Target is a very important chain in the United States that is not doing very well also because of its political choices: in the last twelve months, almost 4 out of 10 consumers in the United States have boycotted a brand.
The real advantage today is not having more data. It is using it before others
Target, one of America’s largest retail chains, has invested $6 billion this year in shops, people and technology. The reason: after years of declining sales, cluttered shelves, high prices and unattractive products, the company has decided that recovery comes through artificial intelligence.
The interesting thing is not the investment itself, but where they are applying it.
The first tool is called Target Trend Brain. It is an AI system that analyses fashion shows, social media and industry reports to identify trendy colours, fabrics and cuts. Designers question it in natural language: “what colours will work next season?” and get answers with visual proposals. What used to take weeks of research is now done in a few hours. In this way, the retailer is able to get trendy products onto the shelves much faster than competitors.
The second area is demand forecasting. The company is using advanced predictive analysis models to estimate the demand for all products and make sure that the right product is in the right shop at the right time. Currently the models work well for products with constant demand, such as cereals, but for seasonal or trend products the challenge is more complex, and that is where they are concentrating their effort.
The third element is the one that impressed me the most: Target has invested in ‘synthetic audiences’, digital profiles built from in-depth interviews with real customers. In practice, they can test their audience’s reactions to a new product or a change in style before they even put it on sale. This is the same approach I talked about in a recent post on the digital twins used by CVS.
Underlying all these investments is the same logic: the advantage is not in having more data, but in being able to use it earlier to make better decisions. Understanding earlier what trends are coming, predicting earlier where demand will move, testing earlier how customers will react.
And this is where the case becomes interesting for an SME as well. It does not have to replicate the tools of a large chain. It has to replicate its method.
On a different scale, it means understanding in advance which customers are slowing down, which products are losing rotation and where the margin is eroding.
Today, this can also be done with accessible, readily available and affordable tools. You don’t need billion-dollar investments. You need to use data in good time, when it can still change a decision.
In your companies, does the data arrive in time to decide or in time only to explain?
Post by Antonio De Bellis
