Last week, I spoke at SMXL Milano on a panel about where social media and influencer marketing are headed. The moderator, Paolo Zanzottera, did (in my opinion) a superb job, but the questions he posed and those from the audience, reminded me of the big challenges influencer marketing faces in 2018 and beyond.
The field is growing… fast. Compared to other practices in the web marketing world, the interest levels (as measured by search volume, at least) are skyrocketing.
Search interest in “influencer marketing” outpaces “social media marketing,” “search engine marketing,” “email marketing,” and even the incredibly hyped practice of “content marketing.” Granted, search numbers aren’t everything — new practices tend to get more search volume than established ones (even if there’s far less overall demand) because people are Googling phrases to learn about them, but even so, this looks big.
Mediakix estimated earlier this year that by 2020, brands would spend $5-$10 billion/year on influencer marketing. The Association of National Advertisers (in the US) surveyed 158 marketers (not a huge number, and potentially a biased sample set) and found 75% currently employ the practice, though only 36% feel the efforts are effective. There’s a well-publicized backlash heating up.
So, I figured I’d try to tackle some of the topics from our panel, some of those in the market, and others that I worry aren’t receiving the attention they should.
10 Big Problems with Influencer Marketing
#1: The Practice is Far Too Narrowly Defined
In 2012 influencer marketing meant: “discover all the sources that influence your audience and do marketing (of all kinds) in those places.”
In 2018 influencer marketing has come to mean: “pay half naked people on Instagram $500 to pose with your product”
— Rand Fishkin (@randfish) August 16, 2018
Honestly, this is my biggest problem with the practice of influencer marketing. The tactic should be broad. It should encompass blogs, websites, events, conferences, podcasts, authors, Facebook (pages & groups), Twitter, LinkedIn, Reddit, Pinterest, Medium, discussion forums, and every other source of influence. Instead, the phrase somehow evolved to mean Instagram with a an occasional side of YouTube. That’s… madness.
In any given field, for any given brand, there are numerous channels and mediums that influence potential buyers and amplifiers. Sometimes, Instagram is the biggest one. Often, it’s not. The worst part isn’t just that “influencer marketing” has come to mean this very limited thing, but that the language and focus have taken marketing attention away from every other source of influence. That’s dumb. And some smart marketers are gonna run circles around their competitors by thinking more broadly, and targeting more intelligently.
#2: Follower Counts + Keywords Are a Terrible Way to Select Influencers
How do most brands, agencies, and tools surface potential influencers with whom to work?
- Step One: Search for keywords the account uses in their bio+posts
- Step Two: Sort by number of followers
- Step Three: Filter based on a manual review (and only the more advanced folks do this)
THAT IS A TERRIBLE WAY TO SELECT INFLUENCERS!
In a rational world, marketers wouldn’t seek out the accounts that happen to use the words+phrases they’re searching for, nor those that have the biggest total followers. Both of those are easily game-able. Neither of those indicate that the account even reaches the people you necessarily want to target.
Instead, it should look like this:
- Step One: Define target audience you’re attempting to reach/influence
- Step Two: Sort by the sources of influence most followed by those groups
- Step Three: Filter based on a manual review
Right now, there aren’t good, affordable tools to do this, so some brands do it manually. But I see this as the only logical future. If you’re trying to reach avid golfers to promote your new golf resort in Scotland, you don’t want to find the account that uses “golf resort” and “golf vacation” while having loads of followers, you want to find the account followed by the largest number of golfers who might reasonably visit Scotland.
#3: Measuring ROI is Remarkably Difficult
Brand lift, direct conversions, pre-conversion touchpoints, increased branded search volume, improved brand awareness, wider amplification, links & mentions… All of these are potential results of a good influencer marketing campaign, but none of them are particularly easy to measure.
That’s doubly true on Instagram, where impressions aren’t available and links are only possible via the profile page.
Thus, we end up in a world where tons of investments are made without any confident knowledge of whether a campaign worked or not. For some agencies, marketers, and influencers, that feels like a good thing (at least in the short term), because it means they’re less accountable to hard numbers and more likely to get continued investment. But in the long run, it will eventually lead to brands fleeing from the strategy when budgets tighten.
To compensate, I believe influencer marketing campaigns will need to be measured more like TV, radio, and outdoor advertising. That is, through geographic and time series tests measured with brand lift/decline, direct website visits by geo+time, branded search volume numbers, and press/blog/social mention tracking. That ain’t easy, but it is possible, and it’s the only way to get a real sense for the full impact of an investment.
#4: Likes, Comments, Impressions & Website Visits Are Mediocre Measures of Impact
When data is available for influencer campaigns on certain platforms (Twitter, LinkedIn, Facebook), the numbers don’t map well to impact. A podcast might be able to show total listeners, but not how many heard a particular advertiser mentioned. A Twitter account can show impressions, clicks, likes, retweets, and replies, but can’t tie that to how many people took action afterward. And both of those sources (along with most blogs, websites, and other social sources) are better than the darling of the “influencer” world, Instagram, where measurability is complete crap.
A part of me wonders if the lack of metrics availability and transparency on Instagram is part of what’s made it so popular for influencer marketing. Maybe if companies knew more about the numbers, they’d switch some of that budget and effort to other sources?
But, apart from Instagram (and YouTube), most sources of influence have some accountability to metrics, and I think that’s actually part of the problem. As I’ve written before, stuck-in-the-middle marketing tactics often get the least investment and impact.
