In our ever-changing industry, staying quick on your feet is paramount to overcoming obstacles. But even by typical standards, this past year was particularly tumultuous. One of the most significant concerns has been the official onset of cookie deprecation, a long-anticipated development that has sent publishers scrambling.
Between the alphabet soup of alternative IDs, addressability concerns and data solutions, it’s no surprise that publishers are bulking up on their approaches to data. But is data hoarding the right path forward? If interoperability, scale and privacy are the goals, how can publishers build a data culture that sets them up for success?
Data interoperability is key
According to various sources, less than 10% of open web traffic is authenticated, meaning that upwards of 90% of users are opting out of sharing personal information with publishers and other website providers. Even when they do log in, this data is fragmented as they might not log in on every website, every time.
This means that publishers who solely rely on deterministic IDs, which use personally identifiable information like email logins to tie multiple devices to a single user, will struggle with scale. Single sign-on solutions can help by allowing users to authenticate themselves across multiple locations at once, but they still aren’t painting the full picture.
The best way to solve this is to bring in multiple data sources and allow them to work together. Using multiple IDs and a mix of deterministic and probabilistic data, and storing that data together, can maximize audience match rates and achieve the ultimate goal of interoperability.
It is difficult to know at a glance if the user who has logged in to a publisher’s site is the same user assigned a specific ID tag. But storing these data sources together and using either an internal mapping system or a third-party partner to match across identifiers and data sets can tie various IDs to the same user.
Interoperability is important for understanding reach, as well as for monitoring ad frequency and creating a more holistic measurement framework to attribute success across campaigns.
Don’t forget about data hygiene
First-party data can be a publisher’s superpower, but only if it’s accurate and, ideally, scalable. There are two ways to clean up first-party data: by focusing on getting new data and by scaling up what you already have.
Receiving new data could be as simple as adding a login page to your website and requesting information about your audience’s demographics and interests. It could also mean integrating with a data management platform (DMP) or a customer data platform (CDP) to segment online traffic or offline data.
Scaling data, meanwhile, involves building tech-driven models through a DMP or CDP partnership to expand your first-party data, creating more detailed audience profiles as well as lookalike audiences. This can also be achieved by joining a data clean room to expand your audience with similar, authenticated data sets.
The more alternative IDs and partnerships you bring into the mix, the cleaner your first-party data can become, as you bridge between distinct data sets, connect users across identifiers and build a stronger audience set that can be shared with advertisers.
For example, if User A in your first-party data set has been assigned an ID5, but, through your data partnerships, you are given access to a new data set that shows a User C with the same ID5, you now know that User A and User C are the same user.
The cleaner a publisher’s first-party data – and the more authenticated it is through partnerships and industry collaborations – the more opportunities there are for buyer education. Publishers should communicate the value of their data to advertisers, allowing them to activate through direct syndication or clean rooms to achieve their goals, while also using authenticated alternative IDs to make the data more portable across the ecosystem and, therefore, more accurate, scalable and interoperable.
Embrace a test-and-learn approach
The various challenges presented by cookie deprecation and data hoarding will require multiple solutions. All publishers are navigating an uncertain future, and now is as good a time as ever to try out different approaches and adjust accordingly.
Solutions like UID2.0 can provide targeted audience engagement, personalized ad experiences and frequency management using deterministic data, while partners like ID5 can boost match rates with probabilistic identifiers. Try out multiple offerings. See what they offer on their own, and determine how they can be incorporated together. This is also the time to test partnerships with DMPs or CMPs, including those that prioritize consumer privacy.
In addition to testing identity solutions, another part of your strategy should involve testing contextual targeting solutions to maximize scale and reach. Maybe you use your first-party data to build a contextual audience in a cookieless and ID-free way, predicting the types of content your audience may want to see based on existing information in your data set.
This could also include testing out AI-powered contextual tools that match ads to content based on audio and visual signals found in the content the consumer is watching, and then using that engagement data to continue modeling lookalike audience segments to scale campaigns.
Changing the industry won’t happen overnight, and it can’t happen in a vacuum. Now is an opportunity to set the tone for a privacy-centric future and a data ecosystem built on collaboration, trust and a test-and-learn mentality, rather than siloed data hoarding.
“Data-Driven Thinking” is written by members of the media community and contains fresh ideas on the digital revolution in media.
Follow Connatix and AdExchanger on LinkedIn.
For more articles featuring Lauren Rose, click here.
