Anatomy of a product page that lets people decide

A product page has one job: to let somebody make a decision. Not to persuade, not to move them, not to tell the brand's story. To remove enough doubt that a person can decide, one way or the other.

Almost all of them do the opposite: they open with an evocative line and bury at the bottom, or nowhere, the three pieces of information that were needed.

Above the fold: the four non-negotiables

On a phone you have little more than one screen before the person decides whether to scroll. Four things have to fit in it, and none is a marketing sentence.

What it is, said the way somebody searching would say it. The title is not the internal product name. "Waterproof 30-litre backpack for 15-inch laptops" says everything; "the Aria model" says nothing to anyone who does not already know you.

What it costs, really. Price, availability and when it arrives. The "when it arrives" is almost always missing, and it is among the first things a person wants to know.

What it looks like. A photo showing the product as it is, with a scale reference. Not the styled scene: that comes later.

How to buy it. The button, visible without hunting, and not covered by a promotion or a banner.

Below: the questions, in the order they arrive

The rest of the page should be built as a sequence of answers, not as a piece of writing. And the right order is not yours to decide: support decides it.

The questions arriving every day about your category are the page's contents list. If the three most frequent are size, material and shipping, those go immediately below the fold, in that order, and not in a specification tab rolled up at the bottom.

It is a less creative way of working and a far more effective one, because it removes the guesswork: you are answering questions you know for a fact somebody asks.

What is almost always missing

The product's measurements, not the packaging's, with a concrete reference. "40 cm" is data; "compact" is an opinion.

What is in the box, including the obvious. If the cable is included, write it: if you do not, somebody will ask and somebody else will buy elsewhere.

Who it is not right for. The single most underrated line there is. It reduces returns, raises the credibility of everything else on the page, and is the part an answer engine quotes most willingly, because it is the one piece of information no competitor writes.

The terms, on the page. Returns, warranty, shipping. They are why somebody abandons at the last step, and putting them three clicks away does not make them less decisive: it only makes them invisible.

What can go

The opening paragraph about the brand. The description that repeats the title in other words. Generic trust badges that sit on every site and that nobody reads any more. Reviews loaded at the bottom with none of the useful information extracted.

They do no harm in themselves: they take the space and attention the four things above needed.

How to tell whether it works

The most honest number is not the page's conversion rate, which depends on a thousand things. It is how many support questions arrive about that product: if a page generates questions it is not doing its job, and the questions also tell you exactly what is missing.

The second number is that item's return rate, for the same reason inverted: a return is a wrong decision made with the information you gave.

Frequently asked questions

How long should a product page be?

As long as the questions there are to answer. A simple accessory is done in a few lines; a technical product needs many more. The same length across the catalogue is the sign that nobody decided, they just filled a field.

Prose or bullet points?

Bullets for data, prose for things that need explaining. The common mistake is the reverse: prose for specifications, which becomes unreadable, and bullets for concepts, which become telegraphic.

Do reviews replace missing information?

No, but they reveal it. If ten reviews say "runs small", that is not an opinion: it is data missing from the page that you are leaving people to discover by chance if they scroll far enough.

Is A/B testing product pages worth it?

On a high-volume store, yes. Below a certain traffic threshold a test takes months to say anything, and in those months you are better off simply adding the information that is missing, which is a certain improvement rather than an uncertain experiment.

Which product do I start with?

The one generating the most support questions, not the one selling most. The first has an identifiable problem; the second already works.