Penetration is the proportion of category buyers who bought your brand at least once in a period. Repurchase — or repeat rate — is the proportion of those buyers who bought it again.
They are the two components of a brand's customer base, they answer different questions, and they respond to completely different investments. Treating them as one thing, or optimising the wrong one, is among the more expensive mistakes in category strategy.
The two questions#
| Penetration | Repurchase | |
|---|---|---|
| Question | How many people buy us? | How often do they come back? |
| Responds to | Availability, awareness, trial, distribution | Product experience, price fairness, service |
| Improved by | Reaching new buyers | Fixing reasons not to return |
| Typical variance between brands | Large | Comparatively small |
| Where growth usually comes from | Here | Rarely here |
Why penetration usually dominates growth#
Within a category, brands of different sizes tend to have fairly similar repeat rates, while their penetration differs enormously. The observable difference between a large brand and a small one is mostly how many people buy it, not how loyally those people buy.
This has a direct consequence for planning. A growth plan built on deepening loyalty among existing buyers has a low ceiling: there are only so many existing buyers, and their purchase frequency is largely set by the category's natural rhythm rather than by how much they like you. A plan built on reaching more buyers has a much higher one.
The honest exception: where a specific product problem is suppressing repeat below what the category supports. That is a real and urgent situation — but it is a repair job, not a growth strategy, and it should be recognised as such. Diagnosing it is what Review Mining is for.
Reading repeat rate without a benchmark is meaningless#
A repeat rate has no absolute interpretation. It depends entirely on the category's replacement cycle.
- Long replacement cycles — appliances, durables: a low twelve-month repeat rate is normal and carries no information about satisfaction. Consumers are not repurchasing because they do not need to.
- Replenishables — household, personal care, consumables: the same figure is a warning, because buyers who were satisfied should have returned within the window.
- Seasonal categories: the window has to align with the season, or the measurement captures the calendar rather than the behaviour.
The only useful comparisons are against the category norm and against direct competitors over the same window. Any repeat figure quoted without one of those is decoration.
What suppresses repeat#
When repeat sits materially below the category norm, the cause is usually one of a small set, and review text finds it faster than anything else:
- Expectation gap — the product does not do what the marketing implied. Highly visible in reviews, and typically the largest single cause.
- A specific variant problem — one size, shade or format failing while the rest perform. Invisible at product level; see SPU vs SKU.
- Price fairness — buyers who purchased at full price and then saw a deep discount often do not return. Frequent in heavily promoted categories.
- Availability at the moment of reorder — the buyer intended to return and could not, then substituted and stayed substituted.
- A better alternative — visible in the comparisons consumers draw in reviews and consideration-stage content.
Only the last is a genuine competitive loss. The first four are self-inflicted and fixable, which is why diagnosing before investing matters.
The interaction with price position#
Price position shapes the two metrics in opposite directions.
Lower price positions typically buy wider penetration and weaker repeat — they attract deal-driven buyers who move to the next offer. Higher positions typically show narrower penetration and stronger repeat — the buyer chose deliberately and is less price-mobile.
Neither pattern is inherently better, but a brand should know which one it has. A promotional strategy that widens penetration while diluting repeat can raise sales and lower the value of the customer base simultaneously, and a sales-only view will report that as success. Reading the two metrics alongside Price Band Analysis is what makes the trade visible.
The same trap appears around promotional events: customers acquired cheaply during a peak are the ones least likely to return, so event performance should always be read with the subsequent repeat behaviour attached. See Singles' Day (Double 11).
Measurement caveats#
Penetration and repeat are panel-native metrics — they require knowing that the same buyer purchased twice, which observed marketplace data does not directly establish. Where they are derived from observed data rather than a panel, they are estimates, and the estimation method should be stated. This is a specific instance of the observed-versus-modelled boundary discussed in Market Intelligence Data Sources: both are legitimate, and a buyer is entitled to know which one they are being shown.
Where to look next#
For diagnosing why repeat is low, see Review Mining and Consumer Insights. For the price-position interaction, see Price Band Analysis. For promotional-event effects on customer quality, see Singles' Day (Double 11). For the sales metric these sit beneath, see GMV.
Common questions#
What is the difference between penetration and repurchase?#
Penetration is the proportion of category buyers who bought your brand at least once in a period — a measure of how many people you reach. Repurchase, or repeat rate, is the proportion of those buyers who bought you again — a measure of how well you hold them. The two answer different questions and respond to different investments: penetration responds to availability, awareness and trial, while repurchase responds to product experience, price fairness and service. A brand can be excellent at one and poor at the other, and the remedies have almost nothing in common.
Which matters more for growth?#
Penetration, in most cases, and by a wide margin. Brands of different sizes within a category tend to have fairly similar repeat rates, while their penetration differs enormously — so the observable difference between a large brand and a small one is mostly how many people buy it, not how loyally. The practical consequence is that a growth plan built on raising loyalty among existing buyers usually has a much smaller ceiling than one built on reaching more buyers. The exception is where a specific product problem is suppressing repeat below what the category supports, which is a repair job rather than a growth strategy.
What does a low repeat rate actually indicate?#
It depends on the category's natural rhythm, which is why the figure means nothing without a benchmark. In categories with long replacement cycles a low repeat rate within a twelve-month window is normal and says nothing about satisfaction. In replenishable categories the same figure is a warning. The useful comparison is always against the category norm and against direct competitors over the same window, never against an absolute standard. Where repeat sits materially below the category norm, review text is usually the fastest route to the reason.
How do these metrics interact with price position?#
Price position tends to shape penetration and repurchase in opposite directions. Lower price positions typically buy wider penetration and weaker repeat, because they attract deal-driven buyers who move on to the next offer. Higher positions typically show narrower penetration and stronger repeat, because the buyer chose deliberately. Neither pattern is inherently better, but a brand should know which it has — a promotional strategy that widens penetration while diluting repeat can raise sales and lower the value of the customer base at the same time.