Hey Reader!
I read a post recently from Sanity called "Content is Data" and it's been rattling around in my head ever since. The argument: content locked inside a traditional CMS, where it is tied to its presentation, copy-pasted between surfaces, and impossible to query, is a bottleneck to everything you want to do next. Your content isn't really content. It's HTML with some text trapped inside.
That got me thinking about how if content is data, how would that interact with marketing data? Are attribution problems really just a content architecture problem?
Think about what analytics can actually tell you right now. Everyone measures what happened to a page, such as sessions, engagement, and conversions. What we’ve been trying to do is determine what happened with the content. Did the pages tagged as "customer story" outperform the ones tagged as "product feature"? Did content written for late-funnel readers convert better than early-funnel (it should!)? Is the blog category driving qualified pipeline, or just traffic?
When we’re able to set this up, it works great but it’s way more work than it should be. Usually it involves meta tags, data labels, lookup tables, or some combination of the above. It can be pretty fragile and needs constant tending.
If instead we approach content as data, not "has a category field tacked on," but actually an integral part of the content, you can join it to behavioural data, to CRM data, to pipeline. Then you can ask which kinds of content drive which kinds of outcomes, without all the manual work.
This is the shift that we’re going to be taking with how we build websites at Kick Point, and I think this is a shift where us analytics practitioners have a bigger role than most of us have been playing. We've been the people who measure what the website does. The interesting frontier is being the people who help decide what the website is, because the structure of the content determines what you'll be able to measure, and that decision gets made long before anyone starts setting up GTM.
Laura made a great point — she feels that the days of the $100K brochure-site era are ending. What replaces it is websites built as structured content systems, with analytics baked into the architecture instead of bolted on afterwards. Doesn’t that sound nice? Let’s make it happen.
We Are Not Data Accountants
You might recall this discussion from a previous newsletter! It's now turned into a full post on Analytics Playbook.
The core idea is something I've been saying for a while: the data accountant mindset, where every session must be accounted for and every conversion number must be exact, sets you up to fail at a job that was never actually yours. The data was never complete. It just used to be less obviously incomplete.
A special thank you to reader Kira Rodriguez, who wrote to me after the original newsletter note went out and shared how this framing helped her land a senior role in her recent job search. Her story is in the post.
Read the full post →
Articles Worth Your Time
———•
|
A Better Funnel Chart for Looker Studio Data Studio
Mehdi Oudjida published a free custom funnel chart visualization for Looker Data Studio that addresses two real limitations of the native funnel chart.
The native funnel chart works fine for basic use cases with one dimension, one metric, and one sort field, but it falls apart the moment your data doesn't fit that shape. And most GA4 e-commerce funnels don't, because your steps live in separate metrics (items viewed, items added to cart, items purchased) rather than rows of a single dimension.
Mehdi's custom viz handles both data models, renders as a structured table with dedicated toggleable columns for % of previous step and % of first step, and lets you control your column widths. It's the kind of chart your stakeholders can read without you having to explain it to them afterwards, which is usually the real test of a visualization.
Looker Studio Is Data Studio Again (Welcome Back)
On April 10, Google announced that Looker Studio is being renamed back to Data Studio, reversing the 2022 rebrand that folded Data Studio into the Looker umbrella.
What's interesting to me about this “rebrand” is the positioning. Google is now explicitly framing Data Studio as the tool for personal data exploration and ad-hoc reporting, and Looker as the enterprise BI platform for governed, semantic-model-powered analysis. Which is what most of us understood the tools to be doing all along? The 2022 unification under one brand confused buyers more than it clarified anything, and it's good to see Google acknowledge that and adjust.
Server-Side Tagging Isn't a Fix, It's an Upgrade
I absolutely loved this piece from Simo Ahava at Simmer on how to actually position server-side tagging, because it pushes back on one of the major “sales” points of server-side tagging, which is data quality.
The idea is that with server-side tagging, you’ll "get data back" from users who block trackers. Simo argues, pretty persuasively, that this is the wrong case to make. Ad platforms are already modeling against data gaps, so feeding them a few extra percentage points of row-level data isn't the breakthrough some server-side vendors are selling.
The real value sits in three places: client-side performance (every vendor script you move off the browser is a load you remove from your users' devices), control over what the vendor actually receives (stripping sensitive parameters, normalizing inconsistencies, enforcing your own data governance at the point of collection), and enrichment at scale without exposing internal APIs to the browser.
If your reason for going server-side is to maximize data collection from users who don't want to be tracked, you're building something that might cause problems for you down the line. But if you're looking at it as a performance and governance upgrade, that is the right move for a lot of organizations.
Where You Can Find Me
———•
|
I Was on the Search Party Podcast
I joined Amanda Milligan on the Search Party Podcast to talk about building analytics for the zero-click world we're all living in now. We covered the traffic-to-business-outcomes shift that local SEOs have been doing for years (and that everyone else now has to catch up on), why Google Search Console has never been the source of truth you thought it was, and the specific GA4 setup mistakes I fix in client audits on the regular.
If you've been trying to figure out how to reposition your reporting away from traffic as the hero metric, this conversation is a good starting point.
That's it for this edition of The Huddle. As always, if you have questions or want to share what you're working on, just hit reply!