Influencer marketing ROI

A recent report from Digiday which looked at influencer fraud, highlighted that a single day’s worth of Instagram posts tagged #sponsored or #ad was found to contain over 50 percent fake engagements. Of 118,007 comments, just 20,942 were not made by bot followers.

It’s not surprising that an increasingly competitive industry has fostered dirty tactics to help influencers lure in sponsorship deals. But, of course, bots don’t click, buy, comment or think; because they’re not real people.

As influencer marketing thrives and more marketers look to invest in the value of influencers in driving awareness and establishing genuine legitimacy, we see an increasing focus on using a plethora of data to establish the most effective collaboration, and, by proxy, the most engaging content. But as the landscape moves faster than many are able to keep up with, we must ask: are we focusing on the right data to get our money’s worth?

Influencer marketing demands a robust approach which uses real-time data to identify the best content creators to create sponsored content that truly resonates. Influencers themselves are under more scrutiny than ever- particularly given controversies around Logan Paul et al, and the imperative deep delve into historic content and performance.

But part of this process, and a box that is often left unticked, is close scrutiny of the influencer’s audience.

Without a clear indication that audiences could be potential customers, the content could (and is likely to) fall on deaf ears.

It’s widely understood that an influencer’s audience demographics are fundamental in identifying whether content will speak to the right people.  Depending on the content platform, there are many tools to help with this- Peg.co, for example, offers a breakdown of audience age, gender and location.But we need to go deeper. We need to be looking at whether the audience is an audience at all.

Whilst it is certainly difficult to accurately determine fake followers, due to increasing sophistication of bots and changing privacy settings from the likes of Instagram and Twitter, there are tools available and manual processes to get a good grasp of credibility.

1. Engagement rates - This is a good indication of how many of their total followers are actually engaging with their content, and exploring whether their reach is valuable at all. Comments should represent about 2 percent of total engagement on Facebook, Instagram and Twitter, according to Digiday.
2. Comments - If you find unrelated comments asking for likes, subscribers, or generally sound spammy, they are likely to come from automated bot accounts
3. Follower growth - If an influencer has an irregular or sudden spike in follower growth, it could mean that they have been purchased. Sites like Socialblade can help monitor this.
4. General credibility tools - There are a host of tools which do some of the above for you, by using AI to establish audience authenticity scores based on a combination of follower growth, engagement rates and followers who will actually see the content on an influencer’s feed. Hypeauditor, Influencerdb and Twitteraudit are examples of this.

These processes are imperative in moving away from impressive but ultimately meaningless metrics that so often entice brands, and naïve marketers.

It’s up to brands and agencies to invest in understanding these pitfalls and how to avoid them, if we are to work toward genuine ROI.