Identify where customers split into different paths

At the same awareness stage, different people doubt different things. That's where segments come from.

Traditional segmentation asks: who are they? Age, role, company size, industry. But these demographics don't predict buying behaviour. Two people with identical demographics can have completely different beliefs and need completely different messages.

Belief-based segmentation asks: what do they doubt? What's stopping them from moving forward? Two companies at the solution-aware stage might both be comparing training vendors, but one doubts effectiveness whilst the other doubts ROI proof. They're different segments requiring different campaigns.

This chapter shows you how to identify belief splits using customer research, spot patterns where different people doubt different assumptions, and name segments based on their specific doubts rather than demographics.

The best way to find splits is through customer interviews. Ask 6 people (2 best customers, 2 churned customers, 2 lost prospects) the same question: "What almost stopped you from buying?"

Their answers reveal belief gaps. Write them down word-for-word. Look for patterns where different people mention different concerns.

For cybersecurity training, you might hear: "We weren't sure training would actually change behaviour" (doubts effectiveness), "We needed it set up immediately after a breach" (doubts speed), "We couldn't prove ROI to get budget approval" (doubts measurability), "We were comparing you to two other vendors" (no doubt about training, just comparing products).

These are four different doubts at potentially the same awareness stage. Someone who doubts effectiveness needs proof that training works. Someone who needs speed wants evidence of fast deployment. Someone who needs ROI proof wants data and case studies. Someone comparing vendors wants feature differentiation.

Map each doubt to the belief journey from the previous chapter. Which assumption are they stuck on? Problem-level (does this problem matter?), solution-level (does this solution work?), product-level (is this the right product?), or purchase-level (can we afford it, can we implement it, is timing right?).

Create a spreadsheet. Rows are the assumptions you documented in chapter 1. Columns are the people you interviewed. In each cell, mark whether this person believes, doubts, or is neutral on this assumption. Patterns will emerge showing which assumptions different people doubt.

Once you've identified the splits, name your segments based on what they doubt. Not their job title, company size, or industry. Name them after their belief gap.

For cybersecurity training: compliance-driven (doubts effectiveness, just needs to check the compliance box), breach-reactive (doubts speed, needs instant deployment after an incident), proactive (doubts ROI proof, needs business case before budget approval), tool-chooser (believes in training, just comparing vendors).

These names immediately tell you what messaging to use. Compliance-driven needs fast, checkbox messaging. Breach-reactive needs "set up in 30 minutes" positioning. Proactive needs data on behaviour change and breach reduction. Tool-chooser needs feature differentiation and comparison tables.

Bad segment names: SMB versus enterprise (demographic), IT versus HR buyers (role), finance versus healthcare (industry). These tell you nothing about beliefs. An IT buyer at a bank and an HR buyer at a bank might be the same segment if they have the same doubts.

Good segment names: price-sensitive (doubts affordability), implementation-worried (doubts technical feasibility), risk-averse (doubts vendor stability), feature-focused (doubts capability match). Each name describes a belief gap that campaigns can address.

Limit yourself to 3-5 segments. Too few and you're too generic. Too many and you can't build distinct campaigns for each. Find the belief splits that matter most (affect buying decisions) and create segments around those.

After naming your initial segments, validate them with more customer conversations. Describe each segment to customers and ask: "Which of these sounds most like your situation when you were evaluating us?"

If they immediately recognise themselves in one segment, you've nailed it. If they say "a bit of all of them" or "none of these really fit", your segments are too generic or too narrow.

Also check: do different segments actually need different campaigns? If two segments would get the same ad creative and land on the same page, they're not different segments. Combine them.

The test is: would you write significantly different messaging for these two groups? If yes, they're distinct segments. If no, they're the same segment with minor variations.

For lead generation tools, you might find: paid-sceptic (doubts outbound works at all), LinkedIn-first (thinks LinkedIn beats email), list-sceptic (doubts list building ROI), tool-chooser (comparing vendors). Each needs completely different proof. Paid-sceptic needs economics showing outbound ROI. LinkedIn-first needs deliverability data proving email still works. List-sceptic needs time-to-ROI case studies. Tool-chooser needs feature comparison.

These are validated segments because the campaigns are completely different. If you tried to collapse paid-sceptic and LinkedIn-first into one segment called "outbound doubters", you'd write messaging that convinces neither.

Different segments appear in different channels and at different awareness stages. This mapping tells you where to find each segment and what stage they're at when you reach them.

Compliance-driven buyers often respond to LinkedIn ads (they're active on professional networks) and Google search for "compliance training" (solution-aware, searching actively). Breach-reactive buyers only appear in remarketing pools after a breach is announced (suddenly product-aware and urgent). Proactive buyers engage with content marketing and organic search (problem-aware to solution-aware, building a case slowly).

Create a matrix: rows are segments, columns are channels. Mark which channels effectively reach which segments. Also note what awareness stage that segment typically enters at in each channel.

Example: paid-sceptic segment in a lead gen tool. They're problem-aware (know they need more leads) but doubt outbound works. They respond to LinkedIn ads showing economics, content marketing explaining outbound ROI, maybe Google search for "lead generation alternatives". They don't respond to cold outreach (they doubt it works) or competitor comparison ads (they're not comparing tools yet, they're comparing solution approaches).

This mapping informs your channel selection in the next playbook. You'll know which channels to test for which segments, and what awareness stage to expect them at, which determines your messaging and landing page strategy.

Belief splits create segments. When two people at the same awareness stage doubt different assumptions, they need different campaigns to move forward.

Find these splits through customer research interviews. Ask "What almost stopped you from buying?" and listen for patterns in the doubts people mention. Name segments after their doubts (compliance-driven, breach-reactive, proactive) not their demographics (SMB, enterprise, IT buyers).

Validate segments by checking if different segments need significantly different campaigns. If yes, they're real segments. If no, combine them.

Map segments to channels and awareness stages to know where to find each segment and what messaging they need. This mapping becomes your targeting strategy in the next playbook on channel selection.