Universal Analytics stopped processing data for standard properties from July 1, 2023. If you open your old reports and see a warning banner, or numbers that suddenly flatten, that is why. Google Analytics 4 (GA4) is now the only version of Google Analytics that will keep collecting new data for most sites.
The good news: your old data is not gone yet, and you still have time to save it. In this guide I explain what changed this weekend, what still works, and the exact steps I’d take this week.
What Happened on July 1, 2023
Google first set this date back in March 2022. Standard Universal Analytics properties (the free version, with tracking IDs that start with UA-) would stop processing new hits on July 1, 2023. A “hit” is any piece of data your tracking code sends, such as a pageview, an event or a transaction.
That date has now arrived. Here is where things stand on Monday, July 3:
- Standard properties: processing of new data is being switched off. This includes standard properties that sit inside accounts which also have 360 properties.
- Universal Analytics 360 properties: Google gave paying 360 customers a one-time extension, so their properties keep processing data until July 1, 2024.
- Old data: Google says you can still access previously processed Universal Analytics data until July 1, 2024. After that, Universal Analytics properties will no longer be available.
Google’s own Universal Analytics help page holds the official dates. I’d bookmark it, because Google updates it as the shutdown moves along.
Why Your Property May Still Show New Data
Many people have noticed that their Universal Analytics property still shows data for July 1 and 2. That is expected. Google said on July 1 that the switch-off “will not happen overnight”, so some properties may keep processing for a while.
Google also said all properties have been queued, and the rest will be handled on a rolling basis. So don’t treat any extra days of data as a sign that the deadline moved. It hasn’t. In my view, the safest plan is to act as if your property could stop at any moment.
What Google Already Did for You
In March 2023, Google started creating GA4 properties automatically for standard Universal Analytics properties that had not set one up and had not opted out. Where possible, it reused your existing site tag, so the new GA4 property may already be collecting data.
That sounds helpful, and it is a start. But an auto-created property only copies basic settings. In my experience, it rarely has the right conversions, filters or links set up. Treat it as a draft that you still need to check.
What Still Works After Universal Analytics Stopped Processing Data
A lot of people panic and assume everything is broken. It isn’t. Here is what you can still do with a standard Universal Analytics property for now:
- View reports: all data processed before the shutdown is still in the interface.
- Export data: you can download reports or pull data through the API.
- Use dashboards: Looker Studio reports that read Universal Analytics will still show historical data, but no new dates.
And here is what you can’t rely on anymore:
- New data: once your property stops, no new sessions, goals or ecommerce sales will appear.
- Editing: Google says you will have Viewer access to historical data after the sunset, not Editor access. So don’t plan on changing settings or building new goals.
- Google Ads goals from Universal Analytics: if you imported Universal Analytics goals into Google Ads, those conversions will stop recording once processing ends. That can hurt automated bidding.
The First 5 Things I’d Do This Week
If you only have an hour, here is my order of priority. I’ve put the steps that can cost you money or data at the top.
- Check that GA4 is collecting data. Open your GA4 property, go to Reports, then Realtime. Visit your own site in another tab and confirm you appear. If nothing shows, your GA4 tag is missing or broken.
- Fix Google Ads conversions. Link GA4 to Google Ads, import your GA4 conversions, and set them as the primary goals your campaigns bid on. Do this before your bidding runs on empty data for weeks.
- Set GA4 data retention to 14 months. Go to Admin, Data Settings, Data Retention. The default is only 2 months, and older event data that has already been deleted can’t be brought back.
- Start exporting Universal Analytics data. You have until July 1, 2024, but I wouldn’t wait. Export the key reports now while you still remember what matters.
- Tell your team and clients. Send a short note explaining that Universal Analytics numbers will stop and GA4 numbers will look different. This saves a lot of confused emails later.
If you followed our GA4 migration checklist earlier this year, most of this should already be done. Use the list above as a final check.
How to Export Your Universal Analytics Data
Historical data does not move into GA4. The two tools use different data models, and there is no import button. If you want year-over-year comparisons next year, you need to save your old data somewhere yourself.
Decide What You Actually Need
Don’t try to export everything. Most businesses use a small set of reports again and again. Before I export anything, I write down the questions people actually ask about the site, such as “how much organic traffic did we get last December?”
For most sites, I’d save these, broken down by month for at least the last two or three years:
- Sessions and users by default channel grouping (organic, paid, direct and so on).
- Landing pages with sessions and goal completions, ideally filtered to organic search.
- Goal completions and conversion rate by goal.
- Ecommerce revenue, transactions and average order value, if you run a shop.
- Device category and country, if these matter for your reporting.
Pick an Export Method
There are three practical ways to get the data out. Which one you choose depends on how much data you have and how you’ll use it.
| Method | Best for | Watch out for |
|---|---|---|
| Export button in reports (CSV, Excel, Google Sheets, PDF) | Small sites and a handful of key reports | Row limits and sampling on large date ranges |
| Google Analytics add-on for Google Sheets | Monthly reports you want to rebuild as tables | Needs some setup and a bit of patience |
| Reporting API into a database or BigQuery | Large sites and teams with developer help | Takes technical time to set up |
For most small and mid-sized businesses, I recommend the Google Sheets add-on. You set up a report once, choose your metrics and dimensions, and pull month-by-month data into a sheet. It’s free and easy to share.
Avoid Sampled Exports
Sampling means Google calculates a report from a part of your data instead of all of it. In the interface, a sampled report shows a note near the top saying it is “based on” a percentage of sessions. Large date ranges and custom segments make sampling more likely.
My rule of thumb is to export one month at a time for busy sites. Smaller date ranges usually avoid sampling, and monthly data is what most people compare anyway.
