How Many Options Should a Shop Offer
Retailers that cut their ranges have counted what followed, and the counts disagree. Cutting items nobody prefers did little harm. Cutting across every category cost one grocer the visits of customers who bought rarely, and the best number depends on whether shoppers already know what they want.
Answer to the question
How many options should a shop offer?
No study gives a number that holds across shops. Cutting low-preference items while keeping category space did little harm in 1998 research, but a 2005 study of a grocer's cuts in every category found customers ordered less often. Shoppers who already know what they want gain from a long range.
Key takeaways
- Susan Broniarczyk and colleagues found in 1998 that retailers could cut items without harming perceptions of the assortment, provided only low-preference items went and category space was held constant.
- Peter Boatwright and Joseph Nunes reported in 2001 that an online grocer's cuts raised sales by 11 per cent on average across 42 categories. Sharad Borle and colleagues later reported that study, with a 2004 correction note, as no change in sales: a rise of 1.5 per cent with a standard error of 1.4.
- In Borle's 2005 data, covering every category, households offered the reduced range were expected to wait about 25 per cent longer between orders and spend 4.8 per cent less per order, and rarely bought categories lost the most.
- Alexander Chernev and colleagues counted preference uncertainty among four moderators of choice overload in a 2015 meta-analysis of 99 observations; Barry Schwartz's 2016 edition concedes that for people who know exactly what they want there is no such thing as too much choice.
- Avni Shah and George Wolford reported in 2007 that pen purchases peaked at about ten options, a figure that reached this piece at second hand and comes from a stall, not a shop.
You sell things, and someone has told you the range is too long. How many options should a shop offer, and does the answer change for a menu or a product page? What happens to sales if you cut?
The advice usually traces to a table of jam. That study, and what became of it, is in Choice Overload After the Jam Study. Barry Schwartz's The Paradox of Choice carried it into general circulation, and its 2016 edition added a concession in the body text: "when people know exactly what they want, there is no such thing as too much choice" (p. 109). The same edition says the arguments of 2005 stand (p. xiii), so the concession sits beside the thesis and does not change it. A shopkeeper's version of how many options is too many has evidence of its own, from retailers that cut their ranges and counted what followed.
Cut what few choose, then check the staples
Two studies published in 1998 asked how many options could go before shoppers judged a grocery category thinner. They tested two cues, whether a favourite product is still on the shelf and how much shelf space the category keeps. Susan Broniarczyk and her co-authors concluded that retailers might make substantive reductions without harming perceptions of the assortment or store choice, as long as only low-preference items were eliminated and category space was held constant.
As long as only low-preference items are eliminated and category space is held constant. (Broniarczyk, Hoyer and McAlister, 1998)
Peter Boatwright and Joseph Nunes later summarised that earlier work as showing cuts of up to 54 per cent of low-selling items with no significant effect on perceived variety or sales.
An online grocer cut the number of items in 94 per cent of its categories, mostly low sellers, and Boatwright and Nunes studied the result in 2001. Across the 42 categories they examined, sales rose by an average of 11 per cent. Sales rose in more than two thirds of the categories, and 75 per cent of households spent more overall.
Customers welcomed the removal of redundant items.
Category sales still depended on how many items were left, and on whether the attributes that matter most stayed available, brand and flavour above all.
What household records showed four years later
Four years later a team that included both authors went back to household records from the same grocer, this time for every category in the store. The cuts ran from 24 to 91 per cent of the items in a category. The 840 households offered the reduced range were expected to wait about 25 per cent longer between orders and to spend 4.8 per cent less per order than the 378 households that kept the full range.
The cut affected how often households bought in about 74 per cent of categories.
The fall in order frequency cost more than the fall in order size.
Categories bought less often lost the most, and staples such as milk, bread and cereal lost the least.
The 2005 paper cites the 2001 study together with a 2004 correction note (Journal of Marketing 68, July, p. iv) and describes the result as no impact on sales in 42 categories: an estimated rise of 1.5 per cent, with a standard error of 1.4. The 11 per cent above is therefore the figure as first published, and 1.5 per cent is how the same authors later reported it. The correction note itself could not be obtained for this piece, so what it changed, and whether it is the source of the lower figure, is not established here.
Sharad Borle, who led the 2005 team, put the gap down to coverage. The 2001 analysis used a subset of categories and households, mostly frequently bought staples, and the new model fitted to the 2001 households but with all categories returned the new result.
