Two years ago we wrote an article on “Marketing vs product analytics, and why you should need both”. The article summarised the difference between both marketing and product analytics as well as the complementarity products in each field have towards each other.
A lot has happened in those two years. The barriers separating both marketing and product analytics have been crumbling and will continue to do so. This article reflects on the evolution of those crumbling barriers as well as the role that iOS17 could play in tearing down the walls at an accelerated pace.
Note – This article represents research done by the author in order to form an opinion and trigger reflection. Readers are invited to use this work not necessarily as the truth but to use it as basis for their research and forming of opinions.
Tearing down the barrier, brick by brick
Let’s reconstruct some of the main events that led to the teardown of the barrier between marketing analytics and plot them in a simplified (!) table. Note that we don’t take into account any regulatory breakdowns etc. as the objective is to compare 2 main fields - marketing analytics and product analytics. Marketing analytics will be represented by the dominant marketing analytics platform Google Analytics and product analytics will be represented by Mixpanel and Amplitude.
The objective is to communicate a general idea or trend, therefore granularity of the visual representation can be reduced.
Around 2 years ago, when we wrote the original article on the differences between marketing and product analytics, Google Analytics 4 was still very new and Universal Analytics was still the dominant Google Analytics version used by most organisations. This is the start of our journey and the zeitgeist in which the original article was published.
From a marketing analytics perspective Universal Analytics was very strong (and therefore also covering the basics) in its field. On the other hand it was very weak in the product analytics field. Therefore no boxes have not been checked.
From a product analytics perspective Amplitude and Mixpanel have been very strong in the product analytics field (and therefore also covering the basics).
Enter Google Analytics 4 (GA4). With GA4 comes the introduction of event-based measurement. A measurement model that Mixpanel and Amplitude have been using for years and is representative for the field of product analytics. Where GA4 introduces the model it fails to make data accessible with the ease other tools in the product analytics market are able to. Therefore only covering the basics is checked. Yes, much can be said on GA4’s visualisation and analysis capabilities but we’re using GA4 as a representation of an industry and therefore simplifying. As you can see there’s a huge opportunity for GA4 to push through and become a dominant tool in both markets if they could overcome this challenge.
In the product analytics field Mixpanel and Amplitude have, in the last 6 months, both announced attribution capabilities (Mixpanel link and Amplitude link). As attribution is the key element in marketing analytics, this makes them move further into the field of marketing analytics capabilities. That being said, Google still has the edge as they have some specific features amongst which are closely-knit integrations with their advertising platforms - using gclid and dclid parameters which can automatically be enabled for attribution purposes.
I would argue that this is the playing field before the release of iOS17.
Both are strong in their core industries but crossovers start to happen.
iOS17 and Link Tracking Protection (ITP) - levelling the attribution playing field
To understand why iOS 17 is important to breaking down the barriers between product analytics and marketing analytics we need to understand the impact of iOS17. More specifically of a feature called Link Tracking Protection or (LTP)
Long story short:
“Link Tracking Protection is a new feature automatically activated in Mail, Messages, and Safari in Private Browsing mode. It detects user-identifiable tracking parameters in link URLs, and automatically removes them.” - source
I highly recommend the article written by Peter Jakus who investigates the impact of iOS17 on different types of tracking. This research in turn is based on research previously done by - amongst others Cory Underwood who has compiled a list of tracking parameters that are impacted (see Peter’s article for further resources).
Current test also indicate it impacts gclid and dclid parameters which give GA4 an edge by reducing friction for marketers.
Why the fuzz? Mainly because Safari is the second most popular browser world wide and the usage of Apple iPhone devices is at an all time high and projected to grow even further.
It also means that, to be safe, marketers will want to move to basic UTM tracking. And that’s in the favour of product analytics platforms such as Mixpanel and Amplitude. Mixpanel and Amplitude have rolled out pageview tracking which makes reproducing certain insights, usually linked to Google Analytics, easier.
All of a sudden the dominant position of GA4 over product analytics tools such as Amplitude and Mixpanel does not feel as absolute as it used to be.
What else does Google have that they don’t have, and how hard is it to counter those measures?
Google Signals - Fairly impossible to reproduce by other vendors
Integrations with Google Advertising platforms - Estimated to be feasible
Consent Mode for data modelling - Estimated to be feasible
Data-driven attribution which is Google’s standard attribution model - Estimated to be feasible at it’s based on Markov chains, which are widely documented
“But GA4 is free” is what most people in the market still state. Yes, but people watching the product analytics vendors making moves in the field of attribution might also have noticed that they have been changing their pricing models to accommodate for this.
Conclusion
Many things can happen between now and the release of iOS17. And I might very simply be overestimating its impact. Where Universal Analytics defined an industry Google Analytics 4 has not been a worthy successor yet. As a long term Google Analytics user my hope is still for the Google teams to somehow turn the ship around and bring back some of the excitement Universal Analytics has brought us all those years ago. The removal of dclid also punches a massive hole in the 360 proposition Google is offering so positioning that product long term might also become more challenging.
In the meantime the Mixpanel and Amplitude teams, which have some very bright people in them on both sides, are shipping at full speed and creating some wonderful things. Be sure to explore them.
One thing is for sure – the analytics industry is going through a revolution like it has never seen before.
If you would like to explore the details of our work with other clients, please feel free to reach out. We are always eager to engage in insightful discussions.