Based on the evidence, it appears likely that @miss.everywhere purchased likes and comments for many of her paid posts, if not all. Just as many other Influencers are doing to keep sponsors fooled until the money is in the bank. For our test Instagram account, we bought 10,000 likes for 16 photos for a mere $50, and all the likes were delivered to our account within 10 minutes. If fake Influencers can make $10,000 (or sometimes a great deal more) for an Instagram post, spending $50 to guarantee you’ll get the response your client is expecting sounds like pretty good insurance. Especially if these seemingly popular posts and videos get shared outside of Instagram to YouTube, Facebook, and others.
Many paid posts also come with videos, like this one featuring a Lenovo Yoga Book. It is interesting to note it has only 520 views. Many Influencers typically have high volumes of followers and views across many social networks. Of course, all of that can be bought too. It takes a lot of effort to keep all those faux plates spinning.
I know I’m really throwing @miss.everywhere under the bus here, but maybe she can get a paid post from a bus company before it hits.
Depending on the popularity of this book, @miss.everywhere may end up getting heaps of actual followers. So, I suppose, there’s that. Maybe that will get me back on her Christmas Card list, but I doubt it.
Case Study: @amazingthailand
Now that you know how to validate if an Influencer is legitimate, let’s use what we’ve learned to take a look at another account. I picked this one because I took a closer look at them several months ago in regards to a project I did in Thailand, which I talk more about in Appendix A (All about Trey). To summarize: I do fun “photo walk” events around the world to meet fans and we wander around to take photos together. A lot of times, after the walk, I also give a little art talk and do Q&As. On occasion, I invite other social media Influencers out with me to create a bit of a multiplier effect. Usually, I’ll hire these people or they’ll be brought on board by one of our sponsors.
For example, when I was in Vienna, Austria, I had a photo walk and invited the painter Meagan Morrison (@travelwritedraw on Instagram). She was great! Also, she’s totally legit. We had a fun time leading the photo walk together, and, later, when we went back to the ballroom at the Ritz-Carlton (our sponsor), I gave a 30-minute art talk, while she did a fabulous painting. It was a lot of fun, so I decided to try to do more of this in the future.
When I was planning a visit to Thailand, an Influencer agency started recommending possible collaboration partners. You can see one of the slides they sent me.
This is one of many suggestions sent to us by a supposedly trustworthy looking social media agency.
Most people don’t know how to do the legwork to verify that Influencers are legitimate yet. I do. I’m actually incredibly busy working on the art business and I don’t have time to do other people’s work for them, but I often end up doing precisely that. Maybe you know what I mean.
Anyway, I looked up this @amazingthailand account and I eyeballed it. Nice photos, for sure. It was clearly one of those “aggregation” accounts that reuses amazing images from other people. There’s nothing really wrong with that, as long as they own rights to the photo, or if those photos are part of the Creative Commons, and they give credit. That all seemed okay.
But then, when I looked into their data with Socialblade.com, I saw the incredibly obvious follow/unfollow pattern. Take a look at the follower stats for @amazingthailand yourself below. As you may remember, I am making some bold assumptions from the data. You, the reader, can decide for yourself whether you agree with my speculation.
It appears that @amazingthailand has employed an automated script (or bot) to automatically follow the limit of 7,500 people and then unfollow them over time. Source: Socialblade.com
Data-heads can see a pattern immediately. For the rest of us, I’ll break it down. Look at the bottom chart first. You can see the Following count, the number of accounts @amazingthailand is following, shoot up to 7,500 (the cap) very quickly, then drop back down to zero again. Although you can see a few different shapes in the graph, all more or less follow this pattern: following, then unfollowing, as many people as possible within a short time. The sharp fluctuations and regular geometric patterns also indicate that this is an automated process.
Now if you look at the upper chart, you can see the slow accretion of followers. Obviously, this method works very well and that is why so many fake Influencer accounts employ it. It’s important to remember that their followers are completely random people and there is unlikely any significant demographic overlap that would be interesting to a brand.
The Influencer agency was pretty excited to offer me the husband and wife team that runs the account of @amazingthailand as a fantastic package deal. If I hadn’t run the data myself, I might have hired them. From the data, it appears evident to me that this couple was open to employing unscrupulous methods for gaining followers, so I didn’t want anything to do with them. Ethics aside, their followers are also completely worthless to me. It would do nothing for me professionally and even less for the brands I work with, because the follow/unfollow tactic ends up collecting completely random people and bots, rather than building an engaged audience.
I’ve received countless collaboration pitches from agencies for Influencers who turn out to be fraudulent. This was just one example.
And, of course, this follow/unfollow method of gaining followers is explicitly against Instagram’s Terms of Service. You’ll see later in the book that, even though I presented this to Instagram security, they did nothing.
