Guide · FifteenFold
Ecommerce audit checklist: what I check before a redesign
I don’t recommend a redesign until I’ve been through the customer journey, in this order: discovery, category pages, product choice, product data, trust and checkout. Still working out why traffic isn’t turning into sales? Start with the six common reasons online shops stall.
My short answer: when traffic looks healthy but growth has stalled, I want to find the biggest commercial bottleneck before changing anything. It may be the storefront. It may also be product data, merchandising, availability or the traffic mix.
My ecommerce background makes me wary of diagnosing everything as a design problem. A beautiful product page cannot compensate for customers being sent to the wrong products or Google receiving weak information.
Can customers find the right product?
I start with navigation, search, categories, filters and sorting. If discovery is hard, the customer may never reach the product page everyone wants to optimise.
Does the product page answer the buying questions?
Unsure whether your website needs fixing or rebuilding? Get a free website review
Images, condition, options, delivery, returns, compatibility, stock, trust and price context matter. On mobile, the order and density of this information matter even more.
What information is being sent to Google?
Titles, product types, identifiers, availability and price affect how product listings are understood. Feed and Merchant Center work sounds technical, but in plain English it is about making sure Google receives accurate, useful product information.
Where does checkout create doubt?
Unexpected costs, weak delivery information, forced account creation or an awkward payment journey can undo everything earlier in the funnel. I review the complete purchase journey rather than one template in isolation.
Start with the funnel, not an app recommendation
When sales decline, record the period, channel mix, device split, transaction count, conversion rate, average order value and revenue per session. A fall in conversion after a paid-media budget change does not automatically mean the checkout broke: the visitors may be different. Compare comparable segments and annotate promotions, stockouts and site releases.
| Funnel signal | What to inspect | Example hypothesis |
|---|---|---|
| Collection views, few product views | Filter relevance, sort order, availability and product photography | Mobile filters hide the sizes people need |
| Product views, few add-to-carts | Variant clarity, delivery costs, trust and product information | Users cannot tell which size fits their requirements |
| Cart additions, low checkout starts | Unexpected delivery charges, minimum order and cart distractions | Delivery threshold is unclear before the basket |
| Checkout starts, low purchases | Payment method failures, address validation, stock reservation and mobile keyboard | A form error is not visible to the customer |
The product-discovery walkthrough
Choose three representative products: a bestseller, a low-visibility item and a variant-heavy product. Starting on the homepage, can you find each in under a minute? Can you narrow the collection by actual product attributes rather than invented marketing categories? Open a product page and check title, format, size, stock, price, delivery and returns alongside the main buy action. Confirm that the selected variant is the one placed into the basket.
For a store with several oils sold in 3ml, 6ml, 12ml and 24ml, a variant control must show the price and selected quantity clearly. A beautiful photograph cannot compensate for putting the wrong variant in the cart. Our Hayati Essence redesign study discusses these decisions, without claiming a measured sales increase.
Check product data and Merchant Center as part of the journey
Compare product feed titles, prices, identifiers, images and availability to the actual sellable variant. An inaccurate feed can send people to an irrelevant landing page or prevent a product from being eligible to serve. Look at item-level diagnostics and the 40-point feed checklist before increasing ad spend.
Turn findings into a practical backlog
Score issues by affected traffic, customer impact, effort and confidence. Fix a broken checkout first; do not prioritise decorative animation while mobile payment fails. For each change record the before state, the hypothesis, the pages and devices affected, and a rollback plan. On a low-traffic site, validate with real-user task testing rather than claiming a statistically proven uplift from five conversions. For a broader strategy discussion, read why ecommerce growth can stall.
Next step
Losing sales after the click?
Tell us what is happening in your store. We will help identify whether the priority is product discovery, trust, data or checkout.
Discuss your ecommerce issue