Optimising Ecommerce Pages for Search Engines and Readers

First published in 2014, rewritten in 2026 so the advice holds today.

Your product pages carry the full weight of your ecommerce business. There's no shop window, no member of staff to answer a quick question, and no physical shelf to browse. The page has to do all of that work itself, and it has to do it in a way that satisfies both a human shopper and a search engine that's increasingly likely to be reading the page with an AI model rather than a simple crawler.

The good news is that the two audiences want much the same things. Here's what still matters, and what's changed.

Use better images, in formats that load fast

Larger, higher-quality product images sell more than small, murky ones. This was true in 2014 and it's still true: an improvement to product photography can see ecommerce sales spike by up to 10%, which is a remarkable return for a relatively simple change.

Quantity still matters too. Multiple angles, zoomed detail shots and lifestyle images all help a buyer feel confident about a purchase they can't physically pick up.

What's changed is the technical side. Large, unoptimised images now directly hurt your Core Web Vitals scores, which Google has used as a ranking signal since 2021, and slow pages lose sales regardless of ranking. Serve images in modern formats such as WebP or AVIF, compress them properly, and lazy-load anything below the fold. Add descriptive alt text to every image: it helps accessibility, it helps image search, and it gives AI-driven search results something concrete to describe your product with.

Write in bite-sized, specific language

The original advice here still holds: condense the most important product information into short, scannable chunks near the top of the page rather than burying it in dense paragraphs. Shoppers skim. So do the AI systems now generating summaries and answers on Google's search results pages.

Use bullet points for specifications, keep sentences short, and lead with the details a buyer actually needs to make a decision, such as size, material, compatibility or delivery time. This naturally surfaces the keywords and product attributes that matter, without you needing to write for search engines and shoppers separately.

It's also worth adding structured data (schema markup) for products, price and availability. This doesn't change what a person reads on the page, but it gives search engines and AI Overviews accurate, machine-readable facts to pull from, which increases your chances of appearing in rich results or being cited directly in an AI-generated answer.

Real reviews, attached to the right product

Feedback still needs to sit next to the specific product it's about rather than being lumped into one generic reviews page. A short, genuine snippet from a real buyer next to a product listing is far more persuasive than a general "great brand" testimonial, because it speaks to that exact purchase decision.

Add review schema markup so star ratings can appear directly in search results. This was less common in 2014 but is now a standard, low-effort way to improve click-through rates from search.

Keep pages uncluttered

Cramming every inch of a page with related products, banners and promotional copy still does more harm than good. It slows the page down, dilutes the message, and makes it harder for a shopper to find what they came for. Be selective with cross-sells, upsells and promotional banners. Less genuinely is more, both for conversion rates and for page speed.

Navigation and checkout still decide the sale

Fast, logical navigation and a quick route to checkout remain some of the strongest levers you have, for both search performance and sales. A shopper should be able to move between categories, filter results and reach checkout without friction.

Since Google moved fully to mobile-first indexing, your mobile navigation and checkout experience effectively is your site as far as ranking is concerned, not a secondary version of a desktop site. If checkout is clunky on a phone, that's now both a conversion problem and an SEO problem.

The principle from 2014 hasn't moved: a page built to genuinely help a shopper decide is usually a page that performs well in search too. What's changed is the technical detail underneath it, from image formats to structured data to mobile performance, all of which now need to be right for that principle to actually pay off.