Selling more to those who already bought is cheaper than acquiring new customers. Upselling (a better product) and cross-selling (a complementary product) work when they appear at the right moment, with real relevance. Done badly, they look like insistence and harm trust.
The essentials in a definition
Upselling and cross-selling with ads increases each customer’s value, but depends on the right moment and relevance: suggesting a useful complement after purchase or at decision time is different from pushing products. Communication must be honest (UCPD), measurement tracks average value and margin (key events) and Nielsen Norman Group usability research guides how suggestions are presented.
Difference and logic
Upsell: offering a higher or bigger version than the customer is considering. Cross-sell: suggesting products that complement the purchase. Both work when they help the customer get more value, and fail when they only raise the basket value.
Help first, sell later.
The right moment
In the cart or checkout (low-friction add-ons), right after purchase (add-on offer), during use (when the customer sees the need), at the end of the product cycle (replenishment, upgrade). Choose the moment according to the product. Avoid offering too much on the first interaction.
A good moment depends on context.
Upsell and cross-sell with ads
- Help the customer get more value
- Choose the right moment
- Recommend based on what was already bought
- Measure margin and satisfaction
Ads and data
Use purchase data to recommend relevant complements by product and segment. Catalogue remarketing for complementary products to buyers. Messages by email and in the account. Test bundles with a clear benefit. Respect consent. Avoid ads repeating the product already bought.
Relevance comes from what the customer already bought.
A good upsell helps the customer end up with more than they were looking for.
Measuring with margin
Measure average order value, acceptance rate, incremental margin, returns and satisfaction. Beware of offers that slow checkout or generate regret. Compare test groups. Adjust offers by segment and monitor customer perception.
More revenue with less satisfaction is a bad business.
Decision table
| Moment | Suggestion | Caution |
|---|---|---|
| In the cart | Low-cost, useful complement | Do not overload the decision |
| Right after purchase | Compatible accessory or service | Do not confuse with confirmation |
| When the product is in use | Replenishment or upgrade | Wait for the right moment |
| Loyal customer | Higher version or bundle | Clear value |
Common mistakes to avoid
- Suggesting irrelevant products. It irritates and reduces trust.
- Hiding the additional cost. It is misleading.
- Measuring only revenue. It ignores margin and returns.
How to measure the result
Track average value per customer, suggestion acceptance rate, margin and returns.
Limits and exceptions
Results depend on catalogue and moment, and excessive suggestions harm the experience. Test with control groups and track customer feedback.
Illustrative example
Illustrative example: a photography store suggests, in the cart, a compatible memory card and, two weeks after purchase, a camera bag, by email and ads to customers with consent. It measures average value, margin and returns, and removes suggestions that do not work.
In practice: checkpoints
Analyse the products most often bought together and create suggestions based on data, not hunches. Test the position and number of suggestions, and remove those that reduce conversion of the main product. Ask for reviews after purchase to collect feedback on complements. Offer real value: an accessory that improves use sells better than a random one. Compare margin and return rates for suggested items, and avoid pushing products with high returns. Make the suggestion clear and optional, show the total price, and keep follow-up messages short. Review the performance of every rule each quarter, and drop the ones that do not add measurable value for customers or for the business.
Frequently asked questions
What is the difference between upsell and cross-sell?
Upsell: higher version; cross-sell: complementary product.
When to offer?
In the cart, right after purchase or at the moment of need.
How do I measure?
By average value, acceptance, margin and satisfaction.
How do I choose what to suggest?
With co-purchase data and real relevance to use.
When should I avoid it?
When the customer is frustrated or the main product has problems.
Upsell and cross-sell at the right moment, with useful suggestions, clear price and measurement of value, margin and returns.
Sources and recommended reading
- Unfair Commercial Practices Directive (2005/29/EC)
- Analytics Help: key events
- Nielsen Norman Group: usability articles
Last editorial review: October 2026. Informational content; always confirm current conditions in the official sources before deciding.