I love clicks per follower (CPF) as a metric. I’ve advocated for it in the past, but after spending last week doing a bunch of research for a post I wrote for Social Media Explorer, I feel compelled to re-argue my case.

A brief refresher on CPF: If your tweet had 200 clicks and you have 7,000 followers, your CPF is 200/7,000, or 2.86%. It is the closest you can get in social media marketing to a click-through rate.

There are four core questions that CPF helps you answer:

1. Is my performance increasing or decreasing over time? Hopefully your audience will grow, and it’s important that your idea of “good performance” scales with your audience size. CPF allows you to measure this.

2. Is engagement better on @twitter_account_1 or @twitter_account_2? CPF allows you to compare the performance of different accounts that you manage. 

3. What channel performs best? CPF allows you to compare engagement across Facebook, Twitter, G+, etc.

4. How does your engagement stack up to your peers? If you have a peer group benchmark to measure yourself against, CPF allows you to compare yourself externally.

If you measure don’t measure CPF, you can’t answer