How to Vet Influencers if You’re a Brand, Marketer, Agency, or Business
If you were in charge of finding Influencers for a big brand, would you have taken any of the steps we just discussed before engaging in a contract with either of those Influencers? Let’s say you had $75,000 to spend on flying one of these Influencers to an exotic location and paying her to share a few posts on Instagram of her experience. Would you have chosen @theblondeabroad or @miss.everywhere? This is the dilemma faced by thousands of marketing people every day. Brands have to make decisions about how to spend hundreds of millions of dollars to promote their brands and they’re doing it with fairly limited information.
Major brands across every industry—especially fashion, luxury travel, cars, airlines, restaurants, and entertainment—are parting with hundreds of millions of dollars in the same way. They are paying fake Instagram “Influencers” to promote their brand. These fake Influencers are deliberately pretending to be more influential than they are in order to mislead and deceive these companies.
In this section, we talk about how to avoid making this mistake, as a brand, marketing agency, or business.
Whenever you engage with a potential new Influencer, it’s essential to do a deep dive in the “investigation” phase. This will not be something you can take on in about 15 minutes. Plan on spending 4-8 hours really digging and getting into the weeds.
Since none of the techniques I review in the previous section are conclusive on their own, they should be taken together to help create an informed opinion. I recommend the following steps.
Step 1: Look up Their Historical Follower Growth on Socialblade.com Or another Service
Remember that there are two main ways people buy followers:
Buy huge swaths of followers. This is a dumb approach. You’ll easily be able to detect this fraud by the big numbers that spike up in irregular patterns.
Buy followers that are delivered in a steady stream over months and year. This is a smarter approach because these follower acquisition patterns are more difficult to detect.
When you look at historical follower growth, look for the above patterns to detect devious behavior. For more on this, check out the previous section, where I break it down in detail.
Step 2: Ask Your Influencer Lots of Questions
Another way of vetting your Influencer is by asking them some pointed qu
estions. The answers to these questions will not necessarily indicate whether or not your Influencer is 100% authentic, but you’ll get a better sense of their authenticity this way. At the very least, your Influencer will know you keep a close eye on your marketing spend and that you’re a savvy buyer.
Below, I’ve listed a few questions you might think about asking. You should adapt these questions for the particular social network(s) you’ll be hiring your Influencer to use. For instance, it’s appropriate for brands or agencies to specifically request YouTube viewing statistics for all the videos concerned. My friend Jared Polin, who has a huge number of subscribers on YouTube himself, told me over email that he’s seen all sorts of fishy behavior, especially when it comes to money deals. According to Jared, a lot of less scrupulous YouTubers will buy a ton of fake views to make it seem like the brand was getting a good deal.
Here’s a list of questions you might ask your future partner Influencer:
Have you ever purchased a follower, a like, a comment, a video play, or a story play?
Do you use any third party apps or software to automate your responses to comments on a post?
(If the user has a “verified” blue tick) Did you earn or purchase your blue tick from a third party?
Have you ever participated in a “pod” or engagement/follower exchange before?
Please send screenshots of all the Insights for your account that show audience breakdown, profile visits, countries, reach, and impressions. (I have attached an example from my personal account.)
Please share your media kit that illustrates your reach across the social media platforms where you engage. (For an example of this, you can look at my media kit at www.StuckinCustoms.com/media-kit)
Do you have references from other brands and agencies you’ve worked with?
These are some of the stats that all agencies or brands should request from Influencers. These statistics are available in the Instagram app. Now, many of these numbers can be affected by bots, so they can’t be trusted entirely, but this information should be included as part of the overall “investigation” phase. Note: Instagram has recently announced that even more insights and stats will become available to Top Influencers soon. Source: My Insight Screens from @treyratcliff
Step 3: Deep Dive into a Selection of Posts to Read All Comments and Likes
I’ll warn you now that this will be the most time-consuming step. I recommend opening the Instagram website on a computer or laptop because it allows you to open several tabs at once for comparison.
First, look at all your target Influencer’s photos over the last few years. If they appear to have built a substantial following in months, rather than years, that’s a red flag. Assuming the account is more than a year old, do you see steady growth in likes and comments? Of course, it’s natural for accounts to start small then gain more traction over time.
Look to see if some posts have way more comments and likes than others. Why? When Influencers mention a brand, it’s usually a paid post. That’s usually when they’ll buy extra comments and likes to impress and achieve their KPIs (Key Performance Indicators). Sometimes a post garners more comments for innocent reasons such as they are running a contest, have a particularly awesome photo, or there is some other content inside that is more compelling than usual. Some Influencers will blame spikes in engagement on spam and bots, which might be a contributing factor, so what you’re looking for are patterns.
