E-commerce that converts: the details most stores get wrong
Most stores do not lose sales to their design. They lose them to a missing delivery date, a forced account, a slow search and a product page that answers the wrong questions.
Contents
When a store underperforms, the conversation usually turns to the design. Occasionally that is the problem. Far more often the store is losing people at four or five specific, unglamorous points, and each of them is fixable without redesigning anything.
These are the ones we find repeatedly.
The product page answers the wrong questions
Most product pages are written by someone who already knows the product, for someone who does not. The result is a page full of features and short of the things that actually decide a purchase.
The questions a buyer is silently asking are usually: will this fit, will this suit me, when will it arrive, what if it is wrong, and can I trust you. Work through them in order.
- Will it fit. Real dimensions, real weights, a size guide with measurements rather than letters, and where relevant a sense of scale from an image with something familiar in it.
- Will it suit me. Materials, finish, what it is like in use, and photographs that show it in context rather than only on white.
- When will it arrive. Not “standard delivery”. A date, or at minimum a realistic range, shown on the product page rather than discovered at checkout.
- What if it is wrong. The returns terms in plain words, near the buy button, not in a footer link. Hiding an awkward returns policy does not make people buy. It makes them not buy and not tell you why.
- Can I trust you. A real address, a real phone number, real reviews if you have them and nothing if you do not.
Variants and stock are handled badly
Choosing a size should not reload the page, lose your place or silently swap to a different image. Options that are unavailable should be visibly unavailable at the point of choosing, not after pressing add to basket.
When something is out of stock, say when it is expected and offer to tell them when it returns. An out of stock product that captures an email address is worth considerably more than one that presents a dead end.
Search does not work
People who use search on a store are the people most likely to buy, and on most stores the search is whatever the platform shipped with. It fails on misspellings, on synonyms, on product codes and on anything typed the way a customer thinks rather than the way the catalogue is structured.
Test it yourself with twenty things a customer would plausibly type, including two with typos and two using a word your industry does not use. If half of them return nothing, that is the highest value fix available to you and it has nothing to do with design.
The basket creates doubt
The basket is where confidence is either confirmed or lost. It needs to show the full cost including delivery, before the checkout, because an unexpected charge at the final step is the single most cited reason people abandon.
It should also be easy to leave and come back to. Persist the basket across sessions and devices where you can. Do not use a countdown timer or a stock scarcity message you cannot substantiate, because customers recognise both, and a store that manipulates over a small purchase is not trusted with a large one.
Checkout asks for too much
Every field is an opportunity to stop. Ask only for what is needed to take the money and deliver the goods, and ask for it in an order that makes sense.
- Offer guest checkout prominently. Forcing an account before a first purchase removes sales in exchange for a database record.
- Use address lookup rather than five manual fields.
- Use the correct input types so phones present the right keyboard, and support the browser’s own autofill instead of fighting it.
- Validate as the customer types, and put the error next to the field, in words that say what to do rather than what went wrong.
- Never clear a form on error. This is unforgivable and still happens.
- Offer the payment methods your customers actually use, including the wallet options that skip typing a card entirely on a phone.
- Show progress, and keep the order summary visible throughout.
Mobile is treated as the small version
For most stores the majority of traffic is on a phone and the majority of revenue is not, which is usually a diagnosis rather than a fact of life. Check the things that only bite on a phone: tap targets large enough to hit while walking, a sticky add to basket on a long product page, images that do not force horizontal scrolling, and a checkout that works with one thumb.
After the sale, nothing happens
The period between payment and delivery is when a customer is most attentive and most anxious. A confirmation that restates exactly what was bought and when it will arrive, a dispatch notice with real tracking, and a clear route to a human if something is wrong will do more for repeat business than any discount code.
This is also the cheapest part of the whole system to get right, and the most frequently ignored.
How to find your own version of this list
Buy something from your own store, on a phone, on mobile data, as a new customer, paying with a real card. Write down every moment you hesitated. Then ask two people who have never seen the site to find a specific product and buy it, and watch without helping.
That exercise takes an afternoon and produces a better list of priorities than any report, because it is about your store rather than stores in general.