Ecommerce Funnel: The Metrics That Tell You What to Fix Next

An ecommerce funnel shows how shopper sessions move from a store visit to a completed checkout. Use one measurement basis, compare each stage with a defined baseline, and prioritize the leak with the greatest estimated revenue at risk.

Do not choose a fix only because it has the largest percentage drop. A small drop at a high-volume stage can be worth more than a large drop later in the journey. Your calculation must account for the number of sessions at the stage, the expected stage conversion rate, the expected downstream conversion rate, and average order value.

What an ecommerce funnel measures

For a Shopify store, use sessions as the measurement basis throughout the funnel. Do not mix session counts with unique visitors, page views, add-to-cart events, or order counts. Those measures answer different questions and produce different conversion rates.

Shopify’s Conversion rate breakdown report uses four session-based stages: all sessions, sessions with cart additions, sessions that reached checkout, and sessions that completed checkout. Shopify defines a checkout-reached session as one with user input during checkout, such as a key press or mouse click.

Use a closed, session-based ecommerce funnel

Use Shopify’s closed funnel view for diagnosis. In a closed funnel, a session must add to cart, then reach checkout, then complete checkout in that sequence. This prevents a shopper who enters directly at checkout from inflating a later stage.

All sessions: every session included in the selected date range.
Sessions with cart additions: sessions in which a shopper added at least one product to cart.
Sessions that reached checkout: sessions in which a shopper progressed to checkout and interacted there.
Sessions that completed checkout: sessions in which a shopper completed a purchase.

Use the same date range, store, funnel type, and session definitions for every comparison. Shopify explains that its closed funnel only counts subsequent steps when the session completes the preceding steps in order.

Calculate the stage rates that expose a leak

Record the population at each stage. Then calculate the progression rate between adjacent stages. Cart-entry rate equals sessions with cart additions divided by all sessions. Checkout-entry rate equals sessions that reached checkout divided by sessions with cart additions. Checkout-completion rate equals sessions that completed checkout divided by sessions that reached checkout.

A large stage drop is a useful investigation signal. It does not prove that the stage is your largest revenue opportunity. Traffic volume, product mix, average order value, and the likelihood that a recovered session will complete the later steps all affect the value of a fix.

Set a defensible baseline before estimating lost revenue

Choose a baseline that represents normal or desired performance for the same stage. A practical baseline can be the prior comparable period, a stable period before a site change, or a higher-performing segment with similar traffic intent and product mix. Label the baseline clearly before doing the math.

For each stage, define four inputs: the number of sessions reaching the prior stage, the observed progression rate, the baseline progression rate, and the baseline downstream completion rate. Use the average order value from that same baseline and segment, rather than a storewide value that may reflect a different product mix.

Revenue-at-risk formula for a funnel stage

Estimate incremental stage entries first: prior-stage sessions × (baseline stage rate minus observed stage rate). Use zero if the observed rate is at or above baseline. Then estimate incremental completed checkouts: incremental stage entries × baseline downstream completion rate. Finally, estimate revenue at risk: incremental completed checkouts × baseline average order value.

Revenue at risk = prior-stage sessions × max(0, baseline stage rate − observed stage rate) × baseline downstream completion rate × baseline average order value.

This is an estimate of the revenue associated with returning a stage to its stated baseline. It is not a promise of revenue. The calculation makes its counterfactual explicit: the affected sessions progress at the baseline stage rate and convert through the remaining stages at the baseline downstream rate.

Example: prioritize a cart-to-checkout problem

Suppose 1,000 closed-funnel sessions added a product to cart. The observed checkout-entry rate is 40%. A comparable baseline shows a 50% checkout-entry rate. The baseline rate from reaching checkout to completing checkout is 60%, and baseline average order value is $80.

Incremental checkout sessions: 1,000 × (50% − 40%) = 100.
Incremental completed checkouts: 100 × 60% = 60.
Estimated revenue at risk: 60 × $80 = $4,800.

Run the same calculation for the cart-entry and checkout-completion stages. Rank the positive estimates, then investigate the highest-value opportunity first. Keep the assumptions beside the number so your team can challenge the baseline rather than treating the estimate as a fact.

A practical Shopify measurement workflow

Open Shopify admin and go to Analytics, then Reports. Filter for Behavior reports and open Conversion rate breakdown.
Select one complete date range. Avoid comparing a sale period with a non-sale period unless the sale is the change you are investigating.
Use the closed funnel view and record the four session populations: all sessions, cart additions, reached checkout, and completed checkout.
Calculate adjacent-stage progression rates and compare them with your defined baseline.
Calculate estimated revenue at risk for every stage using the same baseline logic.
Segment the analysis before changing the site. Check whether the issue is concentrated by device, traffic source, landing page, campaign, geography, or product group.
Choose one testable cause, make one material change, and compare the same funnel stage against a comparable period.

Shopify also provides sessions-by-landing-page and sessions-by-device reports. Use them to locate where a funnel change is concentrated before you redesign a page that is performing well for most shoppers.

How to diagnose each funnel stage

All sessions to cart additions

Start with the landing pages and products receiving the affected sessions. If the weakness is concentrated in collection or category traffic, review product order, out-of-stock visibility, filters, search results, and the path from a collection to a product page. A product-level view helps separate a weak product offer from a weak traffic source. Read the guide to product conversion rate.

Cart additions to reached checkout

Review the cart experience for the affected segment. Check whether shipping information, delivery expectations, discount logic, cart errors, or a confusing checkout path appear before the shopper reaches checkout. Compare the issue by device because a cart interaction can fail on one device type while working on another.

Reached checkout to completed checkout

Inspect the checkout experience and the orders that do complete. Look for patterns in payment method, destination, device, discount use, and order value. Treat an unusual change in this stage as a reason to investigate checkout friction, price changes, shipping costs, or payment-related issues before assuming one cause.

Use product and collection segments to find the cause

A storewide funnel tells you where sessions stop. Product and collection segments help explain why. A low cart-entry rate limited to one collection can point to its product mix, ranking, or traffic intent. A low rate limited to a single product can point to that product page, price, availability, or offer.

Review high-traffic collections separately from low-traffic collections. Do not let a small collection with a volatile rate dictate a storewide change. Use collection optimization to examine whether strong products are buried or weak products dominate the collection path.

Decision checklist before you fix a funnel leak

Use sessions for every funnel stage and keep the funnel closed when diagnosing sequential drop-off.
Compare the same date length and similar commercial conditions.
Write down the baseline source for each stage rate, downstream rate, and average order value.
Rank stages by estimated revenue at risk, not by percentage drop alone.
Confirm the issue in at least two comparable periods before making a broad site change.
Segment by device, source, landing page, campaign, and product group to locate the affected traffic.
Change one plausible cause first, then measure the same stage again.
Keep winning changes and reverse changes that do not improve the targeted stage.

Once you know the stage and segment with the highest revenue at risk, move to a focused test. For a broader testing framework, see Shopify conversion rate optimization.