Product count alone does not determine purchasing outcomes
When customers say there are too many products to choose from, a sales manager has two possible responses. Reduce the range, or add information that helps customers judge the differences. The reason for their hesitation determines which change deserves a trial.
Choice overload describes outcomes such as postponing a decision or feeling less satisfied when more options are available. Research finds that the effects vary across experiments, with factors such as comparison complexity helping explain the differences. Consider the range and the information needed to compare it together. (2010 choice overload meta-analysis, 2015 meta-analysis)
Before removing products, investigate whether customers cannot find a suitable item or cannot judge the differences between suitable items. This article uses the research to develop practical changes to test on your own product pages.
A meta-analysis statistically combines results from multiple experiments. Scheibehenne and colleagues' 2010 paper covered 63 comparisons from 50 published and unpublished experiments involving 5,036 participants. The average effect of increasing choice was close to zero, with substantial variation between experiments. That average cannot establish that assortment size has no effect in every store. (2010 meta-analysis)
Chernev and colleagues' 2015 paper analyzed 99 observations from 53 studies involving 7,202 participants. Four factors moderated choice overload: the complexity of the assortment, the difficulty of the decision task, uncertainty about preferences, and a goal of minimizing decision effort. (2015 meta-analysis)
The papers used different study collections and analytical approaches. Their difference does not measure a change in consumer behavior between 2010 and 2015. Both combine experimental settings rather than measuring the actual purchase rate of Japanese online stores.
For a business, the useful implication is to investigate the source of decision effort before adopting a smaller range. Neither paper establishes a universally optimal product count.
Match the source of hesitation to a specific change
Identifying what customers cannot compare gives you changes to consider alongside removing products. The following proposals draw on the four factors in the research. Their effects require testing in your own business.
| Where customers get stuck | Change to test |
|---|
| Product descriptions use different attributes, preventing performance comparisons | Add a comparison table using the same attributes and units |
| Finding an affordable item requires opening every product page | Show prices and additional charges in the listing |
| Customers do not know which specifications their use requires | Explain the relationship between uses and required specifications |
| Choosing a product becomes entangled with browsing the entire catalog | Offer entry points by use while keeping a link to the complete catalog |
For example, a printing service might describe paper thickness differently for each stock. Showing thickness in one unit and explaining the resulting finish would allow a direct comparison. This trial changes how existing products can be compared before deciding which ones to retain.
When considering the removal of products with overlapping specifications and uses, investigate what their existing buyers could purchase instead. Differences in color, size, or device compatibility can matter to particular buyers. Low sales alone can conceal a specification that some customers require.
Distinguish changing the display from removing products
Featuring fewer items in a listing and discontinuing them change the available range in different ways. Record how many products are displayed and how many remain available as separate decisions.
A trial that introduces entry points by use can retain the complete catalog. Check whether visitors can reach the specifications they need. If reports of missing products increase, revise the categories or link labels.
A discontinuation trial requires reviewing purchase histories and the uses mentioned in inquiries, then checking which requirements alternatives can fulfill. Products serving uses without a suitable replacement represent demand that a reduction could lose.
Simplifying a listing while simultaneously discontinuing products combines the effects of lower comparison effort and a reduced range. Define whether the trial is evaluating presentation or assortment reduction.
A higher purchase rate can accompany lower revenue
Evaluate order value and profit alongside the proportion of visitors who purchase. In the hypothetical example below, purchase rate means the share of people visiting the comparison page who make at least one purchase within a predefined observation window. Return visits and multiple orders do not add extra people to either count.
Calculation example with hypothetical figures
| Metric | Current display | Revised display |
|---|
| People visiting the comparison page | 1,000 | 1,000 |
| Buyers | 40 | 50 |
| Purchase rate | 4% | 5% |
| Total revenue | ¥400,000 | ¥350,000 |
| Revenue less product costs | ¥160,000 | ¥140,000 |
Purchase rate rises by one percentage point, while revenue and revenue less product costs fall. These figures exclude expenses such as order-related shipping. Include those costs when assessing the business result. Adopting the change based solely on purchase rate would overlook the decline in profit.
The table alone also leaves the cause unresolved. One proposed test would randomly assign visitors during the same period to the current or revised display, keeping each person's assigned version consistent. Define the metrics, purchase observation window, required sample size, and end date in advance, and assess random variation. A comparison of different dates also captures changes in advertising and seasonality.
A decision to reduce the range requires examining both missing comparison information and the uses a reduction would remove. Use consistent descriptive attributes across your product pages, and decide whether to change the number featured or the number sold. Judge continuation using purchase rate, revenue, and the amount remaining after relevant costs.
Publish the uses, prices, and specifications that explain how to choose your products or services. Recording when and how a comparison table changed also makes the next trial traceable.
SpreadSite is a no-code website builder that lets you create, publish, and update a website from Google Sheets using ready-made templates.
To build a website explaining how your products or services differ in use and price, give SpreadSite a try!
Build a website with SpreadSite →
