Universal Analytics stops processing new data on 1 July. That's roughly ten weeks away. If you've been deferring this — and most teams I speak to have — here's the honest separation between what has to happen before the deadline and what can comfortably wait until the autumn.
Must happen before July
GA4 collecting data on every page, with your key conversions defined as events and verified as firing. That's genuinely the minimum. If GA4 isn't collecting on 1 July, you have a gap in your history that no amount of later work recovers. Everything else on the list can be improved afterwards; this cannot be done retrospectively.
- GA4 tag live on every page, including subdomains and any separate app
- Core conversions defined as events and confirmed firing in the debug view
- Consent handling working, so you're not collecting what you shouldn't
- Historical UA data exported and stored somewhere you'll actually find it
Can wait until autumn
Rebuilding your dashboards, sorting out custom dimensions, connecting BigQuery, tidying up the channel groupings so paid social stops appearing in three places. All of this is real work and none of it is urgent, because the data is being collected either way and you can improve the reporting on top of it whenever you have the time. Teams routinely invert this — perfecting reports while the tag is still missing from the pricing page.
Export the history properly
Decide which reports genuinely inform decisions. In practice it's usually a handful: acquisition by channel over time, conversions by channel, top landing pages, and whatever your board deck references. Export those as actual data, not screenshots, at monthly granularity for the last two or three years. It's an hour of work that you will be extremely glad of the first time someone asks a multi-year question.
Set expectations about the numbers
Your GA4 numbers will not match your UA numbers, and the difference can be substantial. Sessions are counted differently, the default attribution model isn't the same, and consent handling may now be excluding traffic you were previously counting. This is not a bug and there's nothing to fix. Tell your leadership team before the first monthly report lands, not afterwards — the credibility cost of an unexplained drop is much higher than the cost of a pre-emptive email.
Get the tag live and the conversions firing. Everything else is a problem you can solve in September with better data than you'd have had otherwise.
Want to apply this to your business?
Book a free 30-minute strategy call and we'll diagnose your specific growth blockers together.
Book a Free Call