How digital performance advertising revolutionized a very established industry. Let’s meet Criteo.

It began with a chance meeting in a Paris incubator. Serial entrepreneur Jean-Baptiste Rudelle (startup record: one failure, one modest success and Criteo), was presenting his movierecommendation idea to business angels. He began chatting

with a guy working on his laptop, who said: “Check out those two guys over there. I think they’re doing something similar to you.”

Those guys were Franck Le Ouay and Romain Niccoli, who became Criteo’s co-founders. Together, the trio developed a website that allowed people to rate films and get customized recommendations. The design was slick, the technology innovative, and the results seemed pretty relevant. Problem was their traffic remained tiny, stagnating at 500 visitors per day, no matter how much they spent on advertising.

Then came the first pivot. Criteo tweaked their business model to sell their technology to companies, rather than going directly to consumers. Initial jubilation when they agreed a deal with AlloCiné, the French equivalent of IMDb, quickly wore off when it was clear their market was limited.

They tried adapting their technology to a customized recommendation app – but they couldn’t figure out a way of making money.

Investors were losing patience. In the middle of a tense discussion an independent director stepped in. “Have you thought of using your algorithm for advertising?” he said. That was the third – and finally successful – pivot.

Technologically, it was possible – but Criteo still had little revenue and they needed to raise money fast. Enter Dominique Vidal, a French VC living in London whose startup Kelkoo, recently acquired by Yahoo, had fought to impose a pay-perclick advertising model on the market. Dom understood immediately what Criteo were trying to do. He had just joined Index Ventures, whose investment was crucial.

What Criteo invented is an entirely new approach to advertising. Instead of targeting people based on what they are (their age, gender, home address, or marital status), Criteo’s algorithm concentrates on what they want (a new phone, a sofa, a beach vacation etc.). In the advertising industry at the time, it was a surprisingly revolutionary concept.

Criteo’s approach required the ability to capture shopping intent, only possible in the digital era when people starting typing key words into search engines. Using sophisticated machine learning technologies Criteo invented a way to harvest other types of consumer data that allowed them to accurately determine high-quality shopping intent in an ever-expanding range of marketing scenarios.

Criteo technology also made it possible for advertisers to make a precise link between their advertising investments and their revenues. This enabled them to optimize media in a way that wasn’t possible before.

So far, Criteo reckon they’ve captured just a small fraction of the vast universe of consumer shopping intent. They continue to invest in R&D to see just how far they can go.

If you’d like to know more about startups in France, click here to read the new Business France book about 35 success stories.

If you’d like to find out more about technology and startups in France, click here.