Segmenting

If different customers value you differently, a single price underperforms. Segment-based pricing that tailors prices to needs and sensitivities improves revenue and profit, especially for new or differentiated offerings. 

A simple pattern (not a law):

• A minority of customers would still buy at meaningfully higher prices. 

• A middle band is more price-sensitive but stable.

• A tail only buys on deals or off-peak.

Segmentation is about moving the first group up, serving the middle fairly, and deciding what you want to do about the tail. 

Takeaway: Assume your demand is varied. Use segmentation to bring price closer to value and cost-to-serve (not to squeeze everyone)

Common operator-friendly segmentation axes include:

• Customer traits (who they are): student/senior discounts, non-profit rates, SMB vs. enterprise tiers, resident vs. tourist pricing. 

• Usage or use case (what they’re doing): commercial vs. personal use, internal vs. client-facing, low-risk vs. mission-critical.

• Volume/commitment: volume tiers, case-pack discounts, annual contracts, multi-year deals.

• Channel: direct vs. distributor vs. outlet store, reflecting different commission structures, service levels, and assortment.

• Time/demand state: peak vs. off-peak, early-bird vs. last minute, weekday vs. weekend.

(You don’t need all of these – too many overlapping axes just confuses everyone. Start with one that tracks a clear value or cost difference you can explain in a sentence.)

Takeaway: Pick one primary axis where value or cost truly differs and build around it first.

Once you pick an axis, you need fences – the practical rules that decide who gets which price. 

• Customer fences: IDs, membership, affiliation (student email, non-profit certificate, loyalty tier).

• Product fences: different versions (features, warranty, speed, flexibility).

• Purchase fences: when, where, and how the purchase happens (advance-purchase rules, channel, minimum quantity, day of week).

Good fences are observable and verifiable, cost-offsetting, hard to resell, and system-enforced.

Takeaway: A segment without a clear fence is a wish, not a pricing strategy. Design fences that your systems and customers can actually live with.

Segmentation can be seller-selected (you decide who gets what price) or self-selected (customers choose their own tier based on your rules). Most operators opt for self-selection because it scales with less sales judgement, it’s easier to explain, and it usually raises fewer fairness and legal concerns (assuming the rules are published and consistently applied). 

Takeaway: Wherever possible, let customers pick their tier by stepping into a fenced offer.

In the U.S., the Robinson-Patman Act restricts giving different net prices or promotional allowances on like-grade goods to competing resellers without cost or competitive justification. Enforcement was sleepy for decades, but a recent lawsuit from the FTC signals renewed attention.

Practical implications:

• If you sell to resellers, offer discounts on equal terms (volume, timing, geography) and document why differences exist (cost savings, competition).

• Avoid segmentation that tracks protected characteristics (race, gender, etc.) or uses opaque personalization that might proxy for them. Focus on behavior and self-selected plans.

• In B2C, the bigger risk is optics: hidden “junk fees,” extreme surge pricing, and unfair-seeming differences. Transparent rules (“weekday bookings 15% off”) travel better than per-person surprises.

Takeaway: Segment by value, cost, and clear rules, then put the rules in writing. If you’d be uncomfortable seeing your fences on the front page, rethink them.

Segmentation isn’t done when the price book changes. Practical metrics include:

• Are higher-value segments contributing more? Do low-price segments cover their serve-cost?

• Are customers gradually moving into the tier that matches their visage and value?

• How often do discounts meant for one segment leak into another via overrides or exceptions?

• Do surveys or customer support tags show resentment?

Takeaway: Build a simple dashboard by segment so you can see revenue, margin, and leakage per fence.