Quick Demo of Google Chrome FLoC & Web Advertising Future
Last week Google announced that its rolling out Federated Learning of Cohorts or FLoC as a part of its initiative to stop supporting third party cookies. It is released as a part of developer origin trial which requires jumping a couple of technical hoops to activate these features as may impact normal behavior.
So I took the risks and jumped through the hoops to enable FLoC to define an ID for user activity. Read on for the demo!!
What is Google FLoC?
Google FLoC or Federated Learning of Cohorts is the upcoming Google Chrome feature that promises to deliver relevant, personalized ads without using 3rd party cookies. Instead of identifying & managing user ID in 3rd party cookies, it identifies user intent based on browser activity and converts them into cohorts which are defined by a FLoC ID. I really love the simple way FLoC is explained & elaborated here.
Okay Its Showtime !!
As of writing this post, this website is configured to capture user activity and assign cohort ID to Chrome browsers which have this feature enabled. Lets uncover how FLoC API works for each step users take for the most common path on my blog
Step 1 : Browse to sagarmandal.com
When you browse to my blog you have shared an intent to know about a technology professional who works across cloud, data & digital transformation. At the load of this page, a request was sent to FLoC API for assignment and as a result my browser was assigned an ID and none of the details are saved in the cookies.
Step 2 : Browse to post about Adobe Campaign
Next lets browse to my blogpost about Adobe Campaign APIs automation. Notice that the browser detects the change in topic based on the URL and assigns a new cohort ID
Step 3 : Browse to post about Adobe Campaign & Snowflake
This is the third and final step where we browse to the most read blogpost in 2021 about Snowflake & Adobe Campaign integration . Again the same outcome where your interest is assigned a particular ID and no user identifiers or browsing history in terms of segmentation is available with the browser.
So what ads will I see?
Let’s assume that you browse to another media website showing ads named – notfake.news right after visiting my blog.
At this time your browser is identified by a cohort ID based on your recent history. The value is made available to websites to get the most relevant ad based on the cohort ID via a new JavaScript API :
cohort = await document.interestCohort(); url = new URL("https://notfake.news/getCreative"); url.searchParams.append("cohort", cohort); creative = await fetch(url);
Lets assume that our good friends at Adobe Campaign & Snowflake have setup ads for the above mentioned Cohort Ids. As your cohort ID is shared, the DSP would present the most relevant creatives.
Predictions for Web Advertising Future
1. Even more investments from brands : For the last 5 years there has been a lot of a negative press about brands reducing/removing web advertising investments primarily due to brand safety reasons. I have been in those meetings where all stakeholders were limited by the capabilities of 3rd party and ad ecosystem to ensure brand safety. I predict that would be changing in the next 5 years as interest based ads would place the right creative based on user interests/intent and not based on cookie history.
2. User Privacy & Increased Chrome Browser Adoption : As we saw above there are no user IDs collected based on the browsing history. Besides driving confidence for users to adopt Google Chrome, advertisers can sleep well with the fact that the capability to capture, retain & manage user identifier within Chrome would no long be available.
3.Better Ad Experience : We all have had those experiences where the ads presented are not relevant to our digital interactions or ads that don’t make sense at all since they are too generic. Intent based cohorts changes the web ads game where advertisers have to build particular ads to get users interested to continue their intent. The only challenge I see here is delivering for use cases like retention & suppression which are dependent on certain user activity .
If you disliked the post tell me, else like and share with others!
Hope you liked it!
1 Comment
Thank you for sharing this one Sagar.