Conversion Rate / Funnel Calculator
Multi-step funnel: enter user count at each stage. Get step-by-step and overall conversion rates with drop-off analysis. Free, no signup.
Conversion Rate / Funnel Calculator
Enter user counts at each stage of your funnel (visited → viewed → added → checkout → purchased). The tool computes step-by-step conversion rates, drop-off at each stage, the overall top-to- bottom rate, and flags the worst-performing step — the highest- leverage CRO target.
| Stage | Users | Step rate | Drop-off |
|---|
How to Use the Funnel Calculator
Pull counts from your analytics
GA4 Funnel Explorer, Amplitude funnel, Mixpanel funnel, or any tool that gives you unique users per event. Make sure each stage is on the same date range and same population (don't mix daily with weekly counts).
Enter stage labels + counts
Use clear labels — "Visited PDP", "Added to cart", "Checkout started". The tool auto-computes each step's rate and the worst-step flag updates live.
Add or remove stages
Funnels can be 2 stages (landing → purchase) or 10+ (registration → email verify → onboarding → first action → activation). Add as many as you need.
Attack the worst step first
The tool flags the lowest-conversion step. That's almost always where the highest CRO leverage is — fixing a 5% step is easier than improving a 70% step, and either takes the overall rate up by the same factor.
Funnel Analysis — The Foundational CRO Skill
Why Funnels Work and What They Hide
A funnel is the simplest possible model of a multi-step user journey: count users at each step, compute the ratio between consecutive steps, identify where you're losing them. The model traces back to AIDA (Awareness, Interest, Desire, Action) marketing-funnel theory from the 1890s, refined into Dave McClure's AARRR pirate-metrics framework (Acquisition, Activation, Retention, Referral, Revenue) in 2007. Every modern analytics platform — GA4, Amplitude, Mixpanel, Heap, PostHog — ships a funnel-builder view as a core feature because the funnel is the universally readable mental model for product analytics.
What funnels hide: they assume a linear sequence (Step A then B then C), but real user behaviour is often non-linear (users go A → B → A → C, or skip directly from A to D). They assume each step is binary (did or did not), but reality has partial completion. They aggregate across user segments that may have very different behaviour (mobile vs desktop, new vs returning, organic vs paid). The right defence: complement funnels with cohort analysis, segment splits, and session replays (Hotjar, FullStory, LogRocket) for the deeper "why" questions a funnel can't answer.
The Worst-Step Heuristic
CRO 101: attack the worst-performing step first. A funnel with 70% → 70% → 70% → 20% step rates has a 6.9% overall conversion. Doubling the 20% step (to 40%) takes overall to 13.7% — a 99% lift. Doubling one of the 70% steps takes overall to 9.8% — only a 42% lift. The math is structural: the steepest drop is where the biggest absolute user count is being lost, and that's the biggest pool of recoverable users. Most CRO programs spend disproportionate effort on the steps that are already working well (button-color tests on a 70% step) because those tests are easier to run — wrong incentive. Force the program to start with the worst step.
E-commerce benchmark steps (Baymard Institute 2024): cart → checkout average ~70% completion (30% cart abandonment), checkout → purchase ~58%, overall visit → purchase 2-3% (median across all retailers). SaaS signup → trial-start ~50%, trial → paid ~10-20% B2B / 2-5% B2C. Mobile-app install → day-7 retention ~25%. Use these as sanity checks, but real-product specific benchmarks beat industry averages — your top-quartile competitor's funnel matters more than the long-tail average.
"The worst-performing step in a funnel is the highest-ROI CRO target. A 70% step that you might lift to 75% adds 7% to overall conversion; a 20% step you lift to 25% adds 25% to overall conversion. Same effort, very different return."
Statistical Confidence in Funnel Comparisons
When comparing funnels between cohorts (Variant A vs B, mobile vs desktop, this week vs last) you need to think about sample size at the worst step. A funnel with 10,000 top-of-funnel and 200 bottom-of-funnel has plenty of significance at the top stages but high noise at the bottom — a 5% movement in the bottom-stage rate is within sampling noise. Use a sample-size calculator (we have one — RT-FIN-137 in the next tool) to validate that any cohort-vs-cohort comparison has enough volume at the narrowest step to draw conclusions. Funnel comparisons reported without confidence intervals are a common CRO failure mode.
Where to Set the Funnel Boundaries
Funnel definitions are a leadership decision more than a technical one. Where the funnel starts (homepage visit vs paid-ad click vs email click) determines the conversion-rate magnitude — a paid-traffic-only funnel will show much higher rates than an all-traffic funnel because the audience is pre-qualified. Where it ends (first purchase vs paid trial vs activation event) determines whether the metric measures acquisition or product-market-fit. Most companies need at least two funnels: an acquisition funnel (homepage → signup) and a product funnel (signup → activated → retained). Conflating them into one big funnel hides the fact that acquisition and product issues require very different fixes.
10 Facts About Conversion Funnels
Step conversion = users at stage[i+1] ÷ users at stage[i]. Overall = stage[last] ÷ stage[first].
AIDA (Awareness, Interest, Desire, Action) is the 1890s marketing-funnel ancestor.
AARRR pirate metrics (McClure 2007): Acquisition, Activation, Retention, Referral, Revenue.
