Incrementality testing can sound like something reserved for large brands with big budgets, specialized agencies, and in-house data science teams. Search “incrementality testing,” and you’ll find lots of companies offering sophisticated platforms and methodologies to measure the true impact of marketing spend.
But the concept itself is pretty simple. Incrementality testing measures what happened because of your advertising versus what would have happened anyway. Put it another way: it measures the difference in revenue (or sales, or leads) when an audience sees your ads versus when they do not. The “incrementality” is additional business your advertising created beyond what you still would have gotten due to organic presence, word of mouth, or simple brand recognition.
Here are some cost-effective ways small- and medium-sized businesses can utilize incrementality testing:
Geo Testing
For a small ecommerce company that ships products nationwide, a good incrementality test would be to find a pair of markets that historically generate a similar amount of revenue and run ads in one, while shutting them off in the other for a set period of time.
Let’s say the Houston and Nashville markets each generate $1,000 in revenue each month, and the monthly ad spend in each is $200. Don’t spend that $200 a month in Houston for a month or two and see how much revenue you’re still getting from Houston each month. The difference between the initial $1,000 and the monthly revenue during the test is your true incremental revenue from your ad dollars. Just make sure to keep in mind other advertising you might be doing in a market (radio ads, email, etc).
And you don’t have to market to a national footprint either. If you’re a home services company, run an incrementality test in two counties you service with similar sales numbers.
Product-Based Testing
That same online retailer I mentioned could run a product-based incrementality test. They could run ads for one group of products and not for the others.
However, be aware that the data won’t always be perfect. Just because someone clicks on an ad for one product doesn’t guarantee that will be the product they choose once they visit the website. Ad clicks for shopping are just the entry point to your website, and you don’t know what they actually bought unless you have cart data conversion tracking enabled on Google. (Cart data conversion tracking allows you to see the product that the user clicked on, and the product that they ultimately purchased.)
Blackout Testing
I have always found that running branded ad campaigns is a low-cost way to play defense against your competitors’ ads. But I occasionally have clients who do not see the need to run paid ads for their brand name, expecting their organic listings to capture all of that traffic for free.
You can find out definitively if that is accurate by running a branded blackout. Stop running branded ads for a period of time and then compare your overall sales before and after the test. You’ll be able to determine if the relatively low amount needed to sustain a branded ad campaign was worth it to capture the difference in revenue.
Another Account Manager here had a client who absolutely hated spending money on their own brand name and frequently discounted the value. They ran a blackout test, turning off brand advertising for a month, and compared their organic traffic and sales before and after. The revenue loss they took from not running ads on their brand was not made up by additional organic sales.
And it doesn’t have to be restricted to just branded campaigns.
Years ago, when conversion tracking wasn’t at the level it is today, or lead gen clients wouldn’t trust using call tracking numbers for optimization purposes, some clients would question the value of their ad spend. I would recommend that we turn off the ads for 30 days, and inevitably the client would reach out before the end of the 30-day period and ask that we turn the ads back on because they are seeing a drop in business. I guess I was doing blackout tracking at that time without even realizing it!
Platform-Based Conversion Lift Tools
Google and Meta both offer Conversion Lift tools to measure incrementality. If you’re looking for some incrementality testing without much data gathering on your side, these might be good to try.
Customer Surveys
“How did you hear about us?” questions during the purchase cycle may offer some broad insights into incrementality while not costing much to enact, and also not requiring you to abstain from marketing to a particular audience segment during the process. If 50 people stated that they heard about you from Facebook, but Facebook’s pixel is claiming 300 conversions, that’s a sign that your results on that platform may be over-reported. But this one is also problematic because a lot of people say they “google” something as a verb, not as the actual platform. We’ve had clients use this and discount other platforms; turning off those other platforms caused a drop in overall site traffic and business.
Putting Incrementality Testing Into Practice
Incrementality testing doesn’t have to involve expensive software, complicated statistical models, or a team of data scientists. At its core, it’s about asking a simple question: What would have happened if I hadn’t spent that advertising dollar? Small and mid-sized advertisers may not be able to run perfectly controlled experiments, but they can still use practical tests like geo holdouts, product exclusions, branded blackouts, platform lift studies, and customer surveys to get much closer to the answer. Even an imperfect test can provide valuable insight into which advertising dollars are truly driving additional business, and which may simply be taking credit for conversions that would have happened anyway.
If you’ve ever wondered the true value of your advertising, give one of these incrementality testing options a try. No army of number-crunchers required.