Ideal Point

What it is

The term describes a shopper's preferences as a point among the available options: the combination of features that would suit them best. A shopper with an ideal point is looking for something. A shopper without one is looking to find out what they like.

Alexander Chernev made the difference a variable in two 2003 papers, one in the Journal of Personality and Social Psychology on product assortment and individual decision processes, and one in the Journal of Consumer Research titled 'When More Is Less and Less Is More: The Role of Ideal Point Availability and Assortment in Consumer Choice'. The second title states the claim. Whether more is less or less is more depends on whether an ideal point is available.

Barry Schwartz carried the idea into the 2016 revised edition of The Paradox of Choice, in a sentence added to the body text and keyed to Chernev's 2003 paper: when people know exactly what they want, there is no such thing as too much choice, because large choice sets let them find exactly what they want and small ones often do not (p. 109).

In effect

The practical effect is that the same range can serve two customers in opposite ways. A customer reordering a known product benefits from depth, because the product is there and easy to find. A customer browsing with no fixed preference benefits from a range they can compare, and may defer the choice if the range is too large to compare.

Chernev and two colleagues reached a related result at the level of the whole literature. Their 2015 meta-analysis of 99 observations counted preference uncertainty among four moderators of choice overload, alongside the complexity of the choice set, the difficulty of the decision task and the shopper's goal. Uncertainty about preferences is, in effect, the absence of an ideal point.

For a retailer the implication is about audience. A range built for people who know what they want and a range built for people who are looking are different ranges, and a cut that suits the second may cost the first.

What it does not say

It does not give a number. Shah and Wolford's 2007 pen study is often quoted for an optimum near ten options, and Schwartz gives eight to ten, but that figure comes from a single stall of pens and tells a shop with thousands of lines very little about its own peak.

It does not say that a shopper either has an ideal point or does not. Preferences can be partly formed, and a shopper may know the brand but not the size.

It does not say the idea has been tested in every setting. The detail of Chernev's 2003 experiments was not read for this page, which relies on the titles, the 2015 meta-analysis and Schwartz's account of the point.

And it does not say that large ranges are harmless to people who know what they want. It says they are not the problem for that group, which is a narrower claim.

Telling the two shoppers apart

A shop rarely knows in advance which kind of shopper is at the shelf, but it can look for signs. Repeat purchases of a named product point to an ideal point, because the shopper is back for the same thing. Browsing across brands, sizes or styles points to its absence. Search terms on a website, and what customers ask staff, say the same thing in a different form. These are practical readings and not findings of the studies cited here, none of which tested them.

The distinction also explains why one range can draw opposite complaints. A regular who cannot find a usual product reports that the shop has too little. A newcomer facing the whole range reports that it has too much. Both reports can be accurate about the same shelf.

Where it sits in the literature

The Chernev 2015 meta-analysis is the nearest thing to a map of the whole field. Its four moderators describe when a large range is likely to hurt, and preference uncertainty is one of them. Seen that way, the ideal point is not a rival account of choice overload but one of the conditions under which it appears.

That is also why the idea is easier to apply than to prove. The conditions can be described in a sentence, but the strength of each is not given here, and the page does not claim to know how much any one of them moves sales in a shop.

---