Customer Segment Value Calculator
Which segment actually deserves the budget? Enter each segment’s reachable size, conversion rate, value, and acquisition cost. This ranks them by modeled value — not headcount — checks LTV:CAC on each, and flags the “segments” that are really just labels because they respond like everyone else.
A customer segment is worth size × conversion × value, not the number of people in it. This calculator models the value of every segment, ranks them, checks each one’s LTV:CAC against the 3:1 bar, and runs the responsiveness test that separates a real segment (one that behaves differently from your blended audience) from a decorative one. It ends with a single answer: the segment to focus on next, and the value personalizing to it could add.
Enter the segments you want to compare
Edit any cell. The defaults are illustrative — four segments an ecommerce brand might argue over — and they are set up to teach: one converts like everyone else, and two spend more than they return. Change them to your own numbers. Use lifetime value in the value column if you have it, or average order value if you don’t.
| Segment | Reachable size | Conversion % | Avg value $ | CAC $ |
|---|---|---|---|---|
Reachable size is the number of people, sessions, or accounts you can actually put in front of this segment’s offer — not your whole database. Conversion is the share who buy. Value is what an acquired customer is worth (LTV if known, else AOV). CAC is what it costs to acquire one.
Segment ranking and result
| # | Segment | Modeled value | LTV:CAC | Responsiveness |
|---|
How to use this calculator
- Name your segments by behavior, not by demographic.“High-intent returning” and “discount-driven” describe what people do. “Women 25–34” describes who they are and rarely predicts what they buy. Start from behavior and need.
- Enter reachable size, conversion, value, and CAC.Be honest about reachable size — the people you can truly put this offer in front of, not your whole list. Use lifetime value in the value column when you have it; average order value is a fine stand-in.
- Read the responsiveness flags.The tool computes your size-weighted blended conversion and flags any segment that converts within 10% of it as decorative. A decorative segment behaves like everyone else, so treating it specially wastes effort.
- Check LTV:CAC on every segment.Anything below 3:1 is a red flag: you’re paying more to acquire than the customer returns. Green segments earn their budget; red ones need a higher value or a lower cost before you scale them.
- Set the personalization lift, then export.Slide the expected lift, read the ranked list and the single segment to focus on, then copy a share link, download the CSV, or print a one-page PDF for the planning meeting.
RGM Expert Says
The first thing we do with a segmentation deck is throw out the headcounts. A slide that ranks “audiences” by how many people they contain tells you nothing about where growth comes from. So we run every segment through the same three numbers — how many you can reach, how many convert, and what each is worth — and let modeled value do the ranking. It is common for the biggest box on the slide to fall to the bottom of the list, and for a small, unglamorous segment of returning buyers to sit at the top. That reordering is usually the whole insight.
Then comes the test almost no deck survives: the responsiveness check. Kotler and Keller are blunt that a usable segment has to be actionable — it must respond differently from the rest of the audience, or there is nothing to act on. So we compare each segment’s conversion to the blended rate. When a “segment” converts within a few points of everyone else, it is decorative: a label someone drew around a group that behaves exactly like the average. Personalizing to it, building creative for it, buying media against it — all of that spends money to reach people who were never distinct. Cutting decorative segments is often the fastest way to free up budget.
The last discipline is to segment by behavior and need, not by who people are. Yankelovich and Meer made this case in the Harvard Business Review two decades ago, and it has only gotten truer: what someone recently bought, how often, and for how much predicts the next purchase far better than age or postcode. When a client insists on a demographic cut, we don’t argue — we model both and let the value ranking and the responsiveness flags settle it. Nine times out of ten the behavioral segments are the ones that respond differently and pay back above 3:1, and the demographic ones light up decorative.
How it works
The tool runs three calculations on every segment, then ranks them. None of it is complicated — the value is in doing it consistently, on the same page, instead of arguing from headcounts. First, the modeled value of a segment, which folds reach, conversion, and worth into one comparable number:
where N is reachable size, CVR is the conversion rate, and V is the average order value or lifetime value of an acquired customer. Next, the unit-economics check — the ratio of what a customer is worth to what it costs to acquire one:
Below 3:1 the segment is flagged red. Finally, the responsiveness test that decides whether a segment is real. We compute the size-weighted blended conversion across all segments — how your average reachable person behaves — then measure how far each segment sits from it:
- CVR̄blended — the size-weighted average conversion across every segment, i.e. total buyers ÷ total reachable people. It is how the “average” person in your audience converts.
- ≥ 0.10 — a segment must convert at least 10% above or below the blended rate to pass. Inside that band it behaves like everyone else and is flagged decorative — a label, not a lever.
- Focus rule — the recommended segment is the highest modeled value among those that are both real and healthy (LTV:CAC ≥ 3:1). Value and viability, not size.
Modeled value and LTV:CAC are standard growth arithmetic. The 10% responsiveness band is RGM’s own rule-of-thumb operationalization of the classic actionability and differentiated-response criteria for a usable segment (Kotler & Keller, Marketing Management) — treat the threshold as a sensible default, not a law of nature.
Most segmentations never create value — because they never test whether the segment is real
Segmentation is one of the most abused ideas in marketing. Daniel Yankelovich, who introduced the concept to business in 1964, returned to it four decades later with David Meer in the Harvard Business Review to warn that the practice had drifted into “a bewildering array of psychographic and attitudinal” cuts that “lack the predictive power of segmentations based on behavior” (Yankelovich & Meer, HBR, 2006). Their estimate was stark: only a small share of segmentations — on the order of one in seven — actually created value for the businesses that commissioned them. The rest were expensive slides.
The failure mode is almost always the same. A team splits the audience by something easy to observe — age, gender, region, a survey attitude — and never checks whether the resulting groups behave differently. This is exactly the trap the responsiveness test catches. Philip Kotler and Kevin Keller list the criteria for a segment worth having: it must be measurable, substantial, accessible, differentiable, and actionable. Differentiable and actionable both mean the same practical thing — the segment has to respond to a distinct offer. A group that converts at your blended rate fails that test no matter how neat the demographic story is.
What does predict behavior is prior behavior. RFM analysis — ranking customers by recency, frequency, and monetary value, popularized by Arthur Hughes in Strategic Database Marketing (1994) — routinely out-predicts demographic cuts because it segments on what people actually did. That is why this calculator asks for conversion, value, and cost per segment rather than a persona’s age: it forces the conversation onto the numbers that move growth. Rank by modeled value, keep only the segments that respond differently and pay back above 3:1, and you have done in one page what most six-figure segmentation projects never manage — separated the real segments from the decorative ones.
Rules of thumb for reading the numbers
Use these as sanity checks, not targets. They come from widely cited public sources; your own data always wins where you have it.
| Signal | Rule of thumb | Read it as |
|---|---|---|
| LTV:CAC by segment | ≥ 3:1 healthy | Below 3:1 is a red flag |
| CAC payback period | < 12 months | Longer strains cash flow |
| Responsiveness vs. blended rate | ≥ ±10% | Inside the band = decorative |
| Share of segmentations that create value | ~14% | Most never earn their cost |
| Segmented email — click rate lift | +100.95% | Behavior-based targeting pays |
What the segmentation field says
“There is only one winning strategy. It is to carefully define the target market and direct a superior offering to that target market.”
“The aim of marketing is to know and understand the customer so well the product or service fits him and sells itself.”
Attitudinal and demographic segments “lack the predictive power of segmentations based on behavior” — the reason to model what people buy, not who they are.