Hey Reader!
Here is a mantra that I want you to carry with you in your professional career: we are not data accountants.
Somewhere along the way, marketers got sold the idea that good analytics means precise analytics. That if the numbers don't add up perfectly, something is broken. That your job is to account for every single session, every click, every conversion, and if you can't, you've failed.
The problem with that whole idea is that the data was never complete. It just used to be less obviously incomplete. Today, between ad blockers, cookie consent refusals, Safari clearing tracking cookies, and cross-device behaviour that we simply cannot stitch together, you could be doing everything right and still be missing a significant chunk of what's actually happening. That's not a you problem. That's just the reality of analytics in 2026.
What this means in practice is that our reports need to shift from counting to reasoning. Instead of asking "how many sessions did we have this month," we ask "is organic traffic trending up or down, and do we understand why?" Instead of obsessing over whether our conversion number is exactly right, we look at whether the ratio of effort to outcome is improving. We use percentages instead of raw counts when comparing across time periods or channels. We round large numbers. We say "approximately" when appropriate. We focus on what the data is telling us, not whether every row reconciles perfectly.
This is the philosophy I've been baking into how we build reports at Kick Point and yes, it takes work to educate clients about all the data discrepancies out there. But it's worth it. If you've ever sat in a meeting defending a number you knew was wrong because someone expected precision you couldn't deliver, know that you are absolutely not alone.
What does this look like in your reporting? Hit reply and let me know — I'd love to hear how you're navigating this conversation with your clients or stakeholders.
New Guide: Track Offline Advertising in GA4 with UTM Redirects
Offline advertising creates a real gap for most businesses. People see the ad, go home, and search for you later. By the time they convert, you have no idea that the billboard was involved at all.
UTM redirects bridge that gap. Instead of sending people to a long, messy URL, you create a clean vanity URL that redirects through your UTM parameters automatically. The person types in something easy to remember, and your marketing gets the credit it deserves.
I just published a complete walkthrough of how to set this up, including how to structure your UTM parameters for offline campaigns and what to watch for when you're reviewing the data.
Read the full guide →
Articles Worth Your Time
———•
|
Tracking AI Overviews and AI Mode Traffic: A GTM Approach
Nicholas Blazer published a detailed method for tracking clicks from Google AI Overviews and AI Mode in Google Analytics using Google Tag Manager and I haven't implemented it yet, but I'm really interested in his approach. Nicholas built this after exploring my URL fragment tracking method, which I appreciate!
Here's how it works: when someone clicks a citation link in AI Overviews or AI Mode, the browser opens a new tab. That behaviour leaves a specific set of signals, such as referrer data, tab history length, navigation type, and visibility state, that can be captured through a custom JavaScript variable in GTM. The method uses all of these together to reduce false positives, like when someone just manually opens a search result in a new tab.
This is exactly the kind of careful, evidence-based technical work that helps us get closer to understanding AI traffic, at least directionally, if not perfectly. Which, as I said above, is all we're really after anyway!
What Real Attribution Data Actually Looks Like
Darren Shaw at Whitespark analyzed nearly three months of trial signup data for their Local Ranking Grids software using a simple question at signup: "Where did you hear about us?"
YouTube and their own newsletter were the top two sources, each at 16% of acquisitions. Web search came third at 14%, followed by ChatGPT at 13%. What I found interesting is that ChatGPT had the lowest conversion rate of the top sources at 7%, compared to 16% for YouTube and their newsletter. Darren's hypothesis is that his personal presence in video and social creates trust that a ChatGPT mention can't replicate.
What I find most valuable about this data is the last line of Darren's post: "I love having this 'how did you hear about us?' attribution data. It's just so clear and simple compared to trying to figure out our marketing attribution from Google Analytics." He's right. GA4 literally cannot show you most of this. Self-reported attribution is imperfect because people don't always remember accurately, and some sources inflate or deflate depending on how the options are presented, but it's directional data that tells a story your analytics dashboard never will. This is a great reminder that GA4 is one input, not the whole picture.
The Business Model Is the Strategy (Everything Else Comes After)
Juliana Jackson published a piece on how the "industry consensus" around marketing strategy operates inside a very specific bubble that is predominantly US-based, and built around a particular type of B2B SaaS business model. If you work outside that bubble, you've probably noticed that a lot of the ideas out there just don't quite fit.
Her core argument is that the business model itself is the strategy and that before you apply any framework or best practice, you need to understand the model you're actually operating in. The "right" approach to content, attribution, or channel strategy looks completely different for a franchise business versus a direct-to-consumer subscription versus a professional services firm.
This resonates with something I see constantly at Kick Point: clients who've been handed generic strategies that don't account for how their business actually makes money. If you've ever implemented a "best practice" that felt slightly off from the start, this is probably why.
Where You Can Find Me
———•
|
Is Anything Real? Podcast with Adam Barney
I recently joined Adam Barney on the Is Anything Real? podcast, and given what I wrote above about data precision — let's just say the podcast title felt very appropriate. Check it out if you want more of my thoughts on navigating analytics reality in 2026.
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!