Price gaps and the competitive set
A price gap is the difference between one item's price and another's, and it is one of the most-watched numbers in commercial retail analysis. Its usefulness depends entirely on whether the comparison item is one a shopper would actually consider instead.
A gap against an item nobody substitutes is arithmetic.
How should the competitive set be defined?
By substitution, established from behaviour rather than from assumption. Items whose sales respond to each other's price and availability belong in the set; items that share a shelf and do not respond do not.
Defining it by shelf adjacency is convenient and frequently wrong. Two items sitting together may serve different occasions entirely, and a gap between them predicts nothing.
What should the gap be measured on?
Price per equivalised unit, so that pack differences do not masquerade as price differences. A gap computed on pack price between items of different sizes is mostly a size comparison.
It should also be clear whether the comparison is on shelf price, average paid price, or base price excluding promotion. Those three gaps can move in different directions in the same period.
What does a gap predict?
Relative movement, weakly and conditionally. Gaps that widen beyond a range the category has historically tolerated tend to shift share, and gaps within the normal range tend not to.
The useful analysis is therefore historical: how has this category responded to gaps of this size before. A gap reported without that context is a fact rather than a finding.
Why do gaps mislead?
Because they ignore everything else that differs. Two items with a price gap also differ in brand, distribution, promotion and shelf position, and attributing a share movement to the gap alone ignores the rest.
They also mislead across time when one item changes pack. A gap that narrowed because a competitor shrank a pack is a different event from one that narrowed because they cut price.
How should gaps be monitored?
As a time series with the competitive set fixed, so that changes reflect prices rather than changes in what is being compared. Re-specifying the set each period produces a series that cannot be read.
Where the set genuinely changes — a new entrant, an exit — the change should be dated on the chart rather than absorbed silently.
What about promotional gaps?
They matter more than base gaps in categories with heavy promotion, because the shopper sees the promoted price. A brand with a favourable base gap and an unfavourable promoted gap is losing the comparison in the moments that count.
Reporting both is the only honest presentation, and the difference between them is often the most useful number in the analysis.
What should a brand do with the output?
Use it as an input to a pricing decision rather than as the decision. Gaps inform; elasticity and margin decide. A gap-driven price move made without a margin calculation is a share decision taken by accident.
Frequently asked questions
How do I establish the competitive set empirically?
Look at cross-item responses: which items' sales move when another's price or availability changes, using the natural variation already in the data. That analysis is noisy at item level and becomes usable when pooled across many stores and periods, and it is considerably better than assuming the set from the shelf.
Should private label be in the set?
Usually yes, because shoppers substitute toward it. Excluding it produces a competitive picture that flatters branded players and describes a market that does not exist, and it is one of the more common ways a gap analysis becomes decorative.
What gap is too wide?
There is no universal figure, and the answer comes from the category's own history. Plot share against gap over several years and the range the category tolerates becomes visible; a number borrowed from another category is a guess wearing a decimal point.
Does the gap matter if the items are not on the same shelf?
Less, because comparison requires the shopper to see both. In small-format retail, where assortment is narrow and shelf position varies by store, proximity varies and a store-level view is more informative than a market-level one.
How often should gaps be reviewed?
Monthly for monitoring and immediately after any competitor price move you become aware of. What matters more than frequency is consistency of definition, since a gap series computed differently each time is not a series.
What about regional price differences?
They are real and they mean the gap differs by market. A national average gap can conceal two regions with opposite competitive positions, which is precisely the kind of thing a national figure is bad at.
Can I use gaps to set price?
As one input. Setting price to hold a gap without regard to cost, margin or elasticity delegates your pricing to a competitor, which is a strategy only when it is chosen deliberately.
Does a favourable gap guarantee share gain?
No. Price is one driver among distribution, availability, support and brand, and a favourable gap in stores where the item is not stocked delivers nothing at all.
What is the most common reporting error?
Comparing pack prices across different sizes. It is easy to do, it looks like a price finding, and it is a size finding. Retail price measurement conventions published by statistical agencies are a useful calibration for how carefully this is usually stated.
### How should a gap analysis be structured?
One chart per competitive pair, on price per equivalised unit, with base and promoted prices separated and the set fixed. That structure makes each comparison legible and it prevents the set drifting between periods.
Add a share line where the data supports it. A gap chart without an outcome beside it describes prices; a gap chart with share beside it describes a market. Price measurement conventions published by the Census Bureau show how carefully definitions are usually stated.
Interested in NRS Insights data? See case studies or contact NRS Insights.