Google Topics API – is less really more?

Some marketers have equated digital to some form of hypertargeting. They are also borderline lazy. Instead of a painful process of cooking a meal with the right ingredients, process and tools , they wanted digital equivalent of fast food – drive in, order, instant gratification of marketing money. Give me this targeting, give me this … they got a menu that was handcrafted and recrafted by agency minions after discussing and exploring many options..

They fuelled this dystopian flywheel – overcollection of data from users, leading to privacy concerns and now we have less data collection opportunities and need methods to share data among third parties thanks to measures

topics api

Apple calling privacy a fundamental right

Google launches Topics API

Topics API is a mechanism for advertising tracking and personalization introduced by Google in Chrome browser. It creates a list of “topics” that are associated with a users browsing history. It is compliant and an improvement over FloC in 3 ways –

1.Users can manually remove certain topics assigned in the browser. There is also an opt out mechanism ( i.e if the users know how to do it, many users still give a lot of permissions due to lack of knowledge)

2. It is also public through a Github repository , so no more coffee lovers type segments that ad tech used to over propagate, so we have some standardization of segments

3. It is also transient – browsing history is for last 3 weeks.

However the Topics are rather broad (e.g. ‘Sports’, ‘Fashion’). BROAD with font size 90 ..Hence marketers may not get the coffee lover who plays sports and has two kids, they will have to use broader segments.

Google’s initial draft says it is thinking around 350 Topics. According to Google , Topics would be selected from a list of items such as “Country Music”, “Make-Up & Cosmetics” or “Vegetarian Cuisine”. These topics would initially be curated by Chrome for testing, but with the goal that the topic taxonomy becomes a resource maintained by trusted ecosystem contributors. The taxonomy needs to provide a set of topics that is small enough in number (currently proposed to be around 350, though we expect the final number of topics to be between a few hundred and a few thousand) so that many browsers will be associated with each topic.

If you compare this with the IAB taxonomy – a commonly used method for category targeting in programmatic advertising , it sounds like a better approach. Why ? IAB has subcategories like ‘Certified ,Pre-Owned ,Convertible, Coupe’ in its Automotive category and it has subcategories like spyware . Which self respecting website will tag itself as spyware. This results in administration issues for publishers – because they have to keep on tagging their sites correctly and its a never ending process and its never MECE( mutually exclusive, collectively exhaustive) leading to either over segmentation – too many granular segments or undersegmentation – many users who are not there in any category. IAB already has 392 subcategories and some of them are antiquated and overcooked and not relevant in non Western markets.

I have 2 main hopes from the Google Topic API

1. Marketers become more practical and go back to first principles and embrace less is more when it comes to segments. If Google can provide the heat map of topics- for instance what % of users are really into food topics and for what period of time. This will lead to richer user insight and really help the industry do a Pareto analysis of what type of topics lead to performance for their subcategory. Hence an industry recognized wiki for interest subcategorization with some insights on campaign performance ( aggregated) will be wonderful to create an accepted standard for interest based targeting compared to the antiquated IAB methodology.

2. Google engineers can solve the problem of content tagging and make Topics an agile methodology by really working with publishers to solve their pain points – rather than dumping Google Analytics on them. This is to solve an upstream problem of publishers mis tagging the content and leading to poor inputs into data segments.

However its highly likely that my hopes will be paid services by Google or services in lieu for using their platforms – thus strengthening the big G again. 🙂