Mapping body zones to product collections without making a mess

A body map is a promise. Click here and you will get products for this area. Break that promise twice and the shopper goes back to the search box, which is the thing you were trying to improve on.

Most of the failures we see are not technical. The iframe loads fine, the events fire fine, and the map sends people to a collection that answers a question they did not ask.

Work backwards from the catalogue

The instinct is to open the zone list and assign a collection to each of the 23 grouped regions. Do the opposite.

Write down every collection you would be happy to send body-area traffic to. For a mid-size support and recovery store that is usually somewhere between eight and fifteen. Then ask, for each one, which body areas should lead here.

You end up with the same mapping, but the failure mode changes. Working from the zone list produces a destination for every zone, including the ones where you had to invent something. Working from the catalogue produces gaps, and gaps are useful information rather than a problem to paper over.

The merges you almost always want

Left and right. Separate zones in the data, one destination in practice. Unless you sell a left knee brace and a right knee brace as distinct products, a left-shoulder collection and a right-shoulder collection with the same items in both looks like a mistake and doubles your maintenance.

Upper and lower arm. Shoppers do not think in those terms. They think elbow, wrist, shoulder. If your catalogue is organised by joint, merge the arm segments into the nearest joint collection.

Leg and knee. The grouped zone is left-leg, but nine times out of ten the product someone wants is a knee support. Send leg clicks to knees unless you have a real calf, thigh or shin range.

The splits worth making

Upper back and lower back. Different products, different shoppers. Posture correctors and thoracic braces for one, lumbar supports and cushions for the other. We see this pair merged more than any other, and it is usually the highest-traffic pair in the catalogue.

Neck versus upper back. Close on the model, far apart in the product range. Cervical collars and pillows are not posture correctors.

Foot versus ankle. Insoles and arch supports on one side, ankle braces and compression sleeves on the other. If you stock both properly, split them.

What to do about gaps

You will find zones with nothing behind them. Head, face, abdomen and pelvis are the usual suspects in a support and recovery catalogue.

In order of preference:

Leave the zone unmapped. The click does nothing, or shows a short “nothing here yet” state. Honest, and it costs you nothing.

Route to the nearest relevant collection. Works when the relationship is obvious to a shopper. Abdomen clicks going to abdominal support belts is fine. Face clicks going to your knee collection is not.

Fill the gap. The interesting option, and the reason to look at the click data before you decide. If abdomen is getting clicks and you have nothing for it, that is a stocking decision with evidence behind it.

Grouped or detailed

Grouped mode has 23 regions. Detailed has 158, down to specific tendons, joint aspects and muscle groups.

The test is simple: can you name at least two products whose difference is only visible at the detailed level. A clinic supplier with separate medial and lateral knee braces passes. A store with one knee sleeve does not.

Detailed mode also changes what you get back. Zone selections carry region, side and aspect alongside the zone key, which is what you want when you are prefilling a clinical intake form. For routing to collections it is usually more resolution than the catalogue can use.

Start grouped. You can switch with one configuration parameter, and the zone keys are stable, so moving later does not invalidate the mapping you already built.

Keep the map somewhere a human can edit it

Whether it lives in the Shopify app admin or in a JavaScript object in your theme, someone who is not a developer needs to be able to read it and say “that is wrong”.

A plain table works:

The product count column is the one that pays for itself. It surfaces the thin destinations before a shopper finds them, and it turns “we should probably stock more foot stuff” into a number someone can argue with.

Then watch what people click

Once it is live, the click distribution is the most useful merchandising signal the body map produces. Not the conversion number, which takes months of traffic to read honestly. The distribution.

You will usually find that three or four zones take most of the clicks, that the ranking does not match your revenue ranking, and that at least one zone with real interest has almost nothing behind it.

That last one is the whole reason to do this properly. A body map is a navigation tool on the way in, and a demand survey on the way out.