeting Is Running On A 1937 Filing System.
TALs aren’t dead, but static ones are
Static target account lists can’t keep up with how quickly buying behaviour moves now. List hygiene has become an ongoing signal-governance job rather than an annual spreadsheet refresh. See TALs Are Dead. Long Live TALs.
Running ABM so it shows up in pipeline
Score engagement, not activity. In the ABM Measurement Glossary I explain why the same account can legitimately score 85 in one platform and 40 in another. The trend matters more than the number. An account moving from 40 to 70 over six weeks is a signal. A 70 on its own is decoration.
Separate sourced from influenced pipeline, and report both with clear labels. This is the source of most marketing-vs-finance arguments. There’s more on this in the measurement hub.
Don’t confuse ABM with B2B. ABM still wins for named, high-value accounts. A lot of B2B growth also comes from signal-based marketing to a wider market that isn’t on any list yet. See B2B Is Not Just ABM, The Signals Era Needs to Deliver Pipeline, Not Just ABM and the ABM 2.0 playbook.
Watch the match-rate claims. Vendors tout “95% match to your TAL” as a badge of honour. Ask what was matched, how, and at what confidence (B2B Is Not Just ABM).
Plan by region. Account-based advertising doesn’t behave the same in every market. See Running Account-Based Advertising Globally.
Mike’s take: The ABM programmes I’ve seen work all had someone whose job was the bit after the dashboard: deciding which accounts got sales attention this week, and why. The ones that stalled had a great platform and nobody holding the pen.
Checklist: before you buy (or blame) an ABM platform
Is your TAL current? When was it last rebuilt from real signal rather than firmographic codes alone?
Is your CRM clean enough for predictive scoring to mean anything?
Mid-market or enterprise? Match the tool to the segment you actually sell into (see the table above).
Who owns the next step when an account surges? Name the person and the response time.
What sits under the “account graph”? Ask the vendor: identity resolution only, or real company structure?
How will you measure it? Agree sourced vs influenced definitions with finance before launch.
Who will run the media every week? If nobody in-house has the time and programmatic skills, factor in managed service or agency support. Otherwise you’ll end up with an expensive dashboard.
What does “match rate” mean in the vendor’s own words, and at what confidence?
If you can’t answer 1, 4 and 7, changing platform probably won’t fix the problem.
Related hubs
FAQ
What is the difference between 6sense and Demandbase? 6sense is best understood as a predictive GTM engine, built around intent, scoring and AI workflows. Demandbase is a unified identity and account data spine. 6sense rewards teams with clean data that act on its predictions. Demandbase rewards enterprise teams that need joined-up account data, but it can be heavy for mid-market.
Is RollWorks good for mid-market ABM? Usually, yes. RollWorks, now unified with AdRoll, is built for simpler, faster activation. It’s lighter on the orchestration and governance that large enterprise programmes need.
Is StackAdapt an ABM platform? Not really. StackAdapt is a mid-market DSP with ABM-style features. It’s good for programmatic activation, but it isn’t a full GTM operating system.
Do ABM tools actually work? They work when the foundations are in place: a current target account list, clean CRM data, and someone accountable for acting on the signals. Most “ABM tools suck” complaints are really about running enterprise tools without those foundations.
Should we run ABM ourselves or use a managed service? Run it yourself if you have in-house programmatic skills and the time to manage it every week. Enterprise teams without that often get better results from expert-managed ABM. I run a managed service, so weigh that view accordingly.
How big is a B2B buying committee? It depends on deal size. Forrester puts complex purchases at about 13 internal stakeholders plus 9 external participants. Gartner cites 6–10 for deals above $100,000, and Forrester cites 14–23 for deals above $1 million.
What is an account graph? An account graph maps how real companies are structured, including parents, subsidiaries, offices and the roles inside them, and then resolves identity onto that structure. An identity graph only stitches devices, cookies and emails together.
Why is job-title targeting unreliable in ABM? Org charts differ between companies and decay quickly. US role tenure averages around 18 months, and B2B emails decay at roughly 25–30% a year. Many real decision-makers don’t hold the title you’re targeting.
What is a target account list (TAL)? A TAL is the set of companies your sales and marketing teams have agreed to prioritise. It only works if it’s rebuilt regularly from current signal, not left as a static spreadsheet.
How do you measure ABM success? Track account-level engagement trends, buying-committee breadth and pipeline. Report sourced and influenced pipeline separately, and treat the trend in an engagement score as the signal rather than the score itself.
Subscribe
The B2B Stack is free. You get one practical piece a week on B2B advertising, ABM and measurement, written from inside live campaigns. Subscribe free.
I’m Mike Harty. I’ve worked in adtech since 2011: I built PowerLinks, and I’m now Co-Founder and CPO of FunnelFuel, a B2B programmatic and analytics business. The B2B Stack is where I write up what I see on live campaigns, without the vendor spin.