Most B2B segmentation is either too vague ("SMB / mid-market / enterprise") or too detailed (12 personas nobody can remember). The version that actually drives action is 3-5 segments defined across 4 axes. Claude can produce this in 20 minutes if you brief it with real customer data. Here is the workflow.
Generic segmentation by company size or industry has stopped being useful for most B2B businesses. Within "mid-market SaaS" you have wildly different buyers with different needs.
The version that works defines segments along 4 axes: who they are (firmographics), what they care about (priorities), how they buy (process), and what success looks like for them (outcomes). 3-5 segments along these axes drives every downstream decision.
Axis 1: Firmographics. Size, industry, geography, stage, tech stack patterns.
Axis 2: Priorities. What is keeping the buyer up at night this quarter.
Axis 3: Buying process. Who is involved, how long it takes, what proof they need.
Axis 4: Success definition. What does "this purchase worked" look like at 6 and 12 months.
I am segmenting our customer base. Here is our customer list with key data: [PASTE — firmographics, ARR, retention status, expansion pattern, win/loss reason] Here is what we sell: [PRODUCT SHORT] Here is our current ICP definition: [LOADED IN PROJECT] Cluster our customers into 3-5 distinct segments using the 4-axis model: 1. Firmographic profile 2. Top 3 priorities for this segment 3. Typical buying process (committee size, cycle time, proof required) 4. Success definition (what does this customer say if asked "did the purchase work" at 12 months) For each segment, also identify: - The single message that resonates most with this segment - The objection this segment brings up most frequently - Whether we should invest more, hold steady, or de-prioritize this segment Be willing to recommend de-prioritizing segments. Not every customer profile is good for our future business.
1. Too many segments. If you have 10 segments, you have zero. Force consolidation to 3-5.
2. Segments based on what we sell, not who they are. "Enterprise customers" is not a segment. The segment is the buyer profile.
3. No willingness to de-prioritize. Real segmentation means saying "we will not pursue X." Most companies refuse this and end up unfocused.
4. Segments that look the same in messaging. If two segments get the same emails, they are not really two segments — they are one segment with different labels.