Now, hopefully, this Influencer has made sponsored posts for other brands. You’ll want to pay particular attention to these by reading all the comments and responses to comments. If you notice most responses are vague and/or a string of emojis, the Influencer is likely paying a service to auto-respond to comments. Auto-responding to comments is a bit “greyer” on the ethics scale. Many people simply believe it is good manners to reply to people personally. But that takes a lot of time on your phone, especially if you’re getting a ton of comments. I can see people paying a bot to respond, but I don’t think it’s because of good manners; I believe most people set up an auto-response bot to deceive brands into thinking there are more comments than there really are.
Watch for reply comments that come in batches like this that appear to be automated to rack up the comment count. Note if you use Instagram on the web, you see comments in more of a chronological order and you can see when the bot scripts check in to dump auto-replies. Note that sometimes the replies are not simple emojis like this. They can say things like “Thanks dear” or empty things of this nature.
Look for common bot activity. Note the unusual comments, and note who made the comment. Also, note if other users are making the exact same comment.
Suspicious comments sometimes stick out like sore thumbs. Remember this sponsored post from Dec 14th? I’ll never forget two of my literary idols, @omyma_mimo and @habiba20bc0 or their, classic catch-phrase: “Snow fairy ”. Oh, stop it you two. Get a room. However, if you scroll further down, you’ll see that same classic line, complete with emojis, repeated by the indomitable @hassanalserafy_.
There’s another set of rather obvious repeats. As you can see above, @havvac508b commented, “I want to join you babe”. That @havvac508b, he’s a real charmer. But so is @shahbaa.aaa, who said the same thing, with the exact same emoticons, only a few seconds earlier. Keep track of these names, as you may see them across many posts.
The metric brands and agencies care most about is the comment count, because it indicates that people are actually taking the time to write a comment rather than just make a lazy like. Since Influencers know very few brand representatives will manually check comments, auto-responses are an easy way to inflate the number of comments. In the final marketing report about engagement on a post, it’s helpful to check that the Influencer’s responses to comments have been subtracted from the total number of comments. If not, the Influencer’s own responses can add up to hundreds and sometimes thousands of extra comments per photo, while they shouldn’t really count as new engagement.
Next, examine what kind of comments the posts are getting. Do all of them sound believable, or are there many short, vague comments that could apply to any photo such as, “What a place,” “Amazing Snap,” and this sort of thing? Huge strings of emojis are a telltale sign of generic comments purchased on the cheap.
Then, do many of the commenters’ names look suspicious? If you are seeing a lot of names like “Create_Travel_” or “Beauty543” or “Fash.ion.Style”, it’s time to start popping open new tabs for each one.
I also suggest diving into the likes for each post and studying a good sample size of the accounts that made those likes. However, this approach is not always practical, since investigating a good sample size of likes for a post with, say, 5,000 likes could take days. However, if you see some fake comments above, chances are that there will be fake likes as well. Anyone that buys fake comments will likely be buying fake likes too.
Step 4: Face-To-Face
Lastly, I always suggest a face-to-face meeting, even if it has to be over Skype. You can tell a lot about a person in this way, especially if you’re good at reading people. To save a bit of time, you can combine this with Step 2 by asking them those challenging questions in person and watching their expression as they respond. When I do this, I treat it like a standard job interview and ask a wide variety of questions
Also, make sure they understand that you will be asking them to turn over 100% of their data and stats for all photos associated with the campaign. It’s perfectly okay for you to do this, as any first engagement should be thought of as a trial.
Bonus Step: Build a Fake Account of Your Own
Have someone at the agency or brand create their own fake account. It can be based on anything and it doesn’t need to be a person. For example, make an account all about cookies and just add photos of cookies.
Why do this? Because you’ll learn, first-hand, how it is done. Try several different services and you’ll begin
to think like the scammers. You’ll become even more observant. Also, you’ll be able to stay on top of the latest trends and clever ways to circumnavigate Instagram’s rules and trick brands.
How to Spot Fake Accounts If You’re Instagram—And What You Should Do About It
As you can see below, eight months after we bought 104,000 followers for our fake account, @genttravel still has over 84,000 of them. That’s quite a bit.
Here’s our favorite assistant, once again. Oh, if you want to follow his real account, where he unfortunately doesn’t post as many selfies of his chiseled visage, go follow @tanegent. He’s turned into a good photographer, no doubt through my excellent and selfless guidance.
Why hasn’t Instagram stepped in to shut down this account or remove its fake followers? Here are a few reasons why Instagram and Facebook may not be tackling this problem head-on:
They like the big engagement numbers as much as the brands do, because it makes Instagram look more successful. Wall Street likes that, too. All the top people at Instagram and Facebook have a lot of stock options, so bigger numbers translate to millions or billions of dollars in compensation for them. So, why would Instagram want to seriously address the fraud problem, and cull bot accounts, if doing so is not in their financial best interest?
Under the Influence- How to Fake Your Way Into Getting Rich on Instagram Page 11