Google and modern marketing measurement: the measurement tripod

By John Koetsier April 29, 2024

Google sees modern marketing measurement as a tripod. The 3 legs of the measurement stool, as Mountain View sees it, are attribution, incrementality, and marketing mix modeling (MMM).

Google just released a 44-page report titled the Modern Measurement Playbook, sharing how it sees the future of paid media attribution and analytics. Since Google is the world’s largest adtech company by revenue — $307 billion for 2023 — and probably the single biggest ad partner for Singular customers by spend, it’s worth understanding how Google looks at marketing measurement in general, and paid media attribution in particular.

Google’s modern marketing measurement tripod has 3 legs:

  1. Attribution
    Micro focus: Assigning credit for conversions
  2. Incrementality measurement
    Meso focus: Determining impact of marketing
  3. Marketing mix modeling (MMM)
    Macro focus: Understanding all factors for revenue

3 forms of modern marketing measurement

The key is understanding how all of these measurement methodologies interact.

Attribution is the most immediate and it’s the one performance mobile marketers and user acquisition managers are most familiar with. There’s a click or a view, and then something happens: an install, a re-install, an engagement, a conversion. The click or view connects action and reaction, or input and output, and you have near-instant signal.

That signal, of course, is changing thanks to privacy, but SKAN and eventually Privacy Sandbox will still provide an attribution signal of varying levels of value.

Incrementality measurement doesn’t capture the immediate causal impact of clicks or views, and that is both its biggest weakness and greatest strength. It’s a weakness, from the perspective of those who want a direct connection in time and space between a marketing stimulus and a consumer response. It’s a strength, from the perspective of those who point out that just because an ad impression or click occurred at some point on the customer journey, other factors were at play. And the conversion may have happened in any case without the ad view or click.

At its best, incrementality measurement tells you the additional value of ad spend and media sources by comparing what happens for specific sets of people or geos who are exposed to marketing campaigns versus those who were not. Ultimately, incrementality measurement should reveal what you got for all your marketing effort and energy that you would not otherwise have achieved.

Media mix modeling is the longest-range, broadest-scope, and highest-level measurement methodology that Google incorporates into its tripod theory of modern marketing measurement. MMM takes into account all marketing efforts including paid ad campaigns and non-paid organic efforts, but it doesn’t stop there. 

None of your marketing efforts happen in a vacuum. Everything around — world events, weather, competitor activity, economic and financial changes — impacts sales and revenue.

Modern marketing measurement: putting them all together?

Google suggests the way forward is using all 3 modern marketing measurement tools together, where attribution gives you real-time data, MMM provides a macro view of everything, and incrementality ensures you only pay for marketing that works (and partners who are effective).

The big question for a digital marketer, however, is whether you need all 3.

At some level, scope and scale provide insight on which tools you need in your bag.

A startup just beginning with modest scale will certainly waste its time with old-school MMM and is likely to find that sophisticated incrementality testing is out of scope. (Simple incrementality testing, however, like just turning the money tap off for a few weeks here and there, is not, and should be done from time to time.)

Even relatively large mobile gaming and app studios may be likely to find that traditional MMM is too big of a beast. (Note: Google suggests the time scale of MMM projects is “usually two years.” Singular offers MMM for mobile marketers that requires less … though it is 100% true that the longer you have data for, the better MMM works.) At this level, however, incrementality is a key way of discovering which ad partners are truly delivering value.

That said, there are many who don’t dive this deep into all the elements of modern marketing measurement.

Ultimately, it’s at your discretion. And a function of how much time, budget, and resources you have to invest in optimizing your marketing investments. Use the ones that make sense, but make sure you understand both the strengths and the limitations of each to avoid over-correcting course when you get results that don’t match your preconceptions.

If you do use all or several of these, Google suggests using each to calibrate the others. One way is to set the bounds as sanity checks on attribution methods.

As Google says:

“For digital click-based channels, in-house attribution typically represents a generous view of that strategy’s contribution (i.e. the upper-bound), and incrementality represents the most conservative view (i.e. the lower-bound). Your MMM’s assessment should fall within the range of these bounds.”

When you put them all together, your marketing plans set the strategy and allocate budgets for various channels, separate out goals for performance versus brand channels, and enable in-channel optimization. 

There’s a universe in which you have all the resources and time you need to create the ultimate modern marketing measurement machine for your specific company and marketing mix. 

Unfortunately, it may not be this universe.

Marketing maturity for you is to adopt the thinking, processes, and tools that make sense for your specific situation.

What Singular is delivering, starting first with iOS, is Unified Measurement. Unified Measurement is a recognition of the fact that while attribution is just 1 leg of the tripod, in itself it contains a world of diversity in methodologies, signals, and strategies. Unified Measurement brings those together, but it doesn’t stop there. Attribution at its best is very good, and Singular’s new Unified iOS Report is indeed now the gold standard in attribution, but there’s more needed.

Incrementality matters. Marketing mix modeling matters.

The insights you can derive from those methodologies have plenty to say about which partners and channels you should use, and can feed into a better marketing strategy. That’s why you can already tap into those capabilities as a Singular customer in a super user-friendly way: Singular is working to incorporate the insights they deliver into Unified Measurement.

An interesting insight from Google highlights how critical the attribution stage is:

“Regardless of the maturity stage or the number of tools you are using, when it comes to optimizing campaigns and tactics, attribution will be the best tool to prioritize since it allows you to monitor relative changes in the performance of your chosen KPI in real time.”

Ultimately, being able to pair that with higher-level methodologies provides insurance that your attribution results accurately reflect reality … especially the parts that are invisible to today’s tracking and measurement methods, like multistage customer journeys that are influenced by offline or disconnected events you will never see.

Much more in the report

If you are using incrementality or MMM, it’s worth checking out the entire Google report, which offers insight on using those methodologies to help you set bids, understand the true value of paid media (which can be greater or lower than attribution tools alone suggest), set hypotheses to test, and more.

There’s also tons of detailed examples of how to use the methodologies together.

Ultimately, it’s a fairly strong overview of modern marketing measurement. That’s not just about what to do as granular signals fade. It’s also about what should always have been paired with direct attribution data.

Stay up to date on the latest happenings in digital marketing

Simply send us your email and you’re in! We promise not to spam you.