Run A/B tests and personalize your store without code. 14-day free trial, no credit card.
Start free trial →In-depth guides and case studies where this concept is put to work.
A conversion funnel is a model of the stages a visitor passes through on the way to buying, drawn as a funnel because each stage contains fewer people than the one before it. A typical ecommerce funnel runs: site visit → product page view → add to cart → checkout started → payment completed.
The shape is the point. If 10,000 people visit, 4,000 view a product, 900 add to cart, 500 begin checkout, and 260 pay, you can immediately see that the steepest proportional loss is between product view and add to cart. That is where attention belongs, regardless of which stage feels most broken.
Sitewide conversion rate tells you that something is wrong. A funnel tells you where. Without one, optimization work tends to drift toward whichever page the team looks at most often, which is rarely the page losing the most revenue.
Funnels also correct a persistent misjudgement about effort. Redesigning a homepage feels significant and takes weeks; removing a mandatory phone-number field from checkout takes an afternoon. If the funnel shows 45% of visitors abandoning at checkout and 8% bouncing from the homepage, the afternoon's work is worth more than the redesign.
The economics reinforce this. A visitor who has reached checkout has already survived every earlier filter — they want the product, they accept the price, they trust you enough to enter an address. Recovering 10% of those people is worth several times more than recovering 10% of homepage bouncers, because the remaining barrier is usually small and mechanical.
A packaged foods brand mapped their funnel and found an unusual pattern: add-to-cart rate was healthy at 11%, but only 38% of people who started checkout completed it — well below the 55–65% they should have expected.
Session recordings of the checkout step showed the failure. The cash-on-delivery option was present but rendered below the fold on mobile, beneath three prepaid options, with no indication it existed. A large share of their tier-2 city customers, who strongly prefer COD, were reaching the payment step, not finding the option they wanted, and leaving.
Moving COD to the top of the payment list for visitors outside metro pincodes lifted checkout completion from 38% to 51%. The funnel found the stage; the recordings explained it.
The funnel does two jobs in an experimentation programme. It decides what to test — the largest drop-off is your test queue's first entry — and it explains what happened after the test runs.
Tracking the whole funnel during a test, rather than only the final conversion, catches problems that a single metric misses. A variant that lifts add-to-cart by 15% but reduces checkout completion by 20% is pushing under-committed shoppers forward to fail later. That nets out negative, and you would only see it by watching every stage. CustomFit.ai reports the full funnel for each variant so shifts between stages are visible rather than inferred.
Run smarter A/B tests with CustomFit.ai — 14-day free trial, no credit card required.