Known knowns and known unknowns in influencer marketing

‘...we know there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns – the ones we don’t know we don’t know.’
— Donald Rumsfeld

In influencer marketing there are many known unknowns we would like to know. As more and more money flows into the industry, pressure to prove its effectiveness builds.

Statistics with dollar or pound signs in front of them are the most alluring, and are often the most fallible. Return on investment (ROI) is sought across all channels by marketers trying to allocate tight budgets optimally.

Sophisticated methodologies have been developed to attribute sales value back to marketing investments. Econometrics / Marketing Mix Modelling is well established for traditional above-the-line media and digital attribution modelling is getting there.

But influencer marketing is still in its infancy. And no existing methodologies exist for accurately attributing sales or return on investment to influencer campaigns.

In the absence of real ROI numbers, some data and insights providers have attempted to conjure up (and sell) proxies. Chief among these is so-called earned media value (EMV).

emv (1).jpg

Pursuing bad metrics like EMV leads to bad decision making. A much better approach is to get hold of reliable, accurate data that correlates strongly with your ultimate goals, then build data around it that improve your understanding of what’s going on.

Known unknowns: why EMV is flawed

One provider describes ‘earned media value’ (EMV) as their ‘holistic approach’ (alarm bells ringing), which ‘assigns a specific dollar value’ to content based on the ‘perceived value of digital word-of-mouth to brands’. It is an attempt at estimating what would you probably have had to pay to get the same number of ordinary paid impressions.

The problems with EMV are many. Firstly, impressions are hard to come by on most social platforms. If you have a business account with Instagram you can login and get impressions numbers on your own posts. Most third party data providers (us included) do not have access to this.

Even if EMV values accurately represented the equivalent cost of ordinary paid impressions it would still be a misleading figure. The whole point of influencer marketing is that third party influencers are perceived as having more credibility than a brand promoting their own products. Influencer impressions should be more valuable.

And no one would say the return on investment on paid social equals the cost. If it did, no one would do it.

money1 (1).png

What’s more, the economics are different. What determines the price of a paid impression is (almost) independent of what determines the price of an influencer impression. The price of a paid impression on Facebook is presumably determined by the supply of paid space (both on FB and other platforms) and the demand for paid impressions. The price of an impression on an influencer post is determined, for the time being, by the negotiating power of influencers and their agents.

And of course impressions don’t equal or even necessarily correlate with sales.

The ideal

In an ideal world we’d be able to attribute profit right back to decisions made in the planning stage. We’d be able to separate out marketing effects from price and other covariates. We’d then be able to split marketing effects into all the constituent parts: TV, print, digital etc, and understand the synergies between them. We’d then be able to explore within digital, and separate out the impact of influencer marketing vs. other paid, owned and earned channels. We’d be able to go deeper still, and understand which influencers drove what percentage of the overall impact and, crucially, why.

To get to this point we need a lot more data than is currently available.

Known knowns

There is plenty of evidence influencer marketing in general has a positive impact on brand KPIs. The data easily available are engagement numbers - counts of likes and comments on posts, view counts (YouTube) and follower counts. From these we can calculate engagement rates: the ratio of engagements to followers.

The strategy should then be to build up data around our engagement rates, which explain and perhaps predict differences in performance. We can use these data to adjudicate campaign performance on a level playing field: where does the campaign’s performance rank within the recent history of campaigns in its vertical? Which post attributes correlate with high performance: do posts shot outside outperform those shot inside? Do black and white posts outperform colour? What is the benchmark average performance of makeup posts shot in bedrooms? And so on.

This metric is a few miles away from our ideal, but it is at least consistent and we can be reasonably confident it is correlated with our KPIs of interest. Even diving deeper within engagement rates can provide greater insight into the success of campaigns (e.g. where are the engagements coming from, what is the sentiment of comments, etc.). For more detail on how CampaignDeus helps increase transparency around engagements, get in touch.

CampaignDeus is the leading independent provider of influencer marketing campaign data for Instagram and YouTube. Our platform identifies and classifies brand sponsored influencer campaign performance metrics, tracking hundreds of thousands of posts.

We use this data to provide Brands & Agencies with industry insights across verticals, benchmark campaigns against vertical & competitor averages, and equip clients with in-depth reporting and recommendations on how to make campaigns more effective. Get in touch for more details.

Duncan Stoddard is the co-Founder and Chief Data Officer of CampaignDeus. Previously a data scientist with DS Analytics, AlphaSights and media agency Mindshare, he specializes in predictive modelling, forecasting, optimisation and data visualisation.