Ecommerce Customer Journey: How Shopify Merchants Can Find and Fix the Drop-Offs

What an ecommerce customer journey actually shows

An ecommerce customer journey is the sequence of decisions a shopper makes from first noticing a product to buying again. It includes the questions they ask, the pages they use to answer them, and the friction that stops them before a repeat order.

For a Shopify merchant, the useful version of that map is not a poster of personas. It is a stage-by-stage list of what the shopper needs to see, what your store currently shows them, and which metric tells you the handoff failed. Discovery, product evaluation, cart, checkout, and post-purchase are the five stages that change revenue most often.

Baymard Institute's checkout usability research finds that about 70% of online shopping carts are abandoned. That number is a checkout symptom, not the whole journey. Many of those carts were already damaged by a thin collection, a missing size, a slow answer on shipping, or a product page that never earned trust.

The five stages that matter for a Shopify store

Academic journey models often use awareness, consideration, purchase, retention, and advocacy. Those labels are fine. Operationally, Shopify stores get clearer actions if each stage is tied to a page type and a decision.

1. Discovery

The shopper learns the store exists. Traffic may come from search, ads, email, social, or a referral. The job of this stage is to land them on a page that matches the promise that brought them in, not on a generic homepage.

2. Product evaluation

They compare options inside a collection or on a product page. Price, photos, reviews, variants, delivery timing, and stock status decide whether the click becomes an add to cart. This is where most paid traffic is wasted if the product cannot convert.

3. Cart

They have chosen something and are testing the total. Shipping cost, discounts, related products, and stock warnings either confirm the decision or reopen it. A cart is not a form. It is a last comparison against leaving.

4. Checkout

They commit payment and address details. Shopify checkout is a hosted flow with limited theme control, so the leaks you can fix are mostly upstream: required information, payment options, shipping rates, and the product state they carry into checkout.

5. Post-purchase

Delivery, unboxing, support, and the next relevant offer decide whether the first order becomes a second one. A journey that ends at the thank-you page describes a transaction, not a customer.

How a customer journey differs from a funnel

A funnel counts how many people move from visit to purchase. A customer journey explains why they moved, stalled, or left, and what they needed at that moment. Both are useful. They answer different questions.

Use the funnel when you need a rate: sessions to product views, product views to add to cart, add to cart to checkout, checkout to order. Use the journey when a rate is weak and you need the cause: wrong collection order, a sold-out hero variant, unclear shipping, or a post-purchase email that never asks for the next order.

The funnel model, and how to read each conversion step, is covered in the ecommerce funnel guide. This article stays on the decisions between those steps.

A falling conversion rate tells you a stage broke. The journey tells you which decision the shopper could not finish.

Map the journey with data already in your store

You do not need a research agency to build a first map. Shopify already records sessions, product views, add-to-cart events, checkout starts, and orders. Pair those counts with a short look at the pages shoppers actually use.

Shopify's analytics reports cover sales, sessions, and conversion behavior in the admin. The practical starting point is the online store conversion funnel and product-level reports, not a custom dashboard. Confirm the exact report names in your admin, because Shopify has moved some reports between Analytics and the newer Shopify Analytics experience.

1. Pick one entry path. Choose paid product ads, organic collection traffic, or email. Do not average every channel into one map.
2. Write the shopper's question at each stage. Discovery: "Is this the thing I searched for?" Evaluation: "Will this version work for me?" Cart: "Is the total fair?" Checkout: "Can I pay the way I expected?" Post-purchase: "Was this worth doing again?"
3. Attach one page and one metric to each question. Collection or landing page with bounce and product-click rate. Product page with add-to-cart rate. Cart with checkout-start rate. Checkout with completion rate. Post-purchase with repeat purchase rate.
4. Note the blocker in plain language. "Size M is out of stock on the bestseller." "Shipping appears only at checkout." "The winning SKU is on page 3 of the collection."
5. Keep the map to one primary persona and one primary product line. A second map is useful only after the first one produces actions.

A simple table is enough. Columns that stay useful are stage, shopper question, page, metric, current blocker, and owner. If a cell has no owner, it will not get fixed.

What to check at each drop-off

Discovery: match the promise to the landing page

Open the top landing pages for the channel you are mapping. If an ad promises a specific product and the click lands on a broad collection, the journey has already split. The shopper has to search again inside your store.

Check three things before you rewrite copy: the first screen shows the product or category from the ad or query, the collection sorts a relevant product above the fold, and out-of-stock items are not the first thing they see. A pretty homepage is a weak answer to a specific search.

Evaluation: make the product page finish the comparison

A product page fails this stage when it does not finish the comparison the shopper came to make. Missing size, material, compatibility, or delivery detail sends people back to search, even when the price is fine.

• Variant clarity: every sellable option is selectable, labeled, and priced. Hidden or sold-out variants should not look identical to available ones.
• Proof near the decision: reviews, materials, dimensions, and a delivery estimate sit where the add-to-cart button is, not only in a tab three scrolls down.
• Comparison help: if shoppers bounce between three similar SKUs, the collection filters and product copy should say how they differ.
• Check your own device mix before you assume desktop is the default. If a large share of sessions is on a phone, and the key spec sits below a long description, treat that as a journey break.