Find out whether your customers know what they want
How many options help depends on the shopper. A shopper who arrives with an ideal point, the product already in mind, is helped by a large range, because it makes that product easy to find. A shopper without one has more to evaluate. Schwartz keys his concession to Alexander Chernev's 2003 paper on assortment and decision processes, and a second Chernev paper that year, on the availability of an ideal point, makes the shopper the variable. In their 2015 meta-analysis of 99 observations, Chernev and two colleagues counted uncertainty about preferences among four moderators of choice overload, alongside the complexity of the set, the difficulty of the task and the shopper's goal.
One study put a number on the peak. In 2007 Avni Shah and George Wolford of Dartmouth College reported that when buyers were offered between two and twenty kinds of pen, purchases peaked at about ten. The figure reached this piece through Schwartz and press accounts of a two-page report in Psychological Science (18(5), 369-370), whose text sits behind a paywall and was not read, and a stall of pens says little about where the peak falls for a shop with four thousand lines. Schwartz's 2016 text puts the same peak at eight to ten (p. 109).
Count visits, not just baskets
Order size moved by under 5 per cent in the grocer's data and order frequency by a quarter. A shop that tests how many options to keep by watching basket size for a week will see almost nothing. Borle's team advise comparing selected categories to limit the damage. Their study covers one city, one delivery fee and existing customers only, so it cannot say who a thinner range fails to attract.
Friction at the point of purchase is a separate cost, taken up in What Cognitive Load Costs at Checkout, and menus have their own evidence on order and middle options in The Compromise Effect on the Wine List. The studies above looked at grocery categories and a stall of pens, so for how many options a menu or a product page should carry they are a model and not a measurement.
Shortening a range shows its cost late, in how often customers visit categories nobody was watching. The research offers one guide: near-duplicates are safer to lose than the things people buy now and then. Cut the near-duplicates first, leave the staples alone, and judge how many options were safe to lose by visits per household over a quarter.
Common questions
How many options should a shop offer?
No study read for this piece gives a number that holds across shops. A 2007 pen study, known here at second hand, reported purchases peaking at about ten options, and a 2005 study of an online grocer found that cutting items in every category lowered how often customers ordered.
Does cutting the range increase sales?
Sometimes. In 2001 Boatwright and Nunes reported an 11 per cent average rise across 42 grocery categories, which later work cited with a 2004 correction note as a 1.5 per cent rise with no significant effect. A 2005 study using every category at the same grocer found shopping frequency fell, with less frequently bought categories hurt most.
How many items can a retailer remove without customers noticing?
Studies summarised in 2001 found cuts of up to 54 per cent of low-selling items did not significantly change perceived variety or sales. The 1998 study behind that finding held category space constant and removed only low-preference items.
Does choice overload apply to menus and product pages?
Not directly. The grocery and pen results describe shelves and a stall, so a menu or a product page needs its own test. Order and middle options on menus are covered in The Compromise Effect on the Wine List.
Sources
Broniarczyk, S. M., Hoyer, W. D., & McAlister, L. (1998). "Consumers’ Perceptions of the Assortment Offered in a Grocery Category: The Impact of Item Reduction." Journal of Marketing Research, 35(2), 166-176. doi:10.1177/002224379803500203
Boatwright, P., & Nunes, J. C. (2001). 'Reducing Assortment: An Attribute-Based Approach.' Journal of Marketing, 65(3), 50-63. doi:10.1509/jmkg.65.3.50.18330
Chernev, A. (2003). 'Product assortment and individual decision processes.' Journal of Personality and Social Psychology, 85(1), 151-162. doi:10.1037/0022-3514.85.1.151
Chernev, A. (2003). 'When More Is Less and Less Is More: The Role of Ideal Point Availability and Assortment in Consumer Choice.' Journal of Consumer Research, 30(2), 170-183. doi:10.1086/376808
Borle, S., Boatwright, P., Kadane, J. B., Nunes, J. C., & Shmueli, G. (2005). 'The Effect of Product Assortment Changes on Customer Retention.' Marketing Science, 24(4), 616-622. doi:10.1287/mksc.1050.0121
Shah, A. M., & Wolford, G. (2007). 'Buying Behavior as a Function of Parametric Variation of Number of Choices.' Psychological Science, 18(5), 369-370. doi:10.1111/j.1467-9280.2007.01906.x
Chernev, A., Bockenholt, U., & Goodman, J. (2015). 'Choice overload: A conceptual review and meta-analysis.' Journal of Consumer Psychology, 25(2), 333-358. doi:10.1016/j.jcps.2014.08.002
Schwartz, B. (2016). The Paradox of Choice: Why More Is Less, Revised Edition. New York: Ecco. ISBN 9780062449924