CONVERSION · SEO · 21 JANUARY 2025 · 9 MIN READ
The product page audit: twenty things worth checking
A product page has one job and about twenty ways to quietly fail at it. This is the list we work through on a store audit.
A high-converting product page answers the four questions a buyer actually has — is this the right thing, will it fit or suit me, when will it arrive, and what happens if it is wrong — above the point where they decide. Most underperforming product pages are not badly designed; they answer those questions somewhere further down, or not at all, and rely on the customer being patient enough to look.
IN SHORT
- Four questions decide a product page: is it right, will it suit me, when does it arrive, what if it is wrong.
- Delivery date is the most under-used conversion lever on most stores — an estimate beats "ships in 2-3 days".
- Returns policy belongs on the product page, not only in the footer. It removes risk at the moment risk is felt.
- The variant picker is where considered purchases are lost, especially when out-of-stock options are hidden rather than shown.
- Most product pages are slow because of apps, and the slowest element is usually the one added to help conversion.
Is this the right thing?
The first job is identification, and it is mostly about images and the first two lines of copy. A buyer who cannot tell within a second whether this is the item they came for will go back rather than read.
- The first image shows the product clearly, on its own, at a size that survives a phone screen. Lifestyle images come after, not first.
- Zoom works on mobile without pinching a lightbox that traps the scroll.
- The title reads the way a customer would say it, not the way the ERP stores it.
- The first paragraph says what the thing is and who it is for, before it says anything persuasive.
- Key specifications are visible without opening an accordion — material, dimensions, capacity, whatever the category cares about.
Will it suit me?
This is where category knowledge shows. Apparel needs sizing; audio needs compatibility; furniture needs dimensions against a room. The question is the same and the answer is different for every catalogue.
The variant picker carries most of this, and it is where considered purchases are most often lost. Hiding out-of-stock variants seems helpful and is not: the customer cannot tell whether their size exists at all, so they assume it does not and leave. Showing it as unavailable, with a back-in-stock option, keeps them on the page.
- Out-of-stock variants shown as unavailable rather than removed.
- Swatches that show the actual colour, not a name a customer has to imagine.
- A size guide that opens without leaving the page, with measurements for the thing being viewed rather than a generic chart.
- Compatibility or fitment stated where the category needs it.
- Reviews that can be filtered to the variant being considered, if you have the volume.
When will it arrive?
Delivery is the most under-used lever on most product pages. "Ships in 2-3 working days" asks the customer to do arithmetic they will get wrong, and most will assume the worse answer.
An estimated arrival date — "order in the next four hours for delivery Thursday" — is concrete, checkable and removes a reason to delay. It requires knowing your cut-off times and your carrier performance honestly, which is the actual work. The component is easy.
The other half is stock truth. A page that says "in stock" when the number is wrong converts well and creates a refund, a support ticket and a customer who does not come back. Inventory accuracy is a conversion feature.
What if it is wrong?
Returns policy belongs on the product page, close to the buy button, phrased as reassurance rather than as terms. It removes risk at exactly the moment risk is felt, which is the whole reason to put it there rather than in the footer.
Same for the things that make a brand feel real: a contact route that is a person, a warranty if you offer one, and a payment badge set that does not look like it was pasted from 2014.
- Returns window stated in plain words on the page.
- Who pays return postage, said explicitly. Ambiguity reads as bad news.
- A visible way to ask a question before buying.
- Warranty or guarantee where the price point justifies it.
- Trust elements that are specific — a real review count beats a generic badge.
And the five that are technical
The last group are not visible, and they are where most of the measurable loss sits.
Product pages are usually the slowest templates on a store, because they carry the most apps. Reviews, upsells, personalisation, wishlists, size guides and back-in-stock widgets each add JavaScript, and each was installed to improve conversion. Measure the page with them and without, and be prepared for the result to be uncomfortable.
- Largest Contentful Paint measured on a real mid-range phone, not a developer laptop. The main image is usually the LCP element and usually unoptimised.
- Product structured data present and valid, including price, availability and — where you have them — reviews.
- Canonical correct on variant URLs, so five variants are not five competing pages.
- The buy button works before the page has finished loading everything else. It is the one element that must not depend on a third-party script.
- Analytics fires add-to-cart accurately, because every decision you make about this page depends on that number being right.
How to use the list
Work through it on your three highest-revenue products rather than on a template in the abstract. Real products expose the gaps — the one with a forty-character title, the one with nine variants, the one with no reviews yet.
Fix in order of the four questions, not in order of effort. A delivery estimate is worth more than a redesigned tab layout, and it is usually cheaper.
Questions this raises
What is a good conversion rate for a Shopify product page?
It depends so heavily on category, price point and traffic source that a benchmark is close to useless. The number worth tracking is your own, segmented by device and by traffic source, measured before and after a change. A store improving its own rate is doing better than one matching an average.
Should out-of-stock variants be hidden?
No. Hiding them means a customer cannot tell whether their size or colour exists at all, so they assume the product is not for them and leave. Showing the variant as unavailable, with a back-in-stock notification, keeps them on the page and captures demand you can act on.
Do product page reviews actually affect conversion?
They affect it most on considered purchases and on new products, and they matter more for the questions they answer than for the star rating. A review that mentions fit, durability or a use case removes a specific doubt; a five-star review with no text removes nothing.
How much do apps slow a product page down?
More than most teams expect, because the cost is cumulative and nobody measures after each install. It is common to find third-party JavaScript outweighing first-party on a product template. The fix is an app audit with measurement per app, not a general instruction to use fewer.
NEXT STEP
Free store audit
A senior Shopify engineer reviews your storefront, theme performance and checkout, then sends a prioritised list of fixes.