Twenty-two.
That’s how many people Forrester found are involved in a typical enterprise B2B purchase decision in 2026. Thirteen inside the company. Nine outside it.
Outbound • Inbound
The 13 inside are who your marketing team is probably already targeting economic buyer, technical evaluator, procurement, finance, legal, IT, security, the champion who started the whole thing. With clean data and a reasonable budget, paid media can reach them. LinkedIn matches your account list. Programmatic platforms ingest your named-account universe. The infrastructure exists.
Industry analysts whose notes get cited in board discussions. Peer references your champion calls before signing. Vertical communities Slack groups, subreddits, professional associations. The CFO’s trusted consultant. The colleague at another company your buyer worked with two roles ago. AI search responses, when someone asks Perplexity, “which vendors should we evaluate?”
Stop there for a second.
The playbook most B2B SaaS companies are still running was built for a world where a few champions could carry a decision through their organisation. That world is gone. The unit of decision isn’t a person anymore. It’s a network of 22 humans and your current programme is probably only talking to half of them.
Here’s the part that doesn’t show up in your ABX dashboard.
Three structural failures keep getting in the way.
The audience problem.
Most ABX personas are built in a conference room. The team brainstorms what the CISO cares about, what the CDO wants, without ever asking an actual CISO or CDO. The persona is a guess that hardens into a strategy. And the contact data underneath it is wrong roughly 14% of the time wrong title, wrong email, wrong person entirely. Then it gets worse: IT and Security show up in only 10% of typical targeting but raise 38% of the objections that lose deals. You may be aiming at the wrong people, with wrong data, and missing the ones who block you.
The content problem.
You have one message for 22 humans with entirely different jobs to be done. The CISO wants breach-cost ROI. The CDO wants integration patterns. Procurement wants vendor risk evidence. Engineers want API documentation. The analyst wants the proprietary research that makes your category claim defensible. Forrester finds 86% of B2B purchases stall in the process. Most of those stalls trace back to the same root cause: the buying network didn’t get the content it needed, when it needed it, in a format each role could actually use. One message for 22 people isn’t a content strategy. It’s a guess.
The execution problem.
Most teams test too slowly, with too little variety. A handful of creative variants cannot cover 22 humans across five buying stages and multiple offer types. The math doesn’t work. What works is cross-platform testing at twice the typical pace the same message variants running in parallel across LinkedIn, Google, and content syndication, not sequentially on one platform. Budget moving to validated tactics the same week the data clears the threshold. And audience surveys running before paid media goes live, so you’re deploying budget against a validated message rather than your best guess.
All three failures share a root cause: the way most marketing organisations are structured. Paid sits with one team. Owned with another. Earned with an agency outside the building. Each optimises for its own metric. A flywheel that connects all three requires the opposite one team, one contact database, one revenue metric, one cadence.
The fix is structural, not tactical.
Stop treating the account as the unit of marketing. Start treating the buying network all 22 named humans as the unit. When you do, the operating model changes. Paid media stops being an awareness exercise and becomes a contact-level database: every LinkedIn impression, every content syndication download, every programmatic engagement produces a named record with engagement context attached. Over 12 to 18 months, thousands of buying-network contacts accumulate in your CRM not just the ones who filled out a form.
That database fuels the owned engine: founder content, employee advocacy, podcast appearances, community work. Which builds the earned corpus: research releases, analyst placements, bylines the material AI search engines pull from when your next buyer types in their question. Which brings in net-new in-market accounts. Which paid captures. Which the whole cycle runs again.
The marketing leaders who will look back on 2026 as the year things shifted won’t be the ones who spent more. They’ll be the ones who changed the unit of measurement from account to network and built the architecture to match.
That architecture exists. The path through this is clearer than it feels right now.
