Segmenting
HOW DO YOU SET PRICES WITH CONFIDENCE?
At a glance:
Some customers happily pay more for speed, certainty, or white-glove service. Others only show up when there’s a deal. The point? One price almost never fits all.
WHAT IT IS
Segmentation is how you stop averaging everyone into the same number and start matching price to value and cost-to-serve. That means charging different prices to different customers in different contexts on purpose. In practice, that means picking an axis (customer type, volume, channel, time, product version), setting clear rules, and building fences (eligibility conditions) so customers self-select into the tier that fits them.
WHY IT MATTERS
Captures more margin without losing a thrifty tail.
Aligns price with value and cost.
Creates strategic choice instead of one blunt number (so you can serve different demos without one group cannibalizing another).
Case File — The New York Times: Standalone → Bundle = Clean, Self-Selected Segments
“[Our] bundle…provides the most value to our users and represents the best opportunity to monetize our digital products.”
Setup. NYT sells multiple “jobs” (news, games, cooking, sports), and those audiences don’t share one willingness-to-pay. The publication has a longstanding need to attract the widest number of subscribers and create multiple reasons to engage every day.
Move. Enter segmentation by product version (standalone vs. bundle) plus a second fence: household sharing. In September 2025, NYT launched family subscriptions. This is segmentation without personalization: customers self-select into higher price points by choosing broader access or more seats.
Outcome. By Q2 2025, NYT >50% of customers had migrated into higher-value segments. In Q3 2025, NYT added ~460,000 digital-only net subscribers and surpassed 12.3 million total subscribers. Subscription revenue rose 9.1% YoY to $494.6M.
Lessons.
Segment with obvious fences you can explain in one sentence – standalone for breadth, bundle for depth, family for households.
Use a “starter” offer to maximize top-of-funnel, and a clearly higher-value bundle to monetize the habit-formers. It saves negotiating price one customer at a time.
WHY SEGMENT? AND HOW DO YOU DO IT?
-
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.
OPERATOR CHECKLIST
◻️ We’ve picked one primary segmentation axis (customer type, volume, channel, time, or version) that maps to real value or cost differences.
◻️ Each discounted tier has a clear fence that’s observable, verifiable, and tied to lower cost-to-serve or lower value.
◻️ Our price book/ERP enforces fences; discounts aren’t living in inbox threads and side deals.
◻️ We can see revenue, GM, and pocket price by segment in a basic report or pivot.
◻️ Exceptions/overrides are logged and reviewed monthly, not shrugged off.
◻️ We have a short fairness story for each fence (“why this tier exists and who it’s for”).
◻️ For reseller business, we’ve run a Robinson-Patman sanity check and documented that similarly situated customers can access similar deals.
◻️ Any dynamic or time-based pricing we use is published ahead of time (not a surprise on the invoice).
SIGNAL TO WATCH
If your pocket prices are all over the place for similar work and nobody can explain why, you’re already segmenting, just badly and informally.
ONE QUICK ACTION
Pick one axis and sketch two tiers with a simple fence: “If you [meet condition], you get [price and service level]. Otherwise, you pay [standard price].” Sense-check it for fairness and legality, then pilot on a small slice.
COMMON TRAPS
Too many axes, leading to contradictions and firefighting.
Soft fences training the market to wait for deals.
Lower prices for segments that don’t actually cost less to serve.
Ad-hoc deals based on shadow segmentation and perceived WRP.
Ignoring legal blind spots (hello, Robinson-Patman enforcement).
EXPERIMENT 1 — HEAT MAP: SEGMENTS
Use when you need…
Clarity: Turns “we should segment” into a prioritized shortlist.
Speed: Requires minimal data to surface obvious opportunities.
Strategic insight: Forces you to articulate the why behind margin differences.
What it’s for: Identify segment clusters where value delivered or cost-to-serve differs enough that margin looks “off,” so you know where segmentation will pay back.
Who it’s for: Pricing/Finance + Sales/Ops leaders who can pull basic segment-level revenue/margin and want a fast shortlist of segmentation targets.
What it does: Helps you spot the mismatch: where you’re charging “average pricing” for customers who are expensive to serve or less price-sensitive, and where you’re undercharging for customers who get outsized value.
EXPERIMENT 2 — DESIGN CANVAS: FENCES
Use when you need…
Clarity: Converts segmentation into a rule you can defend.
Speed: Prevents weeks of debate by forcing the minimum required fields.
Strategic insight: Fairness story reveals whether your segmentation is sustainable.
What it’s for: Design a segment price difference that will actually stick by defining the qualifying rule (fence), rationale, service difference, fairness story, and where it lives in systems.
Who it’s for: Pricing owner + Sales/Ops + Finance (and Legal if needed) who must implement segmentation without creating chaos or backlash.
What it does: Makes segmentation real: a fence that customers can understand, employees can apply, and systems can enforce – with a clear fairness narrative.
EXPERIMENT 3 — PILOT: 90-DAY SEGMENTATION
Use when you need…
Clarity: Turns segmentation into a controlled experiment, not a policy argument.
Speed: A 90-day timebox forces shipping and learning.
Strategic insight: Reveals whether the fence is enforceable and whether customers accept the fairness story.
What it’s for: Run a controlled segmentation experiment on one axis to validate that the new fence improves GM% and price realization without creating leakage, chaos, or fairness blowback.
Who it’s for: A pricing owner with Sales/Ops/Finance support who can implement a limited test slice in pricing systems.
What it does: Makes segmentation testable: define control vs test, implement a limited change, track weekly, then decide to roll out, revise, or stop.
This document is provided solely for general informational and discussion purposes and is intended to offer insight into certain philosophies, principles, and approaches that Permanent Equity may employ when partnering with the management teams of its portfolio companies following an investment. The material is illustrative in nature only and does not describe a required, uniform, or exhaustive set of practices or expectations.
This document is not intended to be, and should not be construed as, legal, tax, accounting, investment, or other professional advice, nor as creating, modifying, or implying any legal rights, obligations, fiduciary duties, standards of conduct, or contractual commitments. Nothing herein should be relied upon as a substitute for individualized advice or negotiated agreements.
This document does not constitute, and should not be construed as, an offer to sell or a solicitation of an offer to purchase any securities, investment vehicles, or advisory services. Any such offer or solicitation may be made only pursuant to definitive offering documents and governing agreements, which will contain material information, risk factors, and binding terms that may differ materially from the general concepts described herein.
The practices, decision making frameworks, expectations, and examples referenced in this document are intended solely to illustrate how Permanent Equity may work with certain portfolio companies in certain circumstances. Actual practices vary based on the specific facts, negotiated contractual arrangements, regulatory considerations, industry dynamics, and stage of the business, and may change over time. No representation is made that any approach described herein will apply to any particular company, management team, transaction, or situation.
Nothing in this document should be interpreted as a commitment, promise, guarantee, or assurance regarding autonomy, governance rights, decision making authority, performance outcomes, distributions, returns, or operational results. Past experience, illustrative examples, and general statements of philosophy are not indicative of future results.
Readers should consult their own legal, tax, financial, and other professional advisers regarding the applicability of any concepts described herein to their specific circumstances.
Permanent Equity undertakes no obligation to update or revise this document, and the material may be modified or withdrawn at any time without notice.