How To Back-Up And Report On Universal Analytics Historical Data With GA4?

As a product manager who has worked with several enterprise companies, I personally believe Google’s shift from Universal Analytics (UA) to Google Analytics 4 (GA4) has been one of the worst I’ve ever seen. There are several reasons why…but here are the biggest gaffs:

1. Historical Data – Google Analytics Universal was retired in July 2023, and you can only access your UA data until July 1, 2024. After that date, your UA data will be deleted. Poof… gone!
2. Historical Analysis – None of your historical data from UA is available in GA4 by default. While I understand the methodology and data structure for the two platforms have significantly changed, it’s simply unacceptable that companies won’t be able to do simple month-over-month or month-over-year reporting with current data and data that was in UA by default.
3. GA4 Data Retention – Many companies don’t realize that GA4 launched with data retention set to 2 months. You have to modify the data settings to 14 months manually. You should do that in every property you have today!
4. Data Limitations – There are also limitations on the amount of data that can be stored in GA4. For example, there is a limit of 50 event-scoped custom dimensions and 50 event-scoped custom metrics per property.

Given these incredible roadblocks and limitations, there’s only one choice that every company needs to make to maintain historical data and to be able to utilize it moving forward:

What To Do About Your Universal Analytics Data

If you wish to be able to utilize historical data from UA, you have to:

1. Back up your UA data.
2. Normalize the data for use with GA4 data.
3. Connect it to a BI solution where you can easily query and report on the data.

If that sounds simple, it’s anything BUT. I’m, again, going to throw my opinion in here. While Google speaks largely on the GA4 launch as a response to regulatory issues and privacy, I personally believe they couldn’t stand that they had a product that was critical to the average business that Google made no profit on. Sure, some enterprise companies utilized Google 360 but the 5-digit licensing was not in the budget for most businesses.

And moving forward, with GA4 only having a maximum of 14 months of data, companies will be forced to incorporate a database and reporting interface moving forward. Conveniently (sarcasm), Google’s BigQuery and Looker are ready for companies to adopt. And, of course, this will allow Google to monetize the use of GA4 moving forward.

To compound things, we’re even seeing that GA4’s free interface has massive delays in activity… while synchronized solutions to BigQuery are operating in almost real-time. Simply put… they want you to pay for Analytics moving forward… one way or another.

The GA4, BigQuery, and Looker Solution

While you can find other solutions, the native connectivity between GA4 and Google’s platforms is already there and will only get stronger moving forward. It’s likely going to be the best solution… but let’s throw another huge roadblock in… how much will you spend on BigQuery and Looker for backing up, processing, and reporting on data. Let’s first break down their pricing:

BigQuery Costs

BigQuery charges for both storage and processing.

• Storage: BigQuery charges a flat rate of $0.02 per GB per month for all stored data. This includes both active and inactive data.
• Processing: BigQuery charges for the amount of data that is processed by your queries. This includes the data that is read, scanned, and written. The processing costs are based on the following factors:
  • The amount of data that is processed
  • The type of data that is processed
  • The complexity of the query

Looker Costs

Looker charges for the number of users, the amount of data stored, and the number of queries run.

• Users: Looker charges a monthly fee for each user. The fee is based on the user’s role, such as admin, editor, or viewer.
• Data: Looker charges for the amount of data stored in your Looker instance. The fee is based on the size of the data, the type of data, and the frequency of access.
• Queries: Looker charges for the number of queries run in your Looker instance. The fee is based on the complexity of the query and the amount of data processed.

So… How Much Are You Going to Spend?

If your company has no experience in backing up, processing, and reporting on this data utilizing Google’s solutions, you could easily spend tens of thousands of dollars on getting this integrated solution up and running. Why?

• You don’t need ALL your data! You may only need a fraction of your data to analyze historical data. Planned effectively, you can back up summary data by a period for channels, campaigns, events, and other elements.
• You don’t need to PROCESS all data! If you plan effectively, you can substantially minimize the costs that will come out of Looker for the reporting you need moving forward.

We recommend working with a consulting firm with the tools and understanding of the data that can interview your business and ensure the right solution moving forward. We’d recommend doing that with an ongoing engagement rather than a one-time push.

Why? Because every month you’re working within these solutions, building reports, and processing data, you risk driving up your costs without any return on investment.

Martech Zone’s sponsor, DK New Media, can assist you with your GA4 backup and reporting through their partnership with a leading Analytics firm and their solution. They developed a simple framework and process to help marketers, analysts, and business owners make it easy to back up their Universal Analytics data.

Google has said that they will not create a Universal Analytics backup solution after permanently moving everyone over to GA4. And because of this, companies need ways to access that old data.

Tim Flint

Using A Managed Service To Back Up Your UA Data

This is a managed service that makes it easy for you to back up your GA data. It is a simple 3 step process:

1. Determine what you want to be backed up. Universal Analytics has so much data; unfortunately, Google doesn’t make it easy to grab all of it. So choose either predetermined reports that we provide, or work with our team to create custom reports to ensure you get every user, custom metric, custom dimension, event, and more out of the system.
2. Determine where you want it backed up. Choose between Google Sheets, Excel, CSV, or – our favorite – BigQuery to send your data to. BigQuery is the best choice for flexibility. Creating an endless time series graph with UA & GA4 combined makes it much easier. Don’t know how to set up BigQuery? Don’t worry. We will handle that for you and give you 100% control.
3. Determine how you present your data. How you access and explore this old data is important. Since this data is going to be in complex tables and not easy-to-use graphs, you have several options, including connecting your data directly to Looker, which allows you to explore and visualize data. You can also make reports using pivot tables in Excel, Google Sheets, or any other visualizer you wish.

A few days ago we were working with a client and were trying to understand why a certain time period had been their most successful. We wanted to be able to replicate that success. If we don’t have a backup of that data, we wouldn’t be able to ask the clients deeper questions to help us find a solution to improve their marketing.

Tim Flint

There are so many reasons why companies need to back up their data. And most revolve around ensuring that you don’t make the same mistakes twice. So here is our list of why you need to get started on backing up that Universal Analytics data before it is too late.

Why it’s important to keep historical Google Analytics data

• Your CEO will ask for historical data to help them make decisions and look at trends. Historical data is needed to help you understand your marketing progress.
• Historical data enables you to understand seasonality. For example, want to know if running that Mother’s Day sale makes sense?
• Historical data helps you know the success or failure of past tests so you don’t have to take the same test twice or waste money on failed projects.
• Data helps your team get onboarded more easily and understand website history, saving valuable training time.
• Historical data helps your third-party partners, such as agencies, quickly know what to do.
• Strong data can help your analysts do better research projects and find correlations that could be helpful to your business.

Bottom line: What will you say when your CEO asks for historical data to compare when things are going poorly? Will you have the data? You only have until July 1st of 2024. The process isn’t simple and we can help – schedule a free consultation with us today!