Online stores are using AI to run faster in the wrong direction

Generative AI is a game changer for e-commerce. Product photos, videos, social posts and entire ad campaigns used to take time, budget and a team. Now they can be made in minutes. For the price of a monthly subscription.

At the same time, everything else is getting more expensive. Advertising costs keep rising. The cost per customer keeps going up. Competition is growing. Buyers have more choice than ever. Even when campaigns sell, the profit left at the end of the day is shrinking for many online stores.

That’s the paradox. Creating content to get attention has never been cheaper. Winning a customer has never been more expensive.

AI could be the answer. It’s the most powerful tool e-commerce has ever had. Yet for most online stores, it isn’t delivering. Not because the technology falls short. But because of how it’s used.

Most stores use AI to do more of what they already did. More ads. More posts. More product videos. When everyone does that, feeds fill up with content that looks alike. Buyers’ attention doesn’t grow with it. So standing out gets harder, not easier.

Buyers are changing as well. They’re becoming more selective about which brands deserve their attention. Their customer journey is changing with it. They search. They read reviews. They ask questions in communities. And they increasingly let AI assistants recommend what to buy. No surprise, then, that Amazon wants its AI assistant to be the way people shop its store. Many consumers even pay YouTube, Spotify and Amazon extra just to avoid ads.

More and more businesses now base their strategy on what AI tells them. Most feed it their own data: Google Analytics, sales figures, campaign results and margins. That data matters. Without it, no store succeeds.

But it only shows what happens inside your store. Not what buyers search for elsewhere. Not what they discuss or what frustrates them. Not where competitors leave gaps.

Feed AI half the picture and it won’t tell you something is missing. It fills the gaps with confidence. You get a strategy that looks complete. It sounds convincing. It’s backed by your own numbers. That’s exactly what makes it dangerous. The better the plan looks, the less likely anyone questions it. Even when it points the wrong way.

The problem isn’t that online stores use too little AI. It’s that they let AI steer with only half the map.

Every signal is one perspective: search demand, social conversations, competitor activity and reviews. On their own, each tells part of the story. Combined, they show where supply and demand miss each other.

In early October, data from e-commerce intelligence startup CortLogic revealed several concrete growth opportunities in the Dutch furniture market. One stood out. Overall search demand for furniture dropped by 24%. But searches for corner sofas for small living rooms grew by 81%. Their share of all furniture searches rose by 141%.

Buyers were actively discussing corner sofas on Facebook and Reddit. Yet stores with a matching assortment were hard to find. One of the bigger brands went from zero to four organic posts about corner sofas on Instagram and Facebook in two weeks. At the same time, it put 16% of its corner sofas on sale. The average discount was 20%.

No paid campaigns were launched on Meta to capture the rising demand. And TikTok was left untouched by the major corner sofa sellers.

Comments and reviews added one more layer. Late delivery is one of the most common complaints in furniture. Buyers said they would pay more for a guaranteed delivery date.

Demand was growing in a shrinking market. The products were largely available. But the offer wasn’t in the right place at the right time. And the strongest move wasn’t another discount. It was adjusting the offer with a delivery promise and showing it where the audience already was, through organic content or paid campaigns.

Every niche has gaps like this. AI can help you find them and turn them into growth. Your own store data should stay leading. But make sure your AI also understands the market: your blind spots, and the chance to meet demand at the right time, in the right place, with the best offer. Giving AI that missing view of the market is exactly what CortLogic was built for.

When stores can’t stand out, price becomes the easy weapon. But only one store can be the cheapest. Every discount buys a sale with margin. And so the race to the bottom in e-commerce continues.

Some stores make that trade on purpose. They give away margin for volume, new customers or first-party data. In categories like online coffee, a customer often only becomes profitable after their third or fourth order. A discount that wins a loyal customer can pay off. A discount that wins a one-time bargain hunter rarely does.

Buyers have grown wary too. They’ve seen online stores raise their prices before a sale. Then they offer a “discount” that isn’t one. A fake deal might win a sale once. But it costs exactly the trust you need for the next order.

In a market where you can’t always be number one in search or the cheapest, growth comes from seeing what others miss. AI can make that possible. Or it can speed up the race to the bottom.

Let’s not forget the first lesson of economics. Supply has to match demand. If you use AI to grow your business, give it the full picture. Otherwise, you could just be running faster in the wrong direction.

Erik Touwslager is a tech entrepreneur with over 16 years of experience in sales, marketing and media. As a commercial leader at Sprout and (show all) Erik Touwslager is a tech entrepreneur with over 16 years of experience in sales, marketing and media. As a commercial leader at Sprout and The Next Web, he knows the ecosystem and the challenges of startups and scale-ups inside out. Today he is the founder of CortLogic, an e-commerce data intelligence platform that reveals where demand and supply in a market miss each other, and which hidden opportunities online stores can seize to grow.

Get the most important tech news in your inbox each week.

Thanks! Check your inbox to confirm your subscription.

By subscribing you agree to our Privacy Statement.

Expertise from selected TNW Council members, admitted through an application and review process to a fee-based program. Opinions expressed by the authors are their own.