E-commerce average: 2-3% visit-to-purchase per Baymard Institute 2024.
Cart abandonment averages ~70% across e-commerce — the largest single CRO opportunity.
Mobile vs desktop: desktop typically converts 2-3× higher than mobile for high-AOV purchases.
SaaS trial → paid: 10-20% for B2B, 2-5% for B2C freemium.
Worst-step focus: doubling a 20% step lifts overall 2×; doubling a 70% step lifts only 1.4×.
GA4 Funnel Explorer is the most-used funnel tool by volume; Amplitude + Mixpanel are the product-led-growth standards.
Funnels assume linear sequence; real behaviour is non-linear. Pair with cohort + session replay for the "why."
Frequently Asked Questions
- Step conversion = users at the later stage ÷ users at the prior stage. Overall conversion = users at the last stage ÷ users at the first stage. Step conversions multiply together to give overall: 50% × 40% × 60% × 30% = 3.6% overall. Drop-off at each step = 1 − step conversion.
- The lowest-conversion step. Doubling a 20% step's conversion (to 40%) doubles your overall funnel. Doubling a 70% step (to 100%) only lifts overall by 43%. The worst step usually represents the biggest pool of recoverable users — and small absolute lifts in low-conversion stages produce big relative improvements overall.
- 2-3% visit → purchase across US e-commerce (Baymard Institute 2024). Top decile retailers run 4-7%. Single-brand DTC stores often run 1.5-2.5% (lower than horizontal retailers because the brand draws less curiosity traffic). B2B e-commerce 0.5-1%. Niche-specialty (luxury watches, golf equipment) 4-8%. Always benchmark within your vertical, not against the global average.
- Mobile conversion is structurally lower for high-AOV, complex purchases: smaller screen makes comparison-shopping harder, mobile sessions are shorter (interruptions, multitasking), payment friction is higher (typing credit card on phone). Mobile-first verticals (fashion, beauty, food delivery) have closed the gap. Desktop-first verticals (B2B SaaS, enterprise software, luxury) still see 2-3× higher desktop conversion. Plan content + flow accordingly per device.
- For acquisition funnels (visit → signup → first purchase), use new-user funnels only. Returning users inflate top-of-funnel without contributing meaningfully to conversion math. For repeat-purchase funnels (browse → cart → repurchase), use returning-user cohorts only. Mixing the two corrupts the math. GA4 lets you segment funnels by new vs returning — always check both views.
- Match the window to your user's typical decision cycle. E-commerce: 7-30 days. SaaS B2C trial → paid: 14-30 days. SaaS B2B enterprise: 30-90 days. Mobile app retention: 1-7-28 day windows. Too short = you miss real conversions. Too long = you include unrelated revisits as funnel completions. Most analytics tools default to 30 days; adjust per your product.
- A linear funnel assumes A → B → C → D in order. Real users often jump steps (skip C), backtrack (D → A), or hit multiple paths. Modern analytics (Amplitude, Mixpanel) support "path analysis" or "user journeys" for non-linear flows. For most everyday use, a linear funnel is still the right starting point — even if user behaviour is non-linear, the funnel tells you where the largest leak is. Treat funnel + path analysis as complementary, not substitutes.
- Use the A/B test sample-size calculator (RT-FIN-137) and significance calculator (RT-FIN-138) — both in this directory. Pick the step you want to improve (the worst step is the right target), estimate a realistic lift (rarely above 20% relative), compute required sample size, and run the test long enough to hit it. Don't peek and stop the test early — that destroys the significance math.
- Not directly — this is a calculator for analysing counts you've already pulled from your analytics. Export funnel counts from GA4 Funnel Explorer / Amplitude funnel view / Mixpanel funnel and paste them here for quick step-by-step analysis. Useful for prep before stakeholder meetings, ad-hoc what-if analysis ("what if we doubled the worst step?"), or comparing two funnel snapshots.
- Singapore + Hong Kong: similar to US (2-3% e-com average) — affluent, digitally-mature markets. Indonesia + Philippines + Vietnam: lower (0.5-1.5%) because of higher mobile-traffic share, lower-trust e-commerce environment, and cash-on-delivery preference creating extra friction. Thailand + Malaysia in between. ASEAN funnels also tend to have higher cart-abandonment because COD allows users to "order" non-committally. Local payment methods (GrabPay, ShopeePay, GCash) are now reducing this friction; conversion rates have lifted 30-50% in the past 3 years.
Related News
You may be interested in these recent stories from our newsroom.
-
Snowflake jumps 36 per cent in a day on an earnings beat and a US$6 billion AWS chip deal
Snowflake had its best day as a public company on 28 May, closing up 36 per cent after a clean first-quarter beat and a five-year, US$6 bill...
-
MAS Scraps Mandatory Financial Advice for Most Complex Product Buyers in Retail Shake-Up
Singapore retail investors buying structured notes, derivatives and investment-linked policies will no longer need mandatory financial advic...
-
SEC Rewrites Float Rules, PSE Moves to Implement Them — Clearing the Path for GCash's USD 1B Philippine IPO
The SEC lowered the public float floor for large Philippine issuers in February 2026. The PSE followed with a consultation paper in April. T...
75 more free tools
Calculators, converters, security tools — no signup.