Save Your Settings Too
Data isn’t the only thing worth saving. Before access ends, I also take a record of:
- Goal names and how each goal was set up.
- View filters, such as internal IP exclusions.
- Custom dimensions, content groupings and channel groupings.
- Annotations, which often explain spikes and drops you’ll forget about.
A simple spreadsheet with screenshots works fine. It helps you rebuild the same logic in GA4 and explain old trends later.
Get GA4 Ready for Daily Use
If GA4 has been running quietly in the background, now is the time to make it the tool you actually use. Here’s what I check on every property.
Confirm the Tag Is on Every Page
GA4 uses a measurement ID that starts with G-. Check that it fires on every page template, including checkout, thank-you pages and any subdomains. I usually see gaps on landing page builders and booking tools that sit outside the main site.
If you use Google Tag Manager, preview your container and click through a full conversion path. If you set GA4 up next to your old tag, as in our guide on running GA4 alongside Universal Analytics, you can now pause the old Universal Analytics tags once you’re sure they’ve stopped. I’d leave them in place for a couple of weeks first, just to be safe.
Rebuild Your Goals as Conversions
GA4 has no “goals”. Everything is an event, and you mark the important events as conversions. For example:
- A contact form thank-you page becomes an event like
generate_lead, marked as a conversion. - An ecommerce sale uses the
purchaseevent, with revenue sent as a parameter. - A newsletter signup might be a custom event like
newsletter_signup.
Go to Admin, then Events, and toggle “Mark as conversion” for the events that matter. If an event doesn’t exist yet, you can create it from an existing event or set it up in Tag Manager.
Check the Basic Settings
These take a few minutes and prevent messy data later:
- Internal traffic filter: define your office IP addresses and activate the filter.
- Unwanted referrals: list payment gateways like PayPal so they don’t steal credit for sales.
- Cross-domain measurement: add any other domains in your user journey.
- Search Console link: link it so you can see organic queries and landing pages inside GA4.
- BigQuery link: optional, but free for standard GA4 properties. It gives you a raw copy of your data that doesn’t expire.
If you care about SEO reporting, I’d turn on the Search Console link first. It pairs nicely with the Search Console Links report when you review how pages earn traffic.
Why Your GA4 Numbers Look Different
This is the question I get most after any migration. GA4 numbers will not match Universal Analytics, and that’s normal. The two tools count things differently.
Sessions and Users
GA4 doesn’t start a new session at midnight or when a visitor arrives from a new campaign, as Universal Analytics did. So GA4 often shows fewer sessions for the same traffic. User counts can also differ because GA4 reports “active users” by default.
Bounce Rate vs Engagement Rate
In GA4, an “engaged session” lasts longer than 10 seconds, includes a conversion, or has at least two pageviews. Engagement rate is the share of sessions that were engaged. GA4’s bounce rate is simply the opposite of that.
So a GA4 bounce rate is not the same metric as your old Universal Analytics bounce rate. A blog post where people read for two minutes and leave was a “bounce” before. In GA4, it’s an engaged session. I’d stop comparing the two bounce rates side by side.
Conversions
Universal Analytics counted a goal once per session. GA4 can count a conversion event every time it happens. If someone submits a form twice in one visit, GA4 may count two conversions where Universal Analytics counted one.
How I Handle Comparisons
For the next 12 months, year-over-year comparisons will be awkward. Here’s how I deal with it:
- Compare trends, not exact totals. If both tools rose 15% in the same month last year, the pattern matters more than the raw numbers.
- Use the overlap period. If you ran both tools together, compare a few months where both collected data to see the typical gap.
- Write the gap down. A simple note like “GA4 shows about X% fewer sessions than UA did” helps anyone reading reports later.
- Use Search Console for organic trends. Clicks and impressions there were not affected by the Analytics switch, so they give a steady view of SEO performance across the change.
What About Reports and Dashboards
Any dashboard that reads Universal Analytics will slowly go flat as dates move past July 2023. This is easy to miss, because the dashboard doesn’t break. It just shows zero for new periods.
I’d go through each Looker Studio report, monthly email report and third-party tool your team uses. For each one, either switch the data source to GA4 or mark it as an archive of old data. Fields have different names in GA4, so expect to rebuild some charts rather than swap a single setting.
If you use SEO or reporting tools that connect to Google Analytics, check that they’re connected to the GA4 property. Many tools added GA4 support over the past year, but they don’t always switch over on their own.
Common Mistakes I’d Avoid
A few errors come up again and again around this kind of change:
- Trusting the auto-created property blindly. It may be collecting data, but without your conversions it won’t answer the questions that matter.
- Waiting until June 2024 to export. Access ends on July 1, 2024. Exports take longer than people expect, especially if you hit sampling.
- Deleting the Universal Analytics property. There’s no benefit, and you lose access to your history straight away. Let Google’s timeline run its course.
- Leaving Google Ads on old goals. If campaigns bid on Universal Analytics goals that no longer record, the bidding has no signal to learn from.
- Forgetting the 2-month default retention. Explorations in GA4 only reach back as far as your retention setting allows.
Conclusion: Treat This Week as Your Real Deadline
Universal Analytics stopped processing data for standard properties from July 1, 2023, even if your property is still ticking along for a few more days. Your old reports stay viewable until July 1, 2024, but nothing new is coming in.
My advice is to check GA4 is collecting data, fix your Google Ads conversions, set retention to 14 months, and start saving your key Universal Analytics reports this week. If you want the background on how we got here, our article on the original Universal Analytics shutdown announcement covers the full timeline. Then pick one GA4 report a week and learn it properly. That habit will make the switch feel normal much faster than you’d expect.



