HOW TO MEASURE INFLUENCER MARKETING (THE RIGHT WAY)

For many brands, influencer marketing is a double-edged sword. It undoubtedly produces results. Just ask any Instagram user who has made a purchase because of a recommendation from their favorite influencer.

However, it's not always easy to measure exactly how effective influencer marketing really is. Marketers have long believed that influencer marketing is effective, but have been unsure how to measure its true value based on audience response.

In most cases, brands give influencers trackable links and promo codes so they can measure how many visits and conversions an influencer contributed to the campaign. However, links and promo codes only provide a piece of the story and don’t account for the true conversion value.

Only tracking traffic and conversions through links and codes is like reading the final chapter of a book and trying to piece together the entire story based on those few pages. Hooray! Luke Skywalker beat the bad guy. But who is this Darth Vader guy in the weird black suit? Doesn't really work, does it?

We know that people are purchasing, but they aren't using attributable links. To date, the best solution has been to identify them by going off of how customers who’ve been acquired tell you (aka stated behaviors), usually through a post-purchase survey. But even this approach provides a relatively small slice of the picture. After all, what percentage of customers actually take the time to respond to post-purchase surveys?

We knew we needed to solve this attribution challenge. If we knew intuitively that we saw more impact than we could currently measure, we needed to figure out a better way to measure it. We wanted to draw firm conclusions based on the data of observed behavior.

We wanted to dig deeper and look at things through the lens of the consumer, not just through the lens of the brand.

But before we could arrive at a solution, we needed to understand the scope of the problem.

LEARNING FROM TV ATTRIBUTION

Let’s set the stage by looking at how a common medium like TV has traditionally approached hard-to-track measurements. Typically, marketers will do spot attribution, because they know that they’re not capturing everybody there. People are coming in beyond those that are using a promo code or unique URL. To account for these people, they look for traffic spikes from specific locations within four to seven minutes of the spot airing.

The same dynamic occurs with YouTube, but there is a problem. YouTube views are not concurrent. Rather, they happen at separate times. As a brand, this means you can’t measure a spike because your views might be spread out over 24 hours, 48 hours, or even up to a month or beyond (especially for the long tail).

What does this mean? It means that there must be a ratio that exists between those who use some form of promo code and then those who just visit either through a direct URL or search. Additionally, when you have a firm grasp on what that ratio is, you can measure your marketing efforts with much more specificity. The problem is, the only way to get a grasp on that data is to look at things from the consumer’s perspective. No attribution data on the brand side can account for every step of the online journey the consumer is taking.

Our hypothesis was that if we could start with data that reflected the consumers’ perspective, we could get the full picture of what is happening. So, to prove this hypothesis, we partnered with the world's largest clickstream data company. They had access to billions of users who had signed up for internet security, web monitoring, and browsing and provided anonymized data with all sensitive data stripped out of it.

The company typically works with brands to (1) help them understand what their customers did in the three to seven days prior to purchase, (2) develop behavioral science around their customer behavior, and (3) identify the clues that led to the purchase.

We flipped that around and gave them 150 different YouTube integrations we had previously run. We asked them how many people in their database of approximately 10 million had seen the video and the number of individuals out of that group who had purchased. Of those 10 million people in their database, their data turned up about 600,000 folks who had witnessed one of our integrations (you can check out the full white paper here).

Based on that data, we sought to establish a ratio between the number of people who watched a video and then used a unique URL to purchase and how many people visited a site through a search or direct URL. To ensure we had the proper sample size, we looked at the data over a period of two weeks.

In addition to tracking those who visited or purchased directly or through a search engine, we also tracked individuals who purchased via Amazon, Best Buy, or Target.

WHAT DID CUSTOMER DATA TELL US?

INFLUENCER DRIVES MORE TRAFFIC AND CONVERSIONS THAN WE REALIZE

Based on the data, we were able to uncover some truly amazing insights. First, for every trackable visit, there were at least five unattributable visits. And for every trackable conversion, there were at least three more unattributable conversions.

In short, typically about 50% of the traffic and conversions that brands see come from that dark pool of unattributed traffic conversions. This sheds a light on where some of that is coming from. Yes, a lot of your other channels want to claim that the entire 50% of traffic is attributed to them, but this is actually proven by data from actual consumer behavior.

The bottom line? Influencer marketing is more powerful than most people even realize. When you run an influencer campaign, you’re getting 5x the traffic and for every conversion, you’re getting 3.5x more conversions.

And you can expect those same kinds of results regardless of your brand's product or service. In our tests, the results were remarkably consistent across brands, timeframe, and types of products.

IT DRIVES SUCCESS ACROSS THE ENTIRE FUNNEL

Not only does influencer drive more visits and sales than brands typically measure, but it also impacts a broader set of metrics. Instead of brand campaigns focusing on impression metrics and performance campaigns focusing on conversion metrics, we see a positive impact on metrics across the funnel, regardless of the goal.

Performance campaigns will increase awareness and awareness campaigns will frequently impact conversions. In short, influencer campaigns don't neatly fit into either branding campaigns at the top of the funnel or performance campaigns at the bottom of the funnel. Rather, they produce results across the entire funnel.

From driving awareness to ongoing education to generating conversions, influencer campaigns are effective. To determine the value of influencer campaigns you have to measure holistically across the entire marketing funnel.

HOW ARE YOU MEASURING INFLUENCER MARKETING?

So what's the big takeaway from this information? Most importantly, when you under-measure the results of a campaign, you're likely to under-invest in that channel. It's like investing in the stock market. If you put your money into a stock and that stock tanks, you're probably not going to keep putting money into it.

But what if it turned out that the stock was actually outperforming the market by 400%? Because that's effectively what happens with influencer campaigns. When you look only at attributable results, you're getting about 25% of the entire picture. Influencer campaigns regularly produce up to four times the measured results. If you saw that one of your campaigns was producing 4x the measured results, you might invest in it differently, right?