GA 360 Roll-Up Reporting: What Is It Good For? Absolutely … Something.

The day Google announced Google Analytics 360 Roll-Up Reporting I was super excited. Finally, a way to consolidate enterprise data without disrupting the delicate ecosystem of how other teams access and use the same data.

The day Google announced Google Analytics 360 Roll-Up Reporting I was super excited. Finally, a way to consolidate enterprise data without disrupting the delicate ecosystem of how other teams access and use the same data.

    The benefits of using Roll-Ups are huge: They support a cleaner analytics deployment and easier governance by removing parallel trackers; reduce hit “cost” because Google Analytics 360 Roll-Ups only incur .5 hits; and provide a flexible way to aggregate different business elements (by geo, brand, main conversion type, etc.) without requiring changes to the architecture of how you have deployed properties.

    That said, I have seen a lot of big clients struggle to get started with using them. People like the idea of seeing all their data in one place, but have a hard time determining and refining what they’re trying to accomplish, how to execute against those goals, and what the use cases are for the output.

    Before you even think about using a Roll-Up report at your enterprise, you need a plan. Making a plan is a lot like making a pie. It seems simple when the finished product is sitting there in front of you. Actually making the pie is a lot less straightforward. To avoid wasting time, money, and jeopardizing your existing reporting systems, you need to plan for what you’re going to aggregate, how you’re going to do it, and where you’re going to apply the data once you have it.

    This series of three posts (two more are still baking!) will address each of those issues in turn. In this first post, I’m going to emphasize the most important component of Roll-Up strategy. This is more conceptual than the other steps, and will essentially deal with my favorite question — Why?


    What to Consider When Building a Google Analytics 360 Roll-Up


    Like every three-year-old since time immemorial, “why” has been my favorite question.  Often it drives people crazy, but it’s probably the most important thing standing between having a plan and having a good plan. Why is having a Roll-Up important for the business? This is where you should just stop until you get a really good answer.

    The answer to ‘why’ is the thing that’s going to determine everything else. Have a very clear understanding of what it is you’re trying to accomplish and how a Roll-Up is going to help you answer questions to get there. And if the answers to those questions aren’t immediately obvious, maybe it’s better to roll back to fundamentals.

    Why would anyone want to make a Roll-Up? Roll-Ups are valuable tools for aggregating and comparing generally equivalent data points that exist in otherwise independent data sets. But you have some homework to do before you can use them.

    Are you ready to roll?

    Data trust

    First and foremost, do you fundamentally trust your data? A Roll-Up is only as good as the source properties it is made from. This is a good opportunity to clean house before building an addition. If you have ecommerce reporting, do these numbers come close to what you report in your back office systems? If you drive leads, are the numbers in Google Analytics similar enough to what goes through the call center that you are willing to use this information to make decisions about budget? Building trust in your analytics data is vital, and one of the best ways to do that is to make sure you have a generally equivalent measurement somewhere else in the business.

    Data structure

    Consider the logical structure of your data, as this is one of the most fundamental pieces of successful Roll-Up reporting. This means that, for example, across all your sites if you have a lead generation conversion it should be tracked in a way that is logically and practically consistent. If you are measuring leads as total events on site 1, and leads as a destination goal on site 2, you are measuring and comparing fundamentally different things. Decide on a method and naming convention, and use that logic across all your data.

    Data consistency

    Consistency is a big deal for product data as well. Is the same product called Banana Blouse on site A and Banana Camisole on site B? The same goes for channel groupings. If different source properties have different default channel settings, how can you best aggregate these into buckets that are consistent and meaningful for your business? What about custom dimensions and metrics? Though you can map different custom dimensions to a common index at the Roll-Up level, it is still important to build a strategy for what the custom dimensions that are meaningful across source properties will be and what you will call them.


    When I’m planning for a Roll-Up, typically I will start with a document that lays out the structure to answer these questions.  I will  first determine the primary use case, and what source properties will be added to support that use case. Next, I will audit any key conversion metrics and determine if they are sufficiently similar to be aggregated in a meaningful way. Then I will determine what custom dimensions will need to be established and what their equivalent mappings from the source properties ought to be. I will plan out what data integrations will need to be made and document these, as well as a plan for any goals that apply at the Roll-Up level.

    Use Cases for Roll-Up Reporting

    A helpful way of answering the “why” question is to determine use cases for the Roll-Up at your enterprise. This list is by no means exhaustive, but here are some common use cases for Roll-Up reporting that I’ve run into with our clients.


    You have multiple brands with common core business objectives

    Your business has multiple brands, each with a subdomain dedicated exclusively to lead generation. You would like to understand the sum total of your leads, as well as which brands are performing better at driving engagement compared to your other brands with the same business objective. But your account is structured in such a way that each brand falls under a different property. While it’s a manual and time-consuming process to aggregate your lead gen data across brands with this set up, it serves most reporting needs better than if all brands were in a single property. Also, retiring numerous properties to unite all of these sites under a single UA-ID would mean that it would be very manual to do year-over-year comparisons with data that has been migrated from a retired property.

    With a Roll-Up, you have the flexibility to choose source properties and have them report to a single place, without changing the architecture of your account set up. This facilitates ongoing use of the data as you’re used to seeing it, and provides the cross-brand lead gen view that you had otherwise lacked.


    You have multiple brands with residential/retail and commercial customers

    In much the same way as described above, you may have an account that reports different brands under a common property ID. However, suppose that you have a few brands and each of them have both a residential and a commercial side to their product offering. You may want to look at data across brands for just commercial, or just residential customers.  Aggregating a business in this way would be valuable because you would get a much “truer” view of averages such as ecommerce conversion rate, avg. order value and avg. quantity when you are looking at customers who you can expect will behave in fundamentally similar ways. Roll-Ups provide more informative and actionable data about different customer groups while allowing you to maintain brand-level account structure.


    You have multiple sites and want to assess the affinity of users between those sites

    In a typical account, if you manage multiple properties you aren’t able to understand to what extent you have users who are visiting multiple sites. Instead, a user who visits two of your sites independently of each other will look like two users, when they are really one. This means you’re missing out on a valuable opportunity to target specific content and messaging to these users, and understand to what extent their interacting with one of your sites may lead to conversions on another site, which could drive different goals and strategies for the sites as digital assets. With a Roll-Up, if you have the same client ID for a user, you are able to understand their journey across your various properties. This provides meaningful data to drive retargeting efforts, goals and resources for each of your sites.


    You have a group of sites that you want to see regionally

    For companies with a global presence, it can be difficult to see how specific regions are performing compared to others. If you have a different property for each country, for example, a Roll-Up can make it possible to aggregate your data by geographic region in ways that make sense to your business strategy. If you are able to connect those disparate properties into a single Roll-Up and see the entire view of APAC, or Netherland, or North America performance, you can more easily tie your digital data to meaningful information in the finance department about the allocation of budgets, merchandising choices, or the development of new regional strategies.



    Ask why. Ask it until you get a good answer, even if you’re talking to yourself (don’t worry, I won’t tell).  Do not build a Roll-Up until you know what questions you’re trying to answer by aggregating your data, and what you will do with the information when you get it. Make sure your data is in a good place to be aggregated.  Plan your attack based off the answer to “why”.

    Tune in for the next installment in this series which will cover the actual mechanics of how to do it.

    In the meanwhile, if you have questions about what you just read, don’t hesitate to give us a shout.


    Latest Articles

    Ricardo Cristofolini July 3, 2024
    Shreya Banker, Data Scientist June 20, 2024
    Ricardo Cristofolini June 10, 2024

    Ready to grow your company to the next level?