A recent New York Times article surfaced a practice that has concerned us since we started CLEVER almost nine years ago: influencers paying for social media followers.

This is one of the most discussed topics when we meet with clients. My co-founders and I have been in this industry for 15 years now–we helped to build it as we know it–and in terms of bought (or bot!) followers, we have seen it all. This is why, since the day we started, we have had a combination of humans and tech (not one or the other) vetting every single influencer who applies to our network.

Gen4 DASH is our proprietary network and program management platform, and we rely on it to scale as necessary. It allows us to manage thousands of influencers with ease. We also use tools (more on this below) to assist with data and analytics from influencer vetting to program conclusion, but our technology is effective because we don’t rely solely on an algorithm to make vetting decisions.

The first step in ensuring our influencers’ followers are who they say they are is to be extremely selective about who we allow into our network. We accept about 5% of influencers who apply. We’re not an open network where anyone with a URL and email address can automatically join our database. Our goal has never been to be the biggest influencer network. Since day one we’ve focused on being the absolute best. We are.

How we vet is also important. We vet for many criteria including quality of content, how long an influencer has been creating content, percentage of sponsored posts vs. unsponsored posts, and more. Since day one, we have vetted for authentic engagement to ensure comments are real, shares are real, and that the engagement-to-follower ratio makes sense. No other agency does this better.

It’s the job of our membership team to know our influencers inside and out; to interact with our network members, to ensure they are real people who live where they say they do, and who create all of the content posted to their social channels. Whenever my co-founders and I are out in the field I make it a point to host a meet-up and invite local influencers from our network to attend so I can meet them in person, and so that our influencers can also see there are caring humans looking out for their interests as well.

We also rely on technology to bolster our vetting process. We currently use Botometer and DeBot. The first uses audience profile metrics to predict the likelihood of followers being bots, the second analyzes posting activity of followers to identify correlations to known bots.

We are also currently in the process of implementing the DARPA-recommended approach which we believe is more effective than tools alone. One of the key takeaways for CLEVER when reviewing this approach was this finding:

The winning strategies are revealing. The teams began by attempting to identify an initial set of bots in the data. Interestingly, none of the teams were able to automate this step and most used significant human input.

In other words, our expert, “human review” will continue to play a key role in detecting any issues with regard to bot/bought followers. Even the best tools cannot replace a human’s expert knowledge and critical thinking skills.

Because we are so discriminating in all aspects of our business–not just vetting influencers for fake followers–we have a reputation for quality and stellar client care that precedes us. We work very hard at it, and I am very proud of this. No other agency vets as carefully, matches influencers to your needs as precisely, and scales a program as professionally as we do.