In 2000, two Columbia professors set up a jam tasting table at a grocery store. Some days they put out 24 jams. Other days, just 6. When the table had 24 options, 60% of shoppers stopped, but only 3% bought. When it had 6, fewer people stopped (40%), but 30% of them purchased.
More choices, fewer sales. It's been one of the most cited findings in consumer psychology for 25 years. Every eCommerce marketer has heard of it. Most treat it as a fun data point and move on. Very few have worked out what it actually means for how their Shopify store shows products.
What choice overload does to a buyer
Choice overload is the cognitive fatigue that sets in when too many options force too many comparisons. When the cost of deciding feels too high, the brain defaults to not deciding at all.
Barry Schwartz expanded on this in "The Paradox of Choice," arguing that beyond a certain threshold, more options reduce satisfaction, increase anxiety, and lower the probability of purchase. He documented this across financial decisions, medical choices, and consumer goods. Online shopping isn't any different — if anything, it's worse. A grocery shopper who walks away from a jam display still has to eat something eventually. An eCommerce visitor who gets overwhelmed just opens a new tab.
Baymard Institute found that 17% of cart abandonments happen because checkout is too complicated, but that figure understates where the real damage happens. The confusion that drives abandonment usually starts earlier, during browse and discovery, when a customer can't find what they're looking for and quietly gives up before they ever reach a cart.
How this plays out on a typical Shopify store
A merchant with 500 products doesn't show all 500 on one page. But they do show 24–48 items per category page, with filters most customers never touch, and a search bar that returns results ranked by keyword match.
A customer who types "summer dress" might get 60 results: sundresses, cocktail dresses with "summer" buried in the product description, and a few items that barely qualify. The customer has to sort through all of it to find what they're actually looking for. Most don't bother, and it shows up in the data as a bounce — not as evidence of a discovery problem.
The average Shopify store has around 450 products. The average visitor sees fewer than 10. That gap is where revenue quietly disappears. Customers who can't find what they want usually don't ask for help. They close the tab.
The wrong fix
The instinct here is to simplify the catalog: cut SKUs, trim the long tail, reduce the product count per page. That misreads the problem.
Catalog depth is a competitive advantage. A customer looking for a western hat with a specific brim width, crown shape, and felt grade shouldn't have to settle for three options. Having 500 products isn't what's hurting conversion. Showing the same undifferentiated products to every visitor, regardless of what they're actually looking for, is.
The Jam Study's real lesson is subtler than "stock fewer options." The researchers weren't arguing that grocery stores should shrink their inventory. They were documenting that each individual shopper benefited from seeing a smaller, curated selection. The right 6 jams, not all 24.
Online, you can show a different selection to every visitor, in real time, based on what you know about them. That's what product discovery is supposed to do, and it's the gap most Shopify stores haven't closed.
What "fewer, better products" actually looks like
A customer who has browsed your outdoor gear store three times, viewed trail running shoes twice, and bought trekking poles last month should see a completely different product set than someone who clicked an Instagram ad for your hiking backpacks for the first time.
Showing those two people the same homepage hero or the same featured products doesn't simplify the experience. It actively fails both of them.
First-time visitors benefit from seeing bestsellers filtered to the traffic source or campaign they arrived from — it anchors them quickly in relevant territory. Returning visitors should see new arrivals in categories they've already browsed, which signals that the store has picked up on what they care about. A customer who has abandoned a cart three times needs recommendations shaped by what they couldn't decide on, not the same homepage they've already scrolled past. And a post-purchase customer has already made one decision — what they need now are products that complement what they bought, not alternatives they've already considered and passed over.
None of this can be done manually at any meaningful scale. It requires a system that understands each customer's behavior and adjusts what they see across every touchpoint.
A real example: Hat Country
Hat Country is a western wear retailer on Shopify with a deep catalog spanning hats, boots, belts, and accessories. Product relationships in western wear are highly specific. A customer looking at a 4-inch brim cattleman crown in wool felt has almost no interest in a straw hat with a pinch front, even though both live in the same "hats" category. Surfacing them as related products is noise, and noise erodes trust.
After deploying AI recommendations trained specifically on Hat Country's catalog, covering how brim shapes correlate with crown styles, how boots pair with belts, and which accessory combinations match different hat types, the right products started reaching the right customers.
The result: a 23% lift in average order value, and a measured 40x ROI. The catalog didn't shrink. What changed was which products each customer actually encountered.
Signs your store is losing revenue to choice overload
A few things worth checking in your own analytics:
Browse-to-cart abandonment is high. Customers spending time on product pages but not adding to cart are often stuck in comparison mode with no clear path to a decision. The page held their attention but not their confidence. What you surface to them next can either push them toward a purchase or lose them entirely.
Search conversion rate is below 3%. Search visitors convert at 3–4x the rate of non-searchers on average, because they've already expressed intent. If your search is returning too many irrelevant results, your highest-intent visitors are being forced to do the sorting work themselves — and most won't. A 1% search conversion rate on a store with strong products is almost always a relevance problem, not a traffic problem.
Category pages have short session durations. If visitors are landing on a category page and leaving within 30 seconds without clicking a single product, the page isn't showing them anything that connects. Category pages with 40+ undifferentiated items and no personalization are a common cause — the customer scans the first row, doesn't see anything immediately relevant, and bounces.
AOV is flat despite a growing catalog. If you've expanded your product range but average order value hasn't moved, the store may not be surfacing complementary combinations to customers who are ready to spend more. A wider catalog should mean more cross-sell opportunity. If it isn't converting that way, the products aren't finding each other.
How AI personalization addresses the problem
AI personalization's job is filtering the catalog to the relevant subset for each visitor, matching what they see to their specific context in that session.
This requires understanding more than purchase history. A customer's browsing sequence, time spent on specific pages, products they've compared, searches they've run, and the traffic source they arrived from all contribute to what "relevant" means right now. AI models trained on your store's specific data can learn these patterns across thousands of signals simultaneously, at a scale no merchandising team can replicate by hand.
PersonalizerAI's recommendation and search systems are trained on each merchant's catalog, order history, and customer behavior. Every visitor sees a version of the catalog shaped by their context. Returning visitors get recommendations informed by prior sessions. Search results rank by personalized relevance, not keyword density. The same 450 products, surfaced differently for each person who visits.
For merchants running this across their full store, the average AOV lift is 23–34%. The catalog stays the same size. What changes is which products each customer actually encounters — and that's where the revenue difference lives.
The real lesson from the Jam Study
The researchers who ran that tasting table weren't telling grocery stores to carry less. They were documenting something fundamental about how people make decisions under pressure: when the cost of choosing feels too high, most people don't try harder. They leave.
Choice overload plays out across millions of eCommerce sessions every day, on stores with strong catalogs and no system for connecting the right products to the right buyers at the right moment. The merchants who've solved this didn't shrink their product range. They built systems that show each customer a different, relevant slice of it.
Most Shopify stores are still showing everyone the same slice. The conversion rate reflects it.
Ready to show every visitor fewer, better products? Try PersonalizerAI free on the Shopify App Store — AI recommendations and search trained on your catalog, live in 30 minutes.