We were at an event recently, and heard one of our customers say a phrase that stopped us in our tracks:
“AI is a magnifying glass, not a crystal ball.”
Ann Kerry, Head of Merchandising
It is simple, memorable and we wish that we had thought of it first. But it actually came from Ann when we were discussing the use of AI in merchandising, and it captures exactly where the conversation around AI in fashion retail needs to go next, in one short sentence.
AI is everywhere right now.
It is on every stand at events, in every product demo, mentioned in every strategy deck and, if we are being honest, is probably being squeezed into a few conversations where it doesn’t belong. The excitement is understandable, because AI promises faster analysis, smarter forecasting, better planning, and fewer missed opportunities. This is undeniably compelling in an industry where timing, stock, margin and customer relevance can make or break a season.
But the danger comes when AI is positioned as something it is not.
AI is not a retail fortune teller. It can not sit in the corner of the merchandising team and whisper next season’s bestseller into someone’s ear. It cannot guarantee what customers will want, how culture will shift, what the weather will do, which product will go viral, or whether a trend will still feel relevant by the time it lands in store.
And it is important to remember that that is fine, because the real value of AI is not that it magically predicts the future, it is that it can help retailers see the present more clearly.
The problem with the crystal ball view of AI
Fashion retail has always been obsessed with prediction. Which category will grow? Which colour will sell? Which store needs more depth? Which product will drive margin? Which trend is worth backing? Which range is going to feel right to the customer? These are the questions merchandising, buying and planning teams have always had to answer. AI and predictive analytics can absolutely support those decisions. They can process huge amounts of data, identify patterns, highlight anomalies and suggest likely outcomes based on what has happened before.
But “likely” is not the same as certain. A forecast is still built on data, assumptions and context. If the underlying data is incomplete, distorted or misunderstood, the answer can be misleading. If availability was poor, demand may be understated. If a product landed late, performance may look weaker than it really was. If a store was under-allocated, the sales history may reflect a stock issue rather than a customer issue.
This is why treating AI as a crystal ball is risky. It can create false confidence. It can make teams believe the answer has been “calculated”, when really it still needs to be interpreted, and in retail, interpretation matters.
A model can show what happened. It can suggest what might happen next. But it cannot fully understand brand direction, customer emotion, creative intent, competitor behaviour or the commercial judgement behind a brave decision.
That is where people still matter.
Why the magnifying glass mindset is more useful
A magnifying glass does not invent what is there. It helps you examine it more closely, which is a much better way to think about AI in merchandising.
Used well, AI helps retailers move beyond surface-level reporting and into deeper commercial understanding. It can help teams identify where performance is being driven by genuine demand versus where it is being shaped by stock constraints, markdown activity, poor ranging decisions or allocation issues.
It can help merchandisers ask sharper questions, such as:
- Why is one region responding differently to another?
- Where are size curves behaving unexpectedly?
- Which products are driving volume but eroding margin?
- Which stores are being held back by availability?
- Are we repeating last season’s mistakes because the insight is buried across too many spreadsheets?
This is where AI becomes powerful, not as a replacement for merchandising expertise, but as a way to focus it. The best teams don’t need AI to make decisions for them, they need it to help them see the trade-offs, risks and opportunities. They need it to reduce the time spent hunting for answers and increase the time spent deciding what to do about them.
That distinction matters, because better data does not automatically create better decisions. Better data creates the conditions for better decisions and judgement still has to come from people who understand the customer, the brand, the product and the commercial reality of the business.
The future is human judgement, sharpened by AI
There is often a false divide in retail between being ‘data-led’ and being ‘instinct-led’, but in reality, the strongest retailers are both.
Instinct without data can become bias, and data without instinct can become noise. AI has an important role to play in challenging assumptions because it can show when a long-held belief is no longer true and it can reveal hidden risks or uncover missed opportunities. It can connect planning, buying, allocation and trading decisions in ways that are difficult to do manually, but it shouldn’t flatten the human judgement that makes great fashion retail distinctive.
Customers do not buy products simply because a model predicted they would. They buy because the product feels right, the timing feels right, the price feels right, the brand feels relevant and the experience connects. AI can help retailers understand more of that picture, but it cannot own it. The opportunity is not to remove people from the process, it is to give them a sharper lens.
For many merchandising teams, too much time is still spent gathering data, reconciling spreadsheets and trying to work out whose version of the truth is actually true. By the time the insight arrives, the opportunity may already have moved on.
AI can change that. It can help teams move faster from ‘what happened?’ to ‘what should we do next?’ It can support scenario planning, demand sensing, allocation decisions, markdown strategy and range optimisation. It can help retailers explore options earlier, test decisions more intelligently and respond with greater confidence.
But the key word is support. AI should inform, challenge and improve decisions. It should not pretend to be the decision-maker.
A sharper view of retail decision-making
The retailers that get the most value from AI will not be the ones who simply add it to every process and hope for transformation. They will be the ones who are clear about where AI genuinely improves decision-making:
- They will use it to see patterns faster.
- They will use it to ask better questions.
- They will use it to challenge assumptions.
- They will use it to connect insight across teams.
- They will use it to give merchandisers, buyers and planners more confidence in the decisions they make.
That is the real opportunity for AI in fashion merchandising. It’s not magic, hype or a crystal ball. It’s a magnifying glass, which in an industry where small decisions can have major commercial consequences, seeing clearly is a very powerful thing.
So perhaps the question for retailers is not, ‘Can AI predict the future?’, it should be ‘Can AI help our people make better decisions today?’
That is where the value and competitive advantage is, and that is where the future of intelligent merchandising should be focused.
Frequently asked questions (FAQ)
What is the role of AI in fashion merchandising?
AI can help fashion merchandising teams analyse data faster, identify patterns, highlight risks and support better decision-making. Its role is not to replace merchandisers, buyers or planners, but to give them clearer insight into demand, stock, margin, allocation and customer behaviour.
Can AI predict future fashion trends?
AI can support trend analysis by identifying patterns in historical sales, customer behaviour and market signals, but it cannot predict the future with certainty. Fashion trends are influenced by culture, timing, weather, social media, brand perception and customer emotion, which means human judgement remains essential.
Why should retailers avoid treating AI like a crystal ball?
Treating AI like a crystal ball can create false confidence. Forecasts are only as good as the data, assumptions and context behind them. If stock availability, late product launches, markdown activity or allocation issues distort the data, AI outputs still need to be interpreted by experienced retail teams.
How can AI improve retail decision-making?
AI can improve retail decision-making by helping teams move from surface-level reporting to deeper commercial understanding. It can show where performance is being driven by genuine demand, where stock constraints are limiting sales, which products are protecting margin and where teams may need to act faster.
Will AI replace merchandisers and planners?
AI is unlikely to replace the expertise of merchandisers and planners. The strongest use of AI is to support human judgement by reducing manual analysis, challenging assumptions and helping teams make faster, more confident decisions. In fashion retail, customer understanding, brand direction and commercial instinct still matter.
