Even if Google doesn’t meet its self-imposed deadline to fully phase out third-party cookies on Chrome by the end of the year, agencies need to be prepared.
Cookie loss is happening, even if it doesn’t feel imminent, and there’s no point in procrastinating, said Sisi Zhang, chief data and analytics officer at Publicis-owned Razorfish.
“We’re anticipating that there will be significant changes this year,” Zhang said, both in terms of what data is available to advertisers and how buyers must now approach targeting and attribution as a result.
Zhang, who was promoted to data chief in January from EVP of data science and analytics, is tasked with advising the agency’s clients, which include Dove, Samsung and Cadillac, on how to productively manage signal loss.
One trend Zhang is seeing is a rising interest from brands in building their first-party data sets and trying alternative identifiers like UID2.
As an industry, “we’ve really been leaning heavily on third-party signals that we won’t have [for much longer],” she said. The next step for advertisers is to replace those signals with first-party ones wherever possible.
Zhang spoke with AdExchanger about how advertisers are bracing themselves for signal loss.
AdExchanger: Where is signal loss hitting advertisers the hardest?
SISI ZHANG: Measurement.
Measurement is so important because it informs advertisers about a consumer’s full purchase journey, which they need to plan future campaigns. But without signals like cookies, visibility into that journey becomes more opaque, and advanced attribution becomes more difficult.
Plus, measurement limitations raise questions about how effective an advertiser’s targeting strategy actually is.
AdExchanger Daily
Get our editors’ roundup delivered to your inbox every weekday.
Daily Roundup
How is Razorfish advising clients on dealing with signal loss?
We recommend advertisers bolster their first-party data.
We know managing customer data platforms takes a lot of time and investment, but the best way to withstand signal loss resulting from data privacy regulations is to build a value exchange that gives consumers a compelling reason to share their data directly with a brand. And building that value exchange may call for new means of data collection [such as prompting consumers to join a loyalty program in exchange for access to services].
But first-party data is not always reliable. How does that pose problems for marketers?
First-party data isn’t perfect. It can be challenging to acquire usable first-party data.
Getting accurate first-party data may require going back to the drawing board to check whether a brand’s data collection mechanisms prompt consumers to share data that’s actually valid.
But even then, integrating a CDP takes months because it requires revisiting a brand’s technical infrastructure. If a marketer has data living in multiple cloud providers, for example, they have to invest in centralizing that data into one internal platform.
So while we do recommend advertisers build a first-party database, there are a lot of steps required to make sure that road map is actually feasible.
What are these required steps?
The big thing we’re seeing right now is brands testing out methods of ID resolution, particularly with clean rooms and alternative identifiers like UID2. Alternative IDs and clean rooms can both help brands match their audiences with data from other partners in a way that’s conscious of privacy without sacrificing accuracy.
We’re also seeing an uptick in media mix modeling for attribution.
As for alt IDs, some advertisers seem hesitant to try UID2 in particular. Why might that be?
I don’t think there’s necessarily one ID leading the way [yet]. For the most part, advertisers and publishers are both still testing a plethora of IDs to figure out which one works best for particular use cases, such as activating CTV campaigns.
What is the broader industry impact of signal loss?
Consolidation is happening because there’s still uncertainty as to which cookieless solutions will work best – and for which use cases. Bigger players are acquiring smaller ones that may have data or tech solutions that address different use cases.
At the same time that advertisers are trying to centralize their own data, there’s a broader trend of data decentralization happening as marketers look to separate the data they’re using for targeting and measurement to test what’s working.
Consolidation through M&A can help make that process a little easier by offering advertisers multiple solutions in one place.
This interview has been lightly edited and condensed.
For more articles featuring Sisi Zhang, click here.
