Retailers are preparing to spend more on AI.
That is not surprising. AI is now touching some of the most important areas of retail: customer acquisition, search, product discovery, content, service, personalisation, forecasting and decision-making.
The pressure to move is real.
But increased investment does not automatically lead to better outcomes. If budgets move faster than strategy, governance and internal capability, the risk of wasted spend increases.
That is one of the clearest signals in Validify Technologies’s latest retailer AI readiness data.
Investment is moving
In our data, 56% expect their organisation’s AI budget to increase in 2026. Only 4% expect it to stay the same, while 33% are not sure or say there is no dedicated AI budget.
So, for many retailers, AI investment is either already moving or about to move.
That creates opportunity, but also pressure.
Senior teams are being asked where AI should sit, who should own it, how much should be invested, which tools should be prioritised and how ROI should be measured. These are not easy questions. They are especially difficult when the market is noisy, use cases are still developing and many teams are experimenting before the business has a clear roadmap.
Strategy and governance are still catching up
The strategy picture is much less mature than the budget.
Around a third say they have no formal AI strategy but are exploring opportunities. Another third say their strategy is in development.
Only 17.4% say they have executive ownership and a clear roadmap.
That should make retail leaders pause.
It means many businesses are likely to be making decisions about AI investment before they have fully agreed what good looks like.
That does not mean they should stop. Waiting for a perfect strategy is not realistic. AI is moving too quickly for businesses to sit still.
But it does mean retailers may need to be more intentional.
The risk is not experimentation. Experimentation is necessary. The risk is disconnected experimentation that does not build towards anything. Activity increases, but the business is still not clear on ownership, governance, measurement or value.
ROI is still unclear for many retailers
Validify’s data also shows that unclear ROI or business case is one of the biggest barriers to adoption, selected by 35%.
This is where the conversation becomes very commercial.
Retailers are not short of AI ideas. They are short of confidence in which ideas will create value.
That matters because AI is now moving closer to the commercial engine of retail. It is no longer only about logistics, productivity or content support. It is moved into areas that influence conversion, service, merchandising, forecasting, inventory and customer experience.
If a retailer is going to invest in AI search, or customer service automation what should success look like? If a retailer is going to use AI for content enrichment, is the goal speed, quality, consistency, organic visibility, conversion, localisation, personalisation, or all of the above?
Without this clarity, ROI becomes vague. And vague ROI makes it harder to secure budget, prioritise properly – or know whether a pilot has worked.
AI needs to connect to business priorities
AI spend needs to be linked to real business priorities. Not vague innovation or pressure from the market.
Real priorities.
In our data, the highest-ranked business priority is AI-native search, discovery and conversion, followed by customer acquisition and ROI.
That tells us something important. AI is no longer sitting on the edge of the business, but moving into the areas that directly affect growth, visibility and conversion.
Other high-ranking priorities include content creation and enrichment, marketplace strategy and expansion, onsite personalisation, channel mix optimisation, loyalty and engagement, product intelligence, returns, customer data, data readiness and business intelligence.
These are not technology experiments. They are commercial issues.
That is where AI needs to earn its place.
Data and legacy systems still matter
It is tempting to talk about AI as if it sits above the existing business. In reality, it depends heavily on what is already underneath it.
In Validify’s data, legacy system integration is selected as a barrier by 41%. Poor data quality or availability is selected by 24%.
Again, that should not surprise anyone in retail.
Many retailers are working with complex technology estates, fragmented data, historic systems, platform constraints and competing operational priorities. AI may create new opportunities, but it does not remove those foundations.
If the data is poor, AI outputs will be limited. If systems don’t connect, use cases become harder to scale. And if ownership is unclear, teams can end up with pilots that work in isolation but do not embed into the wider business.
This is why AI strategy doesn’t sit separately from data, architecture and operating model conversations.
The leadership question
For senior retail leaders, the question is: “What do we need AI to help us improve?”
The answer will vary by retailer. But the discipline matters.
Start AI investment with the business problem, not the tool. It should have an owner and a clear reason for existing. A sensible view of risk, and be measured in a way that the business understands.
That does not mean every AI project needs a heavy business case before anything can happen. There is still a place for testing, learning and small experiments.
But even experiments need a direction.
Spending well will matter more than spending first
Retailers will be under pressure to show they are moving on AI. But moving quickly and spending well are not the same thing.
The retailers that create the most value may not be the ones with the biggest budgets. They may be the ones that are clearest about priorities, most honest about capability gaps and most disciplined about what they choose to test, buy or build.
AI budget is increasing. Now strategy needs to catch up.
