Skip to main content

Optimizing Products Individually or in Groups

Whether to let the AI price each product on its own, or optimize related products together as a set.

When you set up a Price Optimization strategy, one of the choices in the Optimization settings tab is whether products are optimized individually or in groups. This article explains the difference and when to use each.

Individual optimization (default)

Each product is optimized on its own. The AI finds the single best price for each product based on that product's own demand pattern. This is the right choice for most catalogues, because most products behave differently from one another.

Group optimization

Products that belong to the same group are optimized together. Instead of pricing each one separately, the AI finds a single price change that best achieves your objective across the whole group, and every product in the group receives the same percentage adjustment.

When to use group optimization

Use it only when the products are true substitutes, the same item in different sizes or colours, where a customer would happily switch from one to another based on price alone. If your customers treat the blue and the red version as interchangeable, group optimization keeps them moving together.

When not to use it: avoid grouping products that have meaningfully different demand, costs, or customer segments, even if they sit in the same category. A €30 basic t-shirt and a €90 premium jacket shouldn't be group-optimized just because they share a product line: their ideal prices have nothing to do with each other.

How this differs from Pricing Groups

It's easy to confuse the two, but they solve different problems:

  • Group optimization (this article) decides the price during optimization: the AI reasons about the products together and finds one adjustment for the set.

  • Pricing Groups align prices after optimization: the AI prices products first, then the system forces the group to a single final price. See Pricing Groups – User Guide for that feature.

You can use either, both, or neither depending on how tightly your variants need to move together.

Did this answer your question?