Marketing has been about influencing people since the first caveman tried to sell the improved Wheel 2.0. And then, of course, a wheelwright decided to persuade Ogg, the famous hunter and bon vivant, to say how great those wheels were.

Who knows how much a celebrity endorsement cost back then. Today, getting some major name to tweet about your product could run a staggering sum. But even getting a social influencer with a niche following can run between $25 and $75 per thousand followers. So, getting a mention from someone with 100,000 followers could run between $2,500 and $7,500.

There's just a tiny little problem. Those followings are often largely fake. That can happen with big names in music. Last year, MusicBusiness Worldwide commissioned a Twitter audit firm for a rough estimate on how many fake followers some top name pop musicians seemed to have. The accounts ran from 55 percent fake to 67 percent fake. (Go check the list to see the names.)

Followers for the renting

Maybe the count of fakes was a mistake, or possibly bots are mad for top 40. But the idea of people out and out renting many if not all of the accounts that follow them isn't a maybe. According to people I've spoken with who know, the practice is madly widespread because there are a lot of folks who want to be famous but aren't really, and yet they know the endorsement deals, ad money, and even jobs often go to those who at least seem to be popular online.

One example: Lena Katz, a branded content strategist that a colleague introduced to me, finds the entire scene irritating beyond belief and wanted to know if I'd like to follow some proof of how this could be done. She took a picture of a potato, created accounts on the likes of Instagram and Twitter and, within a couple of weeks, had 10,000 followers. It got plenty of likes and even comments on posts -- because Katz was able to write those comments and have them appear from rented followers.

"This is what Potato proved," she told me. "You can look at an account and say, 'For some mysterious reason, 10,000 follow this potato. Within a certain niche, he must have influence.' Those 10,000 followers may be real accounts, but I bought them. They've never actually looked at Potato. None of those people making the comments actually engaged with Potato. I wrote the comments. Marketers and brands don't understand how much of this can be fabricated. Every single thing on the front end in terms of engagement can be fabricated."

Here's a screen shots of a Potato YouTube video of a non-moving spud and an Instagram post:

 inline image
 inline image

And here is a screen shot of a Twitter audit, showing that the rented followers largely appear real:

 inline image

Convincing, eh?

Bands of fakery

The fakery doesn't just happen through purchases. Groups of bloggers will agree to follow one another, comment on each other's work, all in the hopes of helping to create the verisimilitude of an engaged audience. Katz has seen it among groups of mommy bloggers. I've seen it among other writers who trade follows and likes with one another.

Part of the problem is that people get onto social networks, they want to build a following, but to do so organically, if you don't have a name, can take years. So, they look for a jumpstart. Food blogger Genevieve Germaine Dubois, who goes by the name GiGi, told me she did that very early on.

"When I first started YouTube I did but I quickly stopped," she said. "I said, 'This was totally not legitimate.'" And if you want a genuine following of people interested in the work you do, fake followers quickly lose their appeal.

Over time, Dubois did build her brand and following, with a blog that gets between 100,000 and 200,000 views a month, with posts often getting 100 to 200 comments. On YouTube, she has 7,500 followers, because she primarily focuses on the blog, and would rather stick with reality.

Would-be influencers are desperate for work

But you can pay a steep price without that illusion. Landon Ashworth, fiancé of Dubois and a writer, comedian, and actor (as in, recently filmed a guest spot on a primetime network drama) has a YouTube channel. Here's an example:

Ashworth has just under 7,800 subscribers. That's nothing to write home about, and he knows, but he's never bought followers. That can hurt. He went up to the show runner -- the person in charge -- of the show he was on to say thanks for the casting. "It's surprising unless you have a million YouTube followers or Instagram followers, it's hard to get a guest star," Ashworth said. The show runner said, "I'm one of the few in television to fight for the best person for the job." Most in the industry want the social numbers as "proof" of popularity that they hope will tune in to their own production.

"I had 15000 Twitter followers and 25,000 Instagram followers," he said. "I'm a kid from a corn field without his own television show. 15,000, 25,000 seems like a lot. My television agent said, 'You lose out on some parts because you didn't have enough followers.'" The solution? Either buy enough to get to a million or delete the accounts, which he did. YouTube stays because it's a creative outlet and a way to show some of his work.

To buy or not to buy becomes an issue for performers, writers, artists, photographers, and other creatives because big social audiences have become the currency of the commercial realm. Of course advertisers want to see big numbers -- it's the rationalization for spending money to promote through influencers. But every other place that might use some kind of content also wants to see them.

Too many don't know any better

"There is definitely a generation or group of potential influencers so remote from the process of building a brand or followers that they don't know what they're doing wrong," Katz said. "They don't know that if you use an app to get 50,000 followers that it's wrong. They don't understand that if you're in a sewing circle of 15 mommy bloggers who retweet each other's posts, that's not real traffic. They never learned. Nonetheless, the outcome is the same: no real organic uptick in awareness. They don't know how to do that for you."

Sadly, agencies may not ask the questions, nor ask for the proof of data, that would establish whether an influencer has a legitimate following or not. It might be that they feel so pressed for time that they don't think they can go through the process with everyone. Perhaps they want to fill the ordered quota and close their eyes while doing so.

I've heard from a number of agencies a similar story of how companies will sign off on showing video commercials on low-quality sites where a lot of the traffic is fake. Why? Because whoever buys the traffic has a budget, and those numbers let them fill out the totals they're told to obtain without spending too much.

Also, the realization of problems doesn't necessarily spread to the top. I recently spoke to a major ad industry executive about a different topic but also asked her about influencers with fake followers. She was completely taken aback -- seemed to have no idea that the phenomenon existed, let alone that it was widespread.

There is also a problem with some of the tools available. Here's something that Katz received from an analytics tool when she asked if it could determine the percentage of real traffic:

We cannot know if a user is a bot or not. Besides that, even if we had the possibility to know if an account was a bot or not, that would be a waste of resource, as Instagram deletes every month all the bot users / spam posts etc.

Now's the time to remember that Twitter audit from above that showed 87 percent real followers for Potato.

Ultimately, it comes down to the advertisers, or the company looking for content, to dig in and do the necessary research. "I tell brands [that] I wouldn't focus so much on the numbers," Dubois said. "Sometimes they just don't tell the truth." You have to dig in, see how people interact, or not, with the person.

Demand to see the proof

"If you're looking at giving someone a $20,000 trip for free or looking at paying them $20,000, you need to have someone internally do the unsexy work or have an agency that you trust that has done good work for you before," Katz said.

That generally includes asking the "influencer" for screen shots of the relevant system analytics for whatever social network you're considering.

"You need to say may I see X, a screen print of the relevant analytics," Katz said. "When you're testing out new campaigns, you need to arm yourself with that information. Influencers who won't deliver that information or who can't because they don't know how ... you can't do business with them. They're not business people. They either have something to hide or they know so little that they don't treat this as a business and you shouldn't treat them as a potential business partner."