Finance & Analysis

Verified against ChatGPT · 2026-08-14

Calculate LTV:CAC and stress-test whether the inputs feeding it are actually honest

Computes customer lifetime value against customer acquisition cost and, more importantly, interrogates whether the underlying churn, margin, and cost inputs are being measured in a way that would survive scrutiny, rather than accepting optimistic assumptions at face value.

ChatGPT (GPT-5.1)4 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

Calculate the LTV:CAC ratio from the inputs below, and — this is the more important part — check whether the inputs themselves would hold up if someone skeptical looked at how they were derived. A lot of LTV:CAC numbers that circulate are technically calculated correctly from inputs that were quietly optimistic to begin with.

AVERAGE REVENUE PER CUSTOMER
$120/month average across the customer base.

GROSS MARGIN
72% gross margin.

CHURN RATE / AVERAGE CUSTOMER LIFESPAN
4% monthly churn, blended across the whole customer base.

CUSTOMER ACQUISITION COST AND HOW IT WAS CALCULATED
$340 per customer, calculated as total paid ad spend for the quarter divided by new customers acquired that quarter — doesn't include sales team salaries or marketing software costs.

STEP 1 — CALCULATE
Show LTV as (average revenue per customer x gross margin) / churn rate, or the equivalent lifespan-based formula if churn rate wasn't given directly, with the formula and inputs shown. Show the LTV:CAC ratio.

STEP 2 — INTERROGATE THE CHURN INPUT
If churn rate was given as a blended average across all customers, flag that blended churn can hide a bimodal reality — a cohort that mostly churns fast in month one and a cohort that stays for years produces the same average as a cohort that churns steadily, but implies very different LTV — and note this as a reason the LTV figure could be less reliable than it looks if cohort-level churn data exists and hasn't been checked.

STEP 3 — INTERROGATE THE CAC INPUT
Check what's included in the stated CAC calculation. If it only includes paid ad spend and excludes sales team salaries, commissions, marketing headcount, or tools, flag this explicitly as CAC likely being understated, since a narrowly-scoped CAC produces an artificially favorable ratio. State what a fully-loaded CAC would need to include to be considered complete.

STEP 4 — HONEST RATIO
Give the ratio as calculated from the inputs provided, then separately state how the ratio would likely shift (directionally, not with a fabricated precise number) if the flagged issues in steps 2 and 3 were corrected — do not invent a specific corrected number without the real data to support it; describe the direction and rough magnitude of the likely correction instead.

OUTPUT FORMAT
1. LTV calculation with formula shown, CAC as given, and the resulting ratio.
2. Churn input reliability flag.
3. CAC completeness flag, listing what's likely missing.
4. Directional honest assessment of which way the true ratio likely moves once those gaps are addressed.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

The reason this prompt is built around interrogating the inputs rather than just computing the ratio is that LTV:CAC is one of the most commonly gamed metrics in business — not usually through outright fabrication, but through optimistic scoping decisions (a CAC that excludes sales salaries, a churn rate that's a blended average hiding a bad early-cohort problem) that are each individually defensible but compound into a ratio that looks much healthier than the real unit economics support. Flagging blended churn specifically is important because a blended average is mathematically identical whether it comes from a customer base that churns steadily or one that's bimodal — mostly gone in month one, with a smaller group that stays for years — and those two underlying realities imply very different true LTV even though they produce the same average churn number, which is exactly the kind of thing a single input figure can't reveal on its own. Requiring the model to state what a fully-loaded CAC should include, rather than accepting a narrowly-scoped figure silently, catches the single most common way this metric gets flattered: ad spend alone is usually a fraction of true acquisition cost once sales headcount, commissions, and tooling are counted, and a ratio built on ad-spend-only CAC systematically overstates how efficiently the business is actually acquiring customers. The instruction to describe the likely direction and rough magnitude of correction rather than inventing a specific corrected number is a deliberate honesty constraint — GPT-5.1 could easily generate a plausible-looking "corrected" ratio, but without the actual cohort churn data or fully-loaded cost figures, any specific number would be a fabrication dressed up as a more rigorous calculation, which is worse than clearly stating the direction of the likely correction and leaving the precise figure to be calculated once the real data is gathered.

What you get back

LTV = ($120 x 0.72) / 0.04 = $2,160. LTV:CAC = $2,160 / $340 = 6.35:1. Churn flag: 4% is stated as a blended average — if a large early-cohort churn spike exists and hasn't been checked, true LTV for the retained cohort could differ substantially from this blended figure. CAC flag: excludes sales salaries and marketing tooling — a fully-loaded CAC would likely be meaningfully higher, which would pull the ratio down from 6.35:1 toward a less favorable, though still uncalculated without the real cost data, number.

Verified against

ChatGPT GPT-5.1 · 2026-08-14

Changelog

  • 2026-08-14 Initial publish, verified against ChatGPT GPT-5.1.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
All Finance & Analysis prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY