“You should never forget that we are a tech company,” insists Benoît Fouilland, the CFO of Criteo, an online advertising company that tries to help its clients get clicks from potential customers.
True, he acknowledges, the Paris-based firm’s business is digital “performance marketing” — a term that refers to Internet marketing and ad programs in which advertisers and marketing companies like Criteo get paid when a specific action is completed, “such as a sale, lead, or click,” according to the Performance Marketing Association.
Moreover, the company, a foreign private issuer on NASDAQ preparing to become a U.S. domestic issuer on January 1, 2016, touts its competitive advantage in its niche industry much more than a tech company would. In a recent public filing, for instance, Criteo promoted its ability “to charge our clients only when users engage with an advertisement we deliver, usually by clicking on it.”
In comparison, “traditional display solutions typically charge clients when an advertisement is displayed, whether or not the advertisement is seen or clicked on by a user,” according to the company.
Nevertheless, Fouilland likes to hearken back to the early days of the company, when it looked a whole lot more like a tech startup than it does today. After its founding in 2005, the firm spent the first four years on research and development, according to VentureBeat.
“We had a company with smart engineers that had started this company without knowing initially that they would end up in online advertising,” the finance chief told CFO. “Our founders were passionate about machine-learning technology, and what they had in mind was not exactly to deliver performance advertising.”
Still, out of those beginnings arose what the company considers its crown jewel: the Criteo Engine. The machine-learning technology system, which consists of algorithms that yield predictions and recommendations, is supported by hardware and software that enable it to operate in real time. “The accuracy of the prediction and recommendation algorithms improves with every advertisement we deliver, as they incorporate new data, while continuing to learn from previous data,” according to the company filing.
What the engine learns from the scads of consumer data it gobbles up is “whether to buy each ad impression based on the likelihood that a user will both click on an ad and then purchase [the advertised item] on the advertiser’s site,” according to a Criteo press release. The engine then buys the advertising — often a banner ad — and delivers it to the user in under 150 milliseconds.
Keeping the engine purring is a big part of Fouilland’s job. One of the most significant risks the company faces is the potential failure to accurately predict user engagement with the display ads (often banner ads) it buys. Fouilland needs to make sure the technology is adequately funded and dynamic enough to keep up with the constantly changing web. If the engine fails, then Criteo’s cost-per-click pricing model will spawn lower revenue.
“From a CFO standpoint, [my job] is more of a technology challenge than anything else, because we need to have a technology platform and all of the software around that technology platform that enables [our] real-time pricing,” he says.
Indeed, Fouilland, who joined Criteo in March 2012, in time to help guide the company through its October 2013 initial public offering, sees helping the company innovate at a fast enough pace to match its impressive sales growth — and compete in what he considers a highly “disruptive” industry — as one of his biggest challenges.
Since its IPO, the company’s revenue has risen from $358 million in 2012 to $612 million in 2013 to $902 million at the end of 2014, according to its U.S. public filings. During an interview with CFO in late September, Fouilland spoke about that and other aspects of his job and of his company. He also sounded off about Apple’s mid-September introduction of ad blockers. An edited version of the conversation follows.
As the finance chief, what’s your view of how the company generates revenue and prices its service?
We are driving a way of spending for our customer which is very repetitive. We have very sticky customers who are spending with us because they want to generate sales. That creates some recurring revenue flows and gives a good level of predictability to our business.
Our pricing, which is based on the click, is in fact decided by the client. Digital marketers who spend on [Internet searches] are very used to this because this is the way they’re spending on Google. Our clients choose how much they are ready to pay for each click each time we generate engagement of a prospect on the banner.
There is no pricing discussion per se with the client. Clients can change at any time on their interface [with Criteo] based on how much they are ready to pay per click. Depending on their objective of return on investment, they would increase or decrease their cost per click.
To what extent does Apple’s introduction of an ad-blocking feature on iPhones and iPads represent a risk to your company?
There’s probably 5% to 6% of people using ad blockers today. That’s a small group. But this is something that needs to be monitored very closely. It’s important to look at the reason people are using ad blockers. The use of ad blockers is, first, a reaction that shows very clearly that the users are not satisfied. They have had enough of intrusive, non-relevant ads. I’m sure that you’ve been through the experience of having to watch a preroll video while you wanted to access some content on YouTube, or gotten a popup on your screen that you can’t stop. That is the primary root cause of the usage of ad blocking.
We are all about relevant, non-intrusive ads, because that is part of our business model. We want to intimately engage the user and to generate engagement through relevant ads. But we are against the type of format that could generate and create the usage of ad blockers.
The ad blocker is not a perfect solution to the root cause of the problem. When you use an ad blocker, it’s pretty much black and white today. You’re blocking the entire contents, which means that you are breaking the well-established balance of getting free access in exchange for [viewing] some advertising. Our position is very clearly that the industry needs to take the right steps to ensure that this balance is reestablished. And there are various things that can be done for that by publishers. At Criteo we believe that it’s through working within our ecosystem to promote the use of formats.
What company risks keep you up at night? What’s your worst nightmare about the company’s fortunes?
Since I joined the company, it is roughly more than five times larger in terms of revenue than it was at that time. We also have more than four times the number of people. One of the big challenges is just executing on that growth. You already have a company moving that fast that is now being transformed into a public company with lots of exposure globally to a larger advertiser base and larger financial markets. Execution is probably not my worst nightmare, because that’s what’s exciting about working for the company. But it’s definitely what keeps me up at night. To succeed in growing at that pace, you need really to be attentive to multiple things at the same time, because execution has multiple aspects. And clearly the CFO has a critical role to play in a fast execution phase of the company like the one that we are going through and have been going through for many years.
From the technology standpoint, it’s a very important aspect of our competitive advantage to be able to continuously invest. You need to allocate your resources smartly to maximize the impact of the investments.
What are finance’s challenges in terms of executing on that growth?
One is international expansion. We have to open new subsidiaries and new offices in multiple countries. We have a direct presence in 18 countries and we’ve got 27 offices in those 18 countries. With more than 1,600 people today, given the global size that we are, that means a lot of orchestration and coordination. The first critical decision that you make in a geography is the management team. We’ve seen a very clear correlation between the quality of the initial management team in each country and the rapidity of ramping up revenue in each country. We took some steps within my team to set up a PMO, a Project Management Office, to manage the coordination of opening new subsidiaries and new offices. We have to do that sometimes in geographies that have a fair amount of complexity. We opened a subsidiary in China a bit more than a year ago, and we clearly had initially underestimated the administrative complexity of setting up a business there.
How is becoming a public company changing your corporate culture?
We had a clear conflict around material nonpublic information, for example. One of the big changes when you move from private to public is that you need to restrict the access to certain information, especially anything which is nonpublic and can be material to the market. On the other hand, we have a very open, structure and a very rapid, agile type of execution. Our R&D teams have always been used to full access, real-time information on the platform that enables them to derive at any moment the full view of what the company’s revenue is. So a large portion of the R&D team potentially has at any time access to material nonpublic information. And that created a tension between the need for restricting this access and ensuring that we were keeping the same spirit of fast execution and collaboration within the R&D team. Those people are used to being able to be very reactive on the platform and having access to information across the multiple geographies in order to do so.
So we had to go through a lot of discussions internally on access. We decided that we probably had to have a larger team having access to sensitive information than what you would see in a more standard company.