A cart is not the same as a checkout
Cart abandonment can happen after someone adds a product but before they start checkout. Abandoned checkout means they begin checkout without completing the order. Someone browsing a product without adding it is at an earlier stage again.
Shopify distinguishes these stages in its recovery automations. Keep the distinction when you review reports or compare tools: a “recovered checkout” does not describe every visitor who left an item in a cart.
Source: Shopify’s recovery automation guidance.
Start here: compare cart additions, checkout starts, and purchases for the same period. Check a few representative journeys yourself. If the store has almost no cart additions, begin with the traffic-but-no-sales checklist.
Match the investigation to the stage
A drop-off suggests where to look. It does not assign a motive. Use the following as an investigation list, not a set of conclusions about individual shoppers.
| Where they stop | What to investigate |
|---|---|
| Product page or cart | Fit, variant choice, product details, price/value, comparison shopping, and whether the item is available. |
| Cart to checkout | Unexpected total cost, delivery timing, shipping availability, and a cart that is awkward to edit on mobile. |
| During checkout | Address or shipping-rate problems, payment failures, unavailable payment methods, and technical errors. |
Some people use a cart to save or compare items. Some are interrupted. You will not recover a reliable explanation for every abandoned visit, and that is a reason to gather evidence rather than invent one.
Five little doubts. Five very different fixes.
Your dashboard can show what sold, what didn’t, how many, and when. A funnel can show where shoppers stopped. Neither tells you, on its own, what they were unsure about.
The same unfinished cart could hide any of these questions. Treat them as possibilities to investigate, not conclusions about a shopper.
“Is this really worth the price?”
Repeated price checks may be a value question. Can they tell what’s included, how it’s made, and why this option costs more?
A possible fix: explain the value with specific details. If price itself is the barrier, investigate the offer too.
“Will this actually fit me?”
A size chart opened again and again is a clue. It doesn’t tell you which measurement or reassurance is missing.
A possible fix: clearer measurements, fit notes, and answers beside the size selector. More traffic won’t make a confusing chart clearer.
“Will it arrive when I need it?”
“Ships fast” may leave a shopper with a deadline guessing. Processing time and delivery time are different questions.
A possible fix: put realistic delivery expectations and applicable costs where they choose the product.
“Can I trust what I’m buying?”
They may be looking for a clear returns policy, verifiable product details, or relevant feedback from actual customers.
A possible fix: make genuine reassurance easy to find. More persuasive claims aren’t a substitute for evidence.
“Which option is right for me?”
Switching between variants can mean the differences are hard to understand. It can also mean ordinary comparison shopping.
A possible fix: explain the differences beside the choices, then ask what still needs clarifying.
A shopper’s answer adds another piece of evidence. It doesn’t explain everyone who left. For more ways to investigate, read what dashboards can and can’t tell you.
Check the total and the delivery promise
Review the whole amount a customer sees, not just the product price. Are shipping costs, thresholds, and applicable charges understandable before the last step? Does an offer still apply to the chosen variant and destination?
Then check time. A shopper buying for a date needs more than “ships fast.” Distinguish processing time from delivery time, communicate realistic estimates, and avoid promising a date you cannot support.
Try this: use a phone to build a normal cart and a mixed-product cart. Test representative addresses without placing an unintended order. Verify the shipping options, total, inventory, and ability to change quantities. Keep notes of discrepancies instead of relying on memory.
Look for payment problems before changing the price
A failed payment attempt can look like lost interest in a funnel report. It needs a different response. Shopify’s abandoned-checkout timeline can show payment events and failure details; use that evidence to investigate the gateway or checkout issue.
Source: Shopify’s checkout payment troubleshooting.
Try this: inspect recurring errors, affected payment methods, device patterns, and whether the issue can be reproduced in an appropriate test setup. If customers report a failure, ask for the relevant error and time through your normal support process. Do not request full card details.
What recovery email and SMS can do
Recovery messages invite an eligible, reachable customer back after an unfinished visit. A configured workflow can send a reminder and a route back to the relevant cart or checkout. Depending on your provider, it can also apply timing, purchase exclusions, and message limits.
Reach is conditional. A visitor needs to be identifiable through a usable contact method and qualify for the workflow. Contact details, subscription status, channel eligibility, and your provider settings all matter. An anonymous product-page visitor may never supply those details.
Shopify’s current recovery automations target marketing subscribers by default. Its abandoned-checkout email settings have an additional recipient option that is not available for the SMS and WhatsApp versions. Check the guidance for the channel you actually use instead of assuming one setting covers every message.
Source: Shopify’s current automation and recipient settings.
A reminder can bring someone back to the same unanswered question. If fit was the doubt, the size chart still needs an answer when they return.
Review your sequence: does it identify the relevant items, link to the right destination, stop appropriately after purchase, and still describe an available offer? Check delivery and completion reporting as well as the template. A useful reminder does not always need a discount.
Before you give away 10%, find out what was missing.
Before changing the offer, write down your hypothesis. “People cannot tell whether the size will fit” calls for better sizing information. “The delivered total is higher than the page suggests” calls for clearer cost information. Neither becomes clearer merely because a percentage comes off.
- Read recent support questions about delivery, suitability, variants, and price.
- Review genuine feedback for details shoppers expected but could not find.
- Compare the campaign promise with the product and delivered price.
- Change one well-supported issue, record the date, and evaluate a suitable period afterward.
Watch for simultaneous changes in stock, traffic, and promotions. A before-and-after comparison can help you decide what to try next; it does not isolate the cause by itself.
Ask. Understand. Fix. Check.
Recovery campaigns invite shoppers back after an unfinished visit. Sell & Tell adds evidence earlier, while they’re still considering the product. The useful part is what happens after the question.
- 01 / ASK
Iris asks while they’re deciding.
When eligible product-page behavior suggests hesitation, Iris asks a short, relevant question. Shoppers can tap an answer or use their own words. It’s easy to dismiss and never interrupts checkout.
- 02 / TELL
Your Monday Brief makes it actionable.
Shopper responses and behavioral evidence become a prioritized plan: Fix first. Do next. Stop doing. Each recommendation shows its evidence. Dollar estimates wait for enough shoppers and are labeled as estimates.
- 03 / GRADE
You make the change. It checks the result.
Clarify the fit, delivery, or product details, then review the before-and-after comparison. The verdict can be Worked, No clear change, or Made it worse. Too little evidence means waiting, not inventing a win.
A before-and-after comparison is not a controlled experiment or proof that your change caused the result. Sell & Tell does not contact people after they leave, repair checkout errors, or automatically recover carts.
Keep recovery campaigns where they work. Add shopper evidence to understand what may need fixing before the next visit.
A few questions before you start.
Can you tell me exactly why every shopper left?
No. Iris collects answers from shoppers who choose to respond during eligible product-page hesitation. Those answers and supported behavioral evidence help you investigate; they cannot reveal every visitor’s motivation.
Does this replace abandoned-cart emails?
No. Recovery messages encourage eligible, reachable shoppers to return. Sell & Tell helps investigate product-page doubts before they leave. The two approaches can work together.
Do I have to rebuild or migrate my store?
No migration or redesign is required. Connect your existing Shopify store, customize Iris, and review the product-specific questions. You choose which recommendations to act on and make the changes yourself.
Will this improve my conversion rate?
It can uncover questions worth addressing, but an improvement is not guaranteed. The result depends on your shoppers, products, traffic, and changes. Before-and-after movement is useful evidence; promotions, stock, and other changes can also affect it.