IKEA puts applied AI at the core of its commercial decisions with Recommend
Retail runs on decisions made months in advance. Which categories get budget, which products get attention, what the creative should say — all settled in a plan, then executed against a market that has moved on. The plan is not wrong when it is written. It is wrong by the time it lands.
IKEA now makes those decisions differently. Applied AI sits at the centre of the commercial process: reading what the market is doing, deciding what deserves attention this week, producing the work to match, and carrying what happened into the next cycle.
What changed
Over five months:
13.9× return on invested spend
72 hours from decision to live creative
41% higher click-through
6× more efficient consideration
~€20,000 saved in production cost
40 articles and 80 visuals, every other week
The decision moves from the plan to the system
The commercial decision that matters in retail is which categories deserve attention right now. It is normally made once a quarter, in advance, from a forecast.
Applied AI makes it continuously instead. Every product category is held in a live state — what is rising, what is fading, what people are looking for this week — and the catalogue is read against it. The decision is available whenever it needs to be made, not on the calendar.
A decision that arrives late is not a decision
Deciding well is only worth something if the business can act while the decision still holds.
From decision to creative live in market takes 72 hours: the category chosen, the work produced against IKEA's brand book, the pages and ads built and pushed. Every other week, for five months.
More output, lower cost
Producing more, more often, normally costs more. Here it cost around €20,000 less.
When applied AI carries the research, the selection and the assembly, extra output stops being extra expense. Cadence rises and unit cost falls.
Every cycle informs the next
What each cycle earned goes back into the system rather than into a wrap-up deck: which categories carried, which work performed, how fast a decision had to move to still be worth making.
After five months, the next decision is not a fresh guess. It is made against a record of how this market actually behaves — and that record keeps getting more useful.
What's next
The system keeps running. The market keeps moving, the decisions keep being made against it, and each cycle sharpens the next.
That is what applied AI looks like inside a commercial team — not a tool bought, but a capability gained: it reads the market, decides what to do next, does it, and proves what it earned.