Incrementality testing has started to become the new hot buzzword for marketing measurement. But as marketers hunt again for the single, “silver bullet” to solve their measurement challenges, they’re finding a blended approach that includes incrementality testing might be the things that works best.
There has been a lot of discussion around marketing measurement over the last 20+ years of digital advertising. After all, this was supposed to be the most measurable form of marketing that could accurately track every ad impression, click, and all engagement completely time- & date-stamped, connected back to unique users.
That was the promise, anyway. But still, marketer survey after marketer survey continues to report that measurement is still not figured out.
So, the key question persists: What is the right way to measure ad effectiveness?
Attribution was supposed to light the way
Back in the late 2000s, the rise of multi-touch attribution modeling shed light on the problems of “last click tracking” and how marketing measurement was broken because it gave all of the credit of a conversion to the customer’s last interaction. Marketers rejoiced because Measurement had been solved and attribution was put on course to be the accepted industry solution.
But, ten years later, few today would say that MTA has fully cracked this puzzle.
A decade ago, in a desktop-centric, cookie-driven web, tracking user touches with online marketing was actually pretty easy. The wrinkle is that at the same time that multi-touch attribution was being figured out, the proliferation of web-enabled devices—especially smartphones and tablets—changed the game. Almost overnight, a user’s online ID was spread across multiple devices and, unfortunately, cookies tend to crumble between desktop and mobile browsers, thus creating a lot of problems for attribution solutions.
The truth is, marketing attribution as a standalone solution peaked in 2007, right before the mobile explosion. In a recent Kantar study, marketers reported that out of all of the most common measurement methodologies, attribution has the most perceived gaps.
Attribution certainly has its merits, but also its limitations. Marketers are still looking for their silver bullet to solve marketing measurement.
However, could we be just not asking the right question?
Maybe the answer to marketing measurement isn’t a silver bullet at all.
A blended approach vs. a silver bullet
A quick review of measurement methodologies outside of the marketing industry shows that the general standard is not a single solution, but rather a blended approach where numerous viewpoints are combined to form a macro-thesis supported by multiple data sources.
A perfect example of this is the Leading Economic Index (LEI) used by many economic participants to predict what will happen with the economy in the near future. By analyzing the index in relation to the business cycle and general economic conditions, investors and businesses can make better-informed decisions. It’s not perfect, but it’s useful: this single index value has generally proved capable of predicting recessions and other major market fluctuations over the past 50 years.
The LEI is a single score composed of ten economic components whose changes tend to precede changes in the overall economy such as:
LEI Ten Economic Components
- The average weekly hours worked by manufacturing workers.
- The average number of initial applications for unemployment insurance.
- The amount of manufacturers’ new orders for consumer goods and materials.
- The speed of delivery of new merchandise to vendors from suppliers.
- The number of new orders for capital goods unrelated to defense.
- The number of new building permits for residential buildings.
- The S&P 500 stock index.
- The inflation-adjusted monetary supply (M2).
- The spread between long and short interest rates.
- Consumer sentiment based on a household survey and are reported on a monthly basis.
As you can see, any one of these single data points just wouldn’t be enough to depict an accurate view of what’s going on with the economy. Only by looking at all of the metrics together does the real story begin to emerge.
And the data is just the start. The economic experts who decipher these economic indicators have lifetimes of experience so they can add context to the numbers in order to better understand them. In a recent press release, The Conference Group—the non-profit organization that produces the LEI— notes that the US government shutdown was having an impact on the economy, but just a short-term impact that may throw off the numbers now, but shouldn’t have a long-term negative impact on the economy.
The right question to ask about marketing measurement?
Maybe the right question is not “what is the best single form of marketing measurement?” but rather “should a blended approach be the standard?”.
Today’s marketers have a variety of ways to measure their efforts. Attribution is one. There’s also web analytics to understand the behavior of people once they get to your owned web properties. There are also in-channel metrics such as SEM impressions, clicks, average CPC, CTR, etc. There are homegrown solutions and media mix modeling vendors.
More recently, incrementality testing has emerged and proven promising. Incrementality testing compares the marketing results between a test group and a control group. Using this method, marketers can easily isolate the affected variables. They are able to clearly assess immediate business impact and formulate data-driven actions to take with a high degree of confidence.
Some have lauded incrementality testing as the new marketing attribution, a single solution that can get to the measurement nirvana that marketers really want.
But, as good as incrementality testing appears to be at answering some of the harder questions of advertising effectiveness, maybe its best role is not as the silver bullet, but an additional leading indicator that marketers can use along with other existing data points to uncover real insights—as attribution for your attribution model, if you will. Alongside A/B testing that helps to refine creative, auditing and analytics, and, even, sometimes, a good, old, occasionally shaky MTA model, incrementality can be leveraged to pressure-test marketing measurement, highlight discrepancies and isolate impact drivers. It can be the thing that makes new investments, deeper investments, and alternate investments less risky, so you can experiment quickly with a higher degree of confidence.
It might be time to change your view of marketing measurement and try a blended approach that incorporates incrementality testing. While there may no single silver bullet, it’s an arrow you’ll want in your quiver.