Product-level conversion, not storewide conversion, is the metric that belongs here. A store can look healthy while three advertised products convert poorly and absorb the budget. Product analytics is the place to separate those products from the ones that deserve more traffic.

Cart: remove the surprise before checkout

Baymard Institute’s 2026 cart-abandonment survey of 1,083 US adults lists extra costs, forced account creation, and slow delivery among the reasons people leave. 40% abandoned because extra costs such as shipping, tax, and fees were too high, and 20% left because delivery was too slow. Those are journey failures you can often see before checkout analytics.

Show the shipping threshold, delivery window, and return rule on the product page and again in the cart. If a discount code is part of the ad, the cart should not make the shopper hunt for it. Related products belong here only when they answer a real gap, such as a refill or a required size, not as a second catalog.

Recommendation logic that matches the product they already chose is covered in Shopify product recommendations. Use it to complete the order, not to restart evaluation.

Checkout: fix what you can, stop blaming the form

Shopify checkout is standardized for a reason: payment, address validation, and fraud checks stay consistent. Theme edits will not rescue a checkout if the shopper arrived with an unresolved shipping or stock question.

Still check the settings you do control. Offer the payment methods your customers already use. Keep guest checkout available unless you have a clear wholesale reason not to. Put delivery and tax estimates earlier so checkout is confirmation, not a reveal. In the same Baymard survey, 19% abandoned because they did not trust the site with their credit card information. That is a trust problem built across the whole journey, not a field-label problem.

If checkout completion is the weak rate and the earlier stages look clean, the next read is a focused Shopify conversion rate optimization pass on payment options, shipping rates, and error states. Do not rebuild the product page to fix a payment-method gap.

Post-purchase: design the second decision

The first order answers "Will this arrive, and is it what I bought?" The second order answers "Do they still have something I need?" Tracking emails, a clear return path, and a replenishment or complementary offer are the pages of this stage.

Measure repeat purchase rate and time to second order for the product line you mapped. A thank-you page upsell that ignores what they just bought is noise. A timed refill reminder for a consumable, or a fit follow-up for apparel, continues the same journey.

Where availability and collection placement break the path

Two store decisions sit underneath every stage and rarely appear on a classic journey map: whether the product can be bought, and whether the shopper can find it.

A stockout is a journey break with a specific owner. If the variant they evaluated is unavailable, evaluation ends even when the product page converts well on paper. If you reorder after the stockout, you have already paid for the sessions that could not finish. Forecast the reorder from recent sales velocity and lead time, and alert before the hero variant hits zero. That is an inventory decision, not a copy test.

Collection order is the other silent break. A high-converting product buried on page two of a collection never enters evaluation for most shoppers. Sort and merchandising should put products that convert, and that are in stock, where the collection’s traffic actually looks. Collection optimization is the operational version of this stage: which products help the collection convert, and which high-converters are underexposed.

• Flag hero products with less than one reorder cycle of cover. Do not send new ad spend to a SKU that cannot survive the campaign.
• Demote out-of-stock and low-converting products in the collections that receive paid or email traffic.
• Promote in-stock products with a proven add-to-cart rate into the first screen of those collections.
• Recheck the map after a stockout. A sudden drop in add-to-cart rate is often availability, not a creative failure.

Turn the map into a weekly action list

A journey map that is not tied to next week’s work is a workshop artifact. Rank fixes by revenue at risk, not by how interesting the insight is.

1. Export last week’s sessions, add-to-cart rate, checkout completion, and orders for the product line you mapped.
2. List every blocker you can name in one sentence. Keep only blockers you can see in the store or in a report.
3. Estimate impact roughly: sessions on the broken page times the gap to your normal conversion, times average order value. A rough rank is enough.
4. Assign one owner and one change per blocker. Examples: move the in-stock bestseller to position 1, add the shipping threshold above the fold, pause ads on the sold-out variant, rewrite the size guide.
5. Review the same five metrics next week. If the stage rate did not move, the blocker was wrong. Change the hypothesis, not the whole map.

Three fixes usually outrank a redesign. Restore stock on the product that already converts. Stop promoting products that get clicks and no orders. Surface the products that convert and are buried. Those actions sit on the journey, and they are specific enough to finish in a week.

Skymetrics is built for that ranking step. Its agents watch inventory, product conversion, and collection performance, then order the actions by revenue impact: what to reorder, what to promote, and what to demote. The map still needs a merchant who knows the customer. The weekly list should not depend on someone rebuilding the spreadsheet.

A map worth keeping

Start with one channel, one product line, and the five stages above. Write the shopper’s question, the page, the metric, and the blocker. Then fix the highest-revenue break before you add personas, workshops, or another tool.

If the next action is unclear, open the product that spends the most and still fails to convert, check whether it is in stock and visible in its main collection, and change that before you touch checkout design.