The advertising industry ‘has stalled’ as it faces a cookieless future

Quantcast is a global digital advertising, audience insights and measurement platform. Understandably, it has been following the Google Privacy Sandbox saga with close attention. We sat down with their U.K.-based CMO Amit Kotecha in their New York office for a conversation about recent developments, including the latest delay in cookie deprecation and the new report from the U.K.’s Competition and Markets Authority. (Interview edited for length and clarity.)

Q: Google has said it is pushing the complete deprecation of third-party cookies back to 2025. A couple of days after that announcement, we got the new report from the CMA that adopted most of the concerns that the Information Commissioner’s Office had surfaced about privacy. Google surely saw this coming?

A: The timing is a bit too coincidental. This is a complicated industry; there are many players. Google is a significant player, in some ways the backbone, with their browser and their ad server being very commonly used. I do think all the concerns that have been raised need to be addressed, but the deadlines are a bit useless. Our conversation with advertisers has always been — you should be thinking about your company and what you’re doing to be able to use digital advertising to your advantage without cookies.

We will work with any partner in the ecosystem. As a platform, we should take all the complexity that there is and take it away from our customers. As a technology company, we should create our own solutions. That’s not to say we’re not testing Privacy Sandbox, but we can’t wait or we’ll be here next year without any innovation. At the moment, I feel like the industry has stalled; there’s not much innovation hapening because everyone is waiting to see what Google does.

Q: It looks like Privacy Sandbox isn’t going to be up and running any time soon.

A: Yes. There’s some really good ideas in there to protect privacy and continue with advertising, but they need to be tested properly. From the second an ad is created to when it is delivered, there are many “hops” and those “hops” have been created over 20 years and each one provides, or should provide, value. All the companies in this ecosystem need to be connected in order for Privacy Sandbox to work and not everyone has done that work yet. Everyone has to connect to the framework to test it properly.

It’s expensive, too. Do we put the engineering time into Privacy Sandbox, or do we put it into our own solutions which we believe are working well and delivering value for our customers today? That’s not to say we wouldn’t pivot and do something on Privacy Sandbox if a customer was willing to spend time with us to do it, but at the moment there’s very little demand. A lot of people talking about it, but not much activity.

Dig deeper: Google’s Privacy Sandbox: What you need to know

Q: I’ve heard that the conditions for testing are currently so limited that it’s almost meaningless. Google allowed 1% of Chrome users to deprecate cookies, but tests running that 1% don’t tell you what will happen when it’s 70% or 80%.

A: That 1% also has to be online at that given moment in time in order for you to deliver an impression and track whatever performance metric that you’re tracking. We’re at a standstill. We have this conversation every year and it’s exhausting.

Hundreds of millions of dollars are spent in this industry every year, outside the walled gardens, and that money delivers value to the advertiser, delivers value to the consumer if it’s done well, and — most importantly — it funds the open web. This delay, any type of friction in the system, means less confidence among advertisers and they push their money into areas where you have less friction; more money gets funneled into walled gardens and less gets spent on the open web. Advertisers will tell you they don’t want to spend all their budget in walled gardens, but there’s too much chaos here and they don’t understand it well enough.

Dig deeper: Goodbye to cookies: Digital advertising’s leap in the dark

Q: A lot of advertisers can find their existing audience, of course, through first-party data, but they also want to extend their reach and find new customers. What are the solutions on the table for that?

A: First-party data acts as a seed to build what we call “models.” You use whatever data points are available to you to build models to find new audiences. That’s the way Quantcast has always worked. When you don’t have deterministic data, you have to fill the gaps with machine learning. If you have a large dataset, you can apply a basic model on top of it and it will do very well. As the dataset depletes or diminishes, as it will in the cookieless world, the model has to become more sophisticated.

Q: It doesn’t sound like Quantcast is heavily invested in the identity resolution that happens in data clean rooms.

A: I think data clean rooms have a place, but as a standalone solution, no. A lot of folks have invested in clean rooms but haven’t found a way to activate them.

Q: Another thing I am seeing from some platforms is a return to contextual advertising. It’s been around forever, but with AI can do more sophisticated things today.

A: Contextual is in our platform, we just don’t break it out as a separate solution. You can buy a contextual strategy if you want, but it’s just another signal that we add to our overall model. An advertiser can buy just contextual, but we don’t recommend it. To get the best performance you should use all the data available.