GA4: The Two-Year Goodbye

Written in 2026, looking back at the day Google announced Universal Analytics would die, with the benefit of knowing how the whole migration actually shook out.

Sixteen months is a generous runway by software standards. Google gave us that much today: they said Universal Analytics would stop processing data on July 1, 2023, and that everyone should move to GA4 before then. My first reaction was relief that the date was so far away. That relief, I already suspect, is the whole problem.

A deadline with a soft edge

The thing about a deadline that far out is that nothing bad happens if you ignore it for a year. Your old reports keep working. Your dashboards keep filling. The cost of waiting is invisible right up until the moment it isn’t, and human brains are famously bad at pricing invisible costs.

I know how I work. I know how the people I work with work. A hard error, a broken checkout, a site that returns a 500, gets fixed inside the hour. A slow, silent deadline sixteen months away gets a note in a document that nobody opens again. GA4 is squarely the second kind of problem, and I think most of the industry is about to prove it.

Nobody wants to migrate to something worse

There is a second reason the waiting will happen, and it is not just laziness. GA4 right now is genuinely harder to like than the tool it replaces. The interface is unfamiliar. The whole model changed from sessions to events, which is arguably more honest about how people actually use the web but also means half the reports I reach for by muscle memory are either moved or missing.

So the incentive is backwards. You are being asked to leave a comfortable tool for an uncomfortable one, with the reward being that you get to keep having analytics at all. That is not a reward that makes anyone move early. It makes people wait until the comfortable tool is actively being switched off.

The cost hiding in the calendar

Here is the part I want to remember, because I think it is the real lesson. The expensive thing about migrating late is not the migration itself. It is the data.

GA4 only starts collecting from the day you set it up. If a client waits until June 2023 to flip it on, then in July 2023 they have a brand new property with no history, which means no year-over-year comparison for a full year. Every “how did this quarter compare to last year” question just returns a shrug until mid-2024. The people who set GA4 up early, even while ignoring it, will quietly be the ones with usable numbers.

I would like to say I will be one of them on every property I touch. Realistically I will be early on the ones I own and last-minute on a few I manage, because a deadline in July 2023 does not feel real in March 2022. Ask me again next summer.