Skip to content
Web Design

Product Configurator UX: Where Shoppers Get Lost and How to Fix It

6 Min Read

Most shoppers get lost in a product configurator for one of three reasons. They can’t see what they’re building, the price changes without warning, or there’s no way to stop and come back later. Fix those three and most of the drop-off goes with them.

When Willow & Hall first talked to us, they thought their customers were the reason people left. Their buyers were older, they said, and older buyers abandon at checkout. When we dug in, the sofa builder itself was fine. People were spending ages getting their sofa exactly right and then leaving at the point of purchase, because everything around the builder added friction, right up to a slow checkout. This guide covers where people get stuck building a sofa or a bed online, and what we’d change first.

Why do shoppers abandon product configurators?

Building a sofa online is a lot of decisions in a row. Size, shape, fabric, colour, legs and cushion fill each make sense on their own. Stacked up, they ask a lot of someone sat on their current sofa with a phone in one hand.

People don’t move through a configurator in a straight line. They pick a fabric, go back and change the size, then wonder whether the fabric still works. They open a second tab to compare two colours. They get to the last step and can’t remember what they chose at the second one, so they scroll all the way back up to check.

Somewhere in all that, a lot of them close the tab and decide to think about it. Anyone does that when the screen can’t keep up with them, whatever their age.

Can shoppers see what they’re building?

Start with the picture. Shoppers want to see the thing they’re building change as they choose. A static photo of the grey version while they’re picking green velvet leaves them to do the imagining themselves, and most people aren’t confident enough to spend four figures on their imagination.

You don’t need a 3D model for every combination. A good set of photos per fabric, swapped in as people choose, covers most of it. What matters is that the picture keeps up with the choices.

Fabric and finish are where this matters most. A swatch the size of a fingernail doesn’t tell anyone how a whole sofa will look in that colour. Show it on the product, and if you offer free fabric samples, put that option right next to the swatch rather than in the footer.

Does the price surprise anyone?

Show the price at every step and update it the moment something changes. A surprise at the end feels like a trick, even when it isn’t one.

This isn’t a configurator quirk. In Baymard’s cart abandonment research, extra costs being too high is the most common reason people give for leaving at checkout, cited by 40% of shoppers who abandoned. Baymard’s product page benchmark also found that 67% of sites don’t give an estimate of the total cost near the buy section. On a configured product, where the price moves with every choice, that gap is wider.

Delivery belongs in that number too. Large furniture often has its own delivery charge and lead time, so show both before the basket, not after.

Can people save their build and come back?

Furniture is rarely bought in one sitting. Someone will want to show their partner tonight, measure the alcove tomorrow, or wait for payday. If the configurator forgets everything the moment they close the tab, you’ve made them start again, and plenty won’t.

Give people a way to keep what they’ve built:

  • a shareable link that reopens the exact configuration
  • an email of the build, with the picture and the price
  • a saved build in their account, if they have one

None of these needs a login to start. Asking someone to create an account before they can save a sofa is asking for commitment before they’ve made up their mind.

Is the configurator slowing the whole site down?

Every option in a configurator can turn into its own product variant behind the scenes, and that weight lands on every page load. A configurator that takes a few seconds to respond to each tap makes people second-guess every choice.

With Willow & Hall we cut over 100,000 SKUs during the rebuild, which did as much for the experience as any visual change. If your catalogue has ballooned because every fabric and size combination is its own product, it’s worth looking at how the options are built before you redesign the screens.

What should you fix first?

Problem What shoppers do What to change
Picture doesn’t change Open new tabs, guess, leave Swap images as each option is picked
Price jumps at the end Abandon at basket or checkout Live price at every step, delivery included
Can’t save the build Close the tab to “think about it” Shareable link and email, no login needed
Slow to respond Second-guess each choice Review how options are built, cut unused SKUs
Long single page Scroll back up to check choices A running summary of choices that stays on screen

If you only do one thing this month, build a sofa on your own site, on your phone, from start to finish. Note every moment you had to scroll back, guess or wait. That list is your brief.

What changed for Willow & Hall?

We moved Willow & Hall from Magento 1 to Magento 2 and rebuilt the whole experience on Hyvä. The site got 80% faster and sales went up 15%, with conversion reaching 3.6%. The same customers who were supposedly too old to finish now finish.

The part we like best is what happens now. The configurator gets adjusted in days as new data comes in, rather than months, so it keeps up with how people shop on their phones. Get your configurator converting and the next problem is a nicer one, like working out how fast the workshop can make what people are ordering.

Who doesn’t need this?

If you sell a handful of products with two or three options each, you probably don’t need a configurator at all. A well-built product page with clear variant buttons will do the job and cost far less to look after. Configurators earn their keep when the combinations run into the hundreds and the product is expensive enough that people want to see it before they buy.

Frequently asked questions

What makes a good product configurator?

It shows the product changing as people choose, keeps the price visible and up to date, lets people save and share a build, and responds quickly on a phone.

Should the configurator be one page or several steps?

Either works if people can see a summary of their choices at all times and go back without losing anything. Long single pages tend to need a sticky summary. Step-by-step builds need a clear sense of how many steps are left.

Do we need 3D or augmented reality?

Not to start with. Good photography swapped in as people choose covers most of what shoppers need. 3D is worth considering once the basics are working and you’ve got the data to show people are still unsure.

How do we know where people drop out?

Track each step of the configurator as an event in your analytics, then watch session recordings of people who started building and left. The two together show you where and why.

How we can help

Send us a link to your configurator. We’ll build something on it on a phone, record the whole thing, and send you the recording with the five moments we’d fix first. If you’d rather look at the numbers behind it, our Free CRO Review runs your live speed data and shows you the first five issues.

Wondering where people give up on your configurator?

Book my NNP Workshop
Book my NNP Workshop Review my site