So your boss just asked you to “go find some influencers” for a campaign coming up. Seems easy at first, but you find it’s quickly overwhelming – there are tons – but you slug through a few 60-80 hour weeks. You can do it! (Just not all the time).
But what if you screw up? Used to be there was so little visibility into influencer campaigns like engagement, comments, etc. But now the CMO is checking out the posts and asking for those pesky metrics – you know the ones – KPIs, ROIs, conversions, share of voice…So the stakes are much higher. You will be accountable for low engagement, negative comments, and content that makes your brand look cheap. But there are a few simple “tells” that can reduce your risk, and all without buying any platform or services.
Influencer marketing is officially here to grow. Know how we know? Influencers are gaming the system, buying followers to pump up their reach number. And this is your risk; picking fraudulent influencers. Now, although there have been examples of fake profiles signing up on programmatic influencer platforms, you really just need to worry about the ones buying followers, because it’s becoming a problem.
Can you do anything? Sure you can. Maybe not a long term answer, but let’s worry about that later.
How the Platforms Find Fake Audiences
It is complicated, I won’t kid you. You need to detect sudden jumps in following, look at all the countries they’re coming from, the number of times they post, the age of the account, and more. And, you need to do it throughout the year and definitely in the discovery phase for everyone in your pool. We run these algorithms once a week, then double-check before we recommend an influencer to a client.
Should you trust the social platforms like Facebook, Instagram, and YouTube? In a word, no. Yes, they’re good folks and yes, they don’t want to be known for housing fraudulent profiles or those with fraudulent followings. But – they’re only in business to sell advertising and today, social advertising needs reach. So although they are dedicating considerable resources to sniff out these bad actors, it absolutely is not their top priority. And, priorities matter across the business spectrum. Apple is brilliant at device hardware and operating systems, right? But are you listening to Spotify or iTunes now? You can rely on Apple’s hardware like you can rely on Facebook keeping compelling content in front of you. Just don’t depend on Facebook detecting the bad profiles on their platform – that’s up to you.
Primer for Calculating Engagement Percentage (w/o a platform)
One core skill you’ll need to learn is how to quickly calculate engagement percentage. On Instagram, use your web browser and roll over their posts with your cursor – start with posts at least a few days old – and either eyeball it or calculate the average Likes per post. I’d suggest you use over 10 posts, but in case you’re math-challenged like me, the calculation is stepped out below.
Just count the Likes, not including the comments which can include a lot from the poster and repeats from the audience. And, we only calculate engagement using video views if we’re doing a video (naturally). Video engagement can skew a lot of automation engines, so that’s something to look out for, too.
Don’t laugh – I’m kind of math-challenged, so I like to show my work. You can see the simple formula in the example below and again, make sure to browse through about 10 posts (old ones may not show current-enough engagement) – this example only uses three.
3 Dead Giveaways of a Fake Audience
Now to detecting those influencers who can jeopardize your job. Again, these are things you can do with just your laptop – no platform needed.
1. Low Engagement Percentage + Lots of Followers
Fake audiences are only aligned with your brand by luck. And typically, most of the fraudulent followers do not engage – it’s just to pump up the reach. So when you see someone’s engagement is under 1%, you should be wary. In fact, even if their audience is credible, you can almost surely be able to net better absolute engagement from a smaller influencer. Read more here.
2. Sudden Jump/Dive in Engagement
Before you look at that engagement percentage, you can roll over as many posts as you have time – but at least a couple page’s worth. If you see a sudden drop off in engagement (starting from the top, most recent posts), then you should definitely be suspicious. Don’t worry about it being subtle; you’ll notice. Someone getting 200 – 500 likes on the top 3 rows then Likes will dive to 50 to 75, for example. This influencer has spent more on their fraud – buying followers who can programmatically like or comment on their posts. And it’s as close to a sure giveaway as you’ll find.
3. Horrible Content + A Lot of Followers = Badness
You’ve probably seen these with the #ad tag. Out of focus, poorly composed, thoughtless caption, bad engagement, but…tons of followers. These usually pop from programmatic campaigns, where “just enter a budget and we’ll do the rest!” Brands rarely, if ever, see the content because that budget is spread out over 1000s of influencers at $.25 – $1.50 per post. But hey – you get what you pay for, right?
Here’s an example where $0.25 would have been generous:
Bad content is worse than a money-loser, it can damage your brand. Stay clear of these influencers.
So you see that you don’t have to spend a ton of money on a platform or service to vet a few influencers. Services (like us!) do come in handy when you either don’t have the time or you’re fully bought into the influencer channel and do it a lot. In other words, when you’re ready to scale your campaigns.
MATTR is the only full-service influencer marketing provider with detailed audience insights from PersonaMesh™. We go beyond demographics into psychographics such as values and interests so that your influencer campaigns align with your campaign targets.