The follower count is the wrong number

Every brand that finds Exposely asks the same question within two emails. How many followers can we get in front of, and what does that cost.

I understand why. It's the only number on the screen. It's printed right there on the profile, it's an integer, and integers feel like they mean something. If you're a marketing director who has to justify a spend to someone above you, "we reached 400,000 people" is a sentence you can say in a meeting without getting interrupted.

It's also close to useless.

What the number actually measures#

A follower count is a record of every decision anyone ever made to click a button, minus the much smaller number of people who later bothered to undo it. It's cumulative and it's nearly irreversible. It tells you about a person's entire history. It tells you nothing about this week.

I came up on MySpace. I know exactly how this works, because I did it. You can build a very large number by being early, by being aggressive, by being adjacent to somebody who was already big, or by being interesting for six months in 2009 and then coasting. Four different people with the same number on their profile can have four completely different businesses attached to them, and only one of them can sell you anything.

The number that matters is much less flattering and much harder to get: when this person says something, how many people do something about it.

What we look at instead#

None of this is sophisticated. It's just less convenient than reading one integer.

Comments per post, relative to the account's own size. Not the raw count. A 20,000-follower account pulling forty real comments is a different asset than a 400,000-follower account pulling forty. The ratio moves in the opposite direction from the number brands ask about, which is exactly why nobody wants to hear it.

Whether the comments are from people or from the follow-for-follow economy. You can read this in about thirty seconds and no tool needs to be involved. Either there are humans in there having a reaction or there are strings of emoji from accounts that leave the same string on everything.

Whether they've sold anything before. Not "worked with brands." Sold. There's a specific tone that a person who has actually moved product uses, and there's a different tone from someone who has posted a discount code as a favor. The second one is much more common and it's the thing most of a brand's budget ends up buying.

Whether the audience is theirs or the platform's. Some people have followers. Some people have an audience that would follow them somewhere else. Those look identical on a profile page and they are not the same thing at all. I don't have a clean way to measure this yet and I think about it constantly.

Why the wrong number wins anyway#

Here's the part I don't have a good answer to. The follower count keeps winning even after you explain all of this, and it keeps winning for a reason that isn't stupidity.

It's legible. It's comparable across people. It exists before the campaign runs. Everything I listed above requires judgment, takes real time per person, and can't be put in a spreadsheet column that a client will accept as a justification. A brand can be talked out of the follower count in a conversation and then go right back to it when they have to defend the buy internally.

So the actual problem isn't that people don't know follower counts are noisy. Plenty of them do. The problem is that nobody has built a number that's both honest and easy to hand to your boss.

I think whoever builds that number owns this category. I'd like it to be us. Right now we're mostly just refusing to sell the bad one, which is a much smaller business than selling the good one and a much better one than selling the bad one on purpose.