Marketing

Sales Funnel Calculator

Model the whole funnel from impression to customer, and see why stage improvements COMPOUND — ten per cent at three stages is not thirty per cent, it is thirty-three.

Sales Funnel Calculator

Results recalculate instantly on every keystroke. Nothing you type is transmitted.

Funnel
Model
Economics
$
$
End-to-End Conversion
the stages multiply. Improvements compound, they do not add.
Stage 1 — Impression to Click
Stage 2 — Click to Lead
Stage 3 — Lead to Customer
Impressions Per Customer
Customers From One Stage Improving
Customers From All Three Improving
The Compounding Bonus

What this result does not account for

  • Period-based; long sales cycles require cohort analysis.
  • Assumes stage rates hold as volume scales, which they rarely do.
  • Three stages only — real funnels may have more.
Zero-Server Execution Updated 11 Aug 2026 Reviewed by Imran S. Qureshi, CFA IEEE-754 Double Precision

In short: Stage gains compound. 1,000,000 impressions become 518 customers at 0.0518% end-to-end; a 10% lift at all three stages yields 689.5 customers, not the 673.4 that adding would suggest.

Formula

end-to-end = stage₁ × stage₂ × stage₃

three 10% lifts = 1.10³ = 33.1%, not 30%

[('multiplicative', 'each stage acts on the survivors'), ('relative lift', 'same customer gain at any stage'), ('compounding', 'simultaneous lifts exceed their sum'), ('cheapest stage', 'the right target, not the narrowest')]

Worked Example

  1. Compute each stage against the stage before it.
  2. Multiply them for the end-to-end rate.
  3. Model a relative lift at one stage.
  4. Model the same lift at every stage.
  5. The difference is the compounding bonus — it is free.

1,000,000 impressions produce 18,500 clicks (1.8500%), 2,220 leads (12.0000%) and 518 customers (23.3333%) — an end-to-end rate of 0.0518%, or 1,931 impressions per customer. A 10% relative lift at any single stage gives 569.8 customers. The same lift at all three gives 689.5, a 33.1000% improvement rather than the 30% that adding suggests. The 16.1 extra customers are worth 866.62 of contribution and cost nothing — they arise from the arithmetic itself.

Strengths & Limits Of This Model

Where this engine is strong

  • Proves the stages multiply to the end-to-end rate
  • Quantifies the compounding bonus explicitly
  • Shows a relative lift is worth the same at any stage

Where it stops

  • Fixed three stages
  • Constant-rate assumption

Risk & accuracy notice. Averaging stage conversion rates rather than multiplying them overstates funnel performance by a wide margin, and planning against the averaged figure produces volume forecasts that cannot be met.

Practical Use Cases

Modelling a growth plan

Testing whether stage targets reach the customer goal.

Prioritising work

Finding the cheapest stage rather than the narrowest.

Explaining compounding

Showing why simultaneous small gains beat one big one.

Forecasting from impressions

Converting media volume into customers.

Diagnosing a funnel

Locating where the volume actually disappears.

Methodology & Editorial Standards

Each stage is computed against the stage immediately preceding it and the stages are multiplied to reproduce the end-to-end rate exactly. The funnel is validated to narrow at every stage. A relative lift is modelled at one stage and at all three, and the difference between the compounded result and the additive expectation is reported as its own figure, because that gap is the practical argument for broad incremental improvement over concentrated effort.

Computation runs in IEEE-754 double precision at full internal precision; rounding to two decimal places occurs strictly at the display layer, so no cumulative drift enters the result. All monetary outputs use accounting presentation — grouped thousands, two decimals, negatives in parentheses — so figures can be transcribed directly into a model or working paper. Division-by-zero and out-of-domain inputs return an em-dash rather than a misleading number.

This engine was reconciled against an independent reference implementation and hand-verified for the worked example above before release. Our full five-stage review process is published on the About Us page.

Imran S. Qureshi, CFA Head of Quantitative Modelling · ApexConverter

Performance-marketing unit economics and contribution-margin analysis. Last reviewed: 11 August 2026.

Disclaimer. This calculator is provided for informational and modelling purposes only and does not constitute financial, tax, legal, medical, or engineering advice. Verify all figures with a qualified professional before acting on them.


Sales Funnel Calculator — 10 Expert FAQs

10 analyst-written answers to the questions practitioners actually ask — optimised for voice and answer-engine retrieval.

How do you calculate a sales funnel conversion rate?

Compute each stage against the stage immediately before it, then multiply the stage rates together. The product is your end-to-end rate. Averaging stage rates instead of multiplying them is the single most common funnel error and it flatters the result enormously.

Why do funnel improvements compound?

Because each stage operates on the output of the last. Ten per cent more clicks feeds ten per cent more leads, which feeds ten per cent more customers — so three ten per cent lifts produce a 33.1 per cent improvement, not 30.

Which stage should I improve first?

The cheapest one to move. A relative improvement produces identical additional customers at any stage, so the decision is entirely about cost and feasibility. Top-of-funnel creative is usually cheapest, sales capability usually dearest.

Why does the narrowest stage not matter most?

Because you improve stages relatively, not absolutely. Taking a 1.85 per cent click rate to 2.04 per cent and a 23 per cent close rate to 25.7 per cent both represent ten per cent relative gains, and both produce exactly the same number of extra customers.

What if a stage widens instead of narrowing?

Then volume is entering mid-funnel from a source the model does not see — typically direct traffic, sales-sourced deals or an untracked channel. Track those separately rather than folding them in, or the funnel arithmetic breaks.

How many stages should a funnel have?

As many as you can measure reliably and act on separately. More stages give better diagnosis and more opportunities for measurement error, and a stage you cannot influence is a reporting line rather than a lever.

Does the funnel model work for long sales cycles?

Only with cohort analysis. In a period-based model with a long cycle, this month's customers came from earlier months' leads, so the stage rates are mismatched and the end-to-end rate is understated whenever volume is growing.

Should I model absolute or relative improvements?

Relative, because it makes stages comparable. An absolute one-point gain is trivial at a 23 per cent close rate and impossible at a 1.85 per cent click rate, so absolute targets make cross-stage comparison meaningless.

How does spend enter the funnel model?

At the top, and it sets the impression volume everything else scales from. Doubling spend roughly doubles customers at constant rates — but rates rarely stay constant as you scale, because incremental audiences convert worse.

What is the practical takeaway?

Small simultaneous improvements across every stage outperform one heroic effort at a single stage, and the difference is free — it comes from the multiplicative structure rather than from additional work.

Related Marketing Engines