#5: Variance of Results is Too High
Campaign A: Instagram “influencer” (I know, I shudder at the word too, but bear with me) #1 puts up four stories and two posts that feature your product.
Campaign B: Instagram “influencer” #2 (with the same number of followers as #1) puts up four stories and two posts featuring your product.
Campaign “A” could be 10X more successful than “B,” half as successful, or 1/10th as successful and no one would be surprised. For those making marketing investments in these types of campaigns, that sucks. Predictable, consistent channels get loads of attention, even if the ROI is small (just think of all the Facebook ad campaigns and Google Ad buys that have barely positive returns but millions in spend). Tactics with high variability often suffer from inconsistent investment, e.g. content marketing, SEO, and PR, all of which are constantly trying to re-prove their worth after setbacks, even when they have higher ROI.
#6: Discovery of Sources of Influence is Manual & Difficult
Say you’re selling restaurant equipment to chefs in Los Angeles or CAD software to architects in Vancouver. How do you figure out which websites those chefs/architects visit? Which blogs they read? What YouTube channels they subscribe to? What podcasts they listen to? Who they follow on Twitter/LinkedIn/Instagram? What SubReddits they read?
Right now, that information comes either from a very expensive (think $50K+) audience survey, often run by a PR firm, or it’s done entirely manually by a marketer who spends days Googling queries like “top architecture blogs Vancouver,” “Recommended podcasts for LA chefs,” etc.
Until this data is readily available and high quality (which, not coincidentally, is the precise problem SparkToro exists to solve), it will continue to be a roadblock for those looking to invest in the process. And that goes double for any investments in the long tail of influence.
#7: Transparency, Authenticity, and Integrity of Influencers
Whenever influencer marketing is mentioned as a practice, it’s inevitable that someone (almost always someone over 30) will mention their belief that, eventually, these “young people” will wake up to the fact that their celebrity and quasi-celebrity influencers are merely paid shills.
My guess is that some people will always be influenced by certain individuals they look up to (yes, even the most notorious of the “Instagram famous” sets), and many of them simply don’t/won’t care that the relationship between the product and the person is a paid advertisement. That said, I agree that some young people who currently don’t care about paid vs. organic will care in the future. And I think some people (of all ages) will never be interested or influence-able via celebrities (or quasi-celebs/micro-influencers/etc). That’s OK. No form of marketing or advertising reaches everyone, but this emerging format will always reach some people. The big question is whether it’s currently under-invested, over-invested, or at the right level given its potential reach.
#8: The Power of the Networks
Today, you could have 50,000 Instragram followers with 5% daily reach. But if Facebook decides to make a tweak and favor different content, that 5% could turn to 1% or 0.1% overnight, nearly invalidating the entirety of the work you’ve done to build up that account’s audience. This is precisely what happened with Facebook’s pages for businesses. It happened on Twitter, too. I’m skeptical that high reach for non-friends-and-family Instagram accounts will maintain their levels of influence for long. When Facebook needs more ad revenue growth or greater product stickiness, this feels like an inevitable move.
#9: Fake Followers, Fake Influencers, and Fake Numbers
We’ve all seen the reports — 50% of paid influencer post engagement is fake, fake followers are ruining influencer marketing, Unilever is refusing to work with influencers who utilize bots and fake followers, etc. — but until the platforms themselves decide to take action and quash fake accounts, bots, and manipulated engagement, it’s unlikely any real change will occur.
When we put out our Fake Followers tool for Twitter in October, the biggest request by far was a similar version for Instagram. Unfortunately, unlike Twitter, Instagram makes it nearly impossible to collect enough data for a third party program to make a reasonable estimate. That’s intentional. In a closed ecosystem, Facebook alone makes the rules and polices them, accountable to no one else.
I suspect this will be a long term issue that third party tools and marketers will struggle against, just as Google and SEOs struggled against webspam and black hat tactics for the first 15 years of search marketing (only in the last few has that subsided, and only because Google finally took serious action).
#10: Over-Abundance of Some Types of Influencers; Almost Nothing Anywhere Else
There’s a very weird phenomenon in the world of influencer marketing where 99% of “influencers” (the Instagram micro-celeb kind) are in travel, fashion, lifestyle and… that’s pretty much it. Who’s the biggest “influencer” in the accounting for small business field? Or in chemical engineering? Or logistics technology?
Yeah… They don’t exist. At least, not in the same way that lifestyle influencers do.
That overexposure in some fields and underexposure in others creates a practice that can’t be replicated and applied across sectors, which detracts from the process marketers have always used with new tactics: experiment, learn, scale, grow, get a new gig, apply learnings, and progress. Until we either broaden what influencer marketing means (see problem #1) or get those micro-celebrity logistics pros posing with the latest in shipping container crane tech on the ‘gram, the tactic has an artificially constrained market size.
I suspect that influencer marketing will continue to gain adoption as a practice, but it will take another few years (maybe more) before the wild west comes to a close. In the meantime, a lot of dollars will be wasted, a handful of companies and campaigns will have incredible ROI, and a lot of folks will be somewhere in the middle. We’ve seen this movie before in SEO, social media, content marketing, and others. Here’s to hoping this time around, the savvy veterans among us will keep both the irrational exuberance and the jaded naysaying to a minimum.
p.s. If you think there’s already solutions to the above, or bigger problems than the ones mentioned here, I’d love to hear about them in the comments.