Skip to content
All Blogs

Industry perspective: unpacking real-time for 'identity'

We think everyone at this point has seen the term "real-time" on 1000s of vendor websites, and to that end you've then seen 1000s of definitions of "real-time" as well. Candidly, we've contributed to that as a brand but have a strong desire to break things down to a more granular level to drive better conversations with all of you.

So, consider this the first of many blogs from us that will attempt to unpack something in the industry and try to make sense of it so you can have real conversations with your vendors (and hopefully us!). Today we are going to focus on identity, which you might suspect is the result of us launching our 1st Party Identity Graph and Profile Building capabilities in the Celebrus CDP.

I think we'd all agree that Identity is essential to generate value from any use case you can think of. That's why there's seemingly an article per day highlighting the challenges with setting identifiers, persisting those identifiers, or recalling who someone is as he/she/they land on a digital touchpoint based upon those identifiers. The Celebrus CDP newly minted 1st Party Identity Graph provides you with the gold standard solution to this, and now we're going to step through what good looks like when thinking about "real-time."

Let's take a situation where you intend and hope to present offers or assign individuals to a particular segment for your downstream campaigns. When someone arrives on your website, mobile app, or other digital channel, there are 5 things you need to be able to do before you even begin to think about decisioning or testing and these 5 steps form the basis for "real-time" data:

  1. Determine if you know who this person is based upon a persistent identifier on the device and whether this person has been there before upon immediately landing on a digital channel.
  2. Recall the history and profile attributes about that individual.
  3. Append the current session, inclusive of the journey, interactions, and marketing signals that may be relevant to the historic profile as they happen.
  4. Enhance the Identity Graph for this individual if anything is provided (e.g., email address, login, phone number, etc.).
  5. Send relevant data, from the current session or history or both, to the right systems in the right format to allow for decisions, A/B tests, triggered campaigns, audience updates, etc.

When you look at that list, it probably seems quite daunting. What makes it even more challenging is that you have to do the above in milliseconds to keep up with the dwindling amount of patience consumers have. To be more specific, Pega, one of our strategic partners, defines that decisioning must happen within 200ms in a recent paper. The decision would ultimately be Step 6, so that puts into perspective just how little latency you have room for in the first 5 steps above.

You may be saying "Well, why do I need the history then? I can just send in-session information as it happens via my data layer." It's true, not all decisions need history, but we'd argue that most organizations take that stance because they have become pre-conditioned to the limitations of their tag-based solutions and let that dictate use cases and strategy. At a minimum, you know how many devices you use in a given day, and simply having the ability to know that you've left off somewhere on mobile and are trying to pick that back up on web would be insanely valuable.

If you could ensure that:

  • all 5 steps above could be completed in milliseconds,
  • be confident that you had the data granularity to continually power innovation, and
  • solve all your identity challenges while
  • ensuring 100% compliance,

then wouldn't it be worth chatting with our team to learn how we do that for organizations all around the world?