If you work in ecommerce, you've almost certainly heard of the jam study. It's used to argue that fewer options sell more, and it's usually told as if it were a law. The full story is more interesting, and more useful for understanding what happens to someone who lands in your shop and can't decide what to buy.

The jam experiment

In a study published in 2000, psychologists Sheena Iyengar and Mark Lepper described a tasting stand they had set up in a Californian supermarket. Sometimes it had six jams on display, sometimes twenty-four. With twenty-four, more people stopped to look, but far fewer bought: of those who came over, only 3% ended up buying, compared with 30% when there were six.

The result was so striking that it became one of the most quoted ideas in marketing: offer too much and people choose nothing.

What came next

Other researchers tried to reproduce the effect many times, with very mixed results. In 2010, Benjamin Scheibehenne, Rainer Greifeneder and Peter Todd pooled fifty experiments in a meta-analysis and found that, on average, the effect was practically zero, though with wide variation between studies. One important detail: people who already knew what they wanted were less put off by extra options.

In 2015, Alexander Chernev, Ulf Böckenholt and Joseph Goodman reviewed nearly a hundred results and identified the conditions that make overload more likely. Four factors play a part: how complex the set of options is, how hard the decision is, how sure the shopper is of their preferences, and whether they're there to choose or just browsing. With options that are easy to compare and clear preferences, the effect weakens or disappears. It's more likely when no option stands out, they're hard to compare and the shopper isn't sure what they want. In other words, "too much choice" isn't always a problem, but it can be under certain conditions.

When an online shop ticks those boxes

Think of someone buying their first robot vacuum. They open the category and find sixty models. They all look the same in the photos, and the product pages talk about suction in pascals, laser mapping, self-emptying bases and minutes of battery life. They don't know which of those things matter to them, because they've never owned one.

This is where the factors the research links to overload come together: options that are hard to compare, a tricky decision, unclear preferences, and a shopper who's there to buy one, not just browse. And a wrong choice costs a few hundred euros. The same person buying kitchen roll has no trouble at all with forty brands.

That's why overload depends less on the size of the catalogue than on the kind of purchase and how much the shopper knows. In categories where people buy for the first time, or rarely, a big range with no help can turn into a wall.

What to do without cutting your range

Plenty of people took the jam study to mean you should sell fewer products. What the later research suggests is that it matters less how many products you stock than how many options shoppers have to compare at once without knowing how.

  • Group by use before specification. "For small flats", "if you have pets" or "if you'd rather not empty it yourself" turns sixty robots into three manageable groups.
  • Explain what really separates similar models. Two lines on when it's worth paying more help more than a twenty-row spec table.
  • Suggest a recommended option for each type of shopper. It gives people somewhere to start without taking choice away.
  • Help them sort out their preferences. Shoppers often don't know what they want because they don't know which questions to ask themselves. That's exactly where a good shop assistant makes the difference, and it's covered in more detail in finding vs choosing.

An AI shopping assistant can do that work in a conversation: it asks about the things the shopper doesn't realise they need to decide, and narrows sixty options down to one or two that fit, without removing anything from the catalogue.

SmartShop AI, a sales assistant for Shopify and WooCommerce stores, does exactly that: it works out what the shopper is after and, if a key detail is missing, asks a single question before recommending one or two products from your catalogue with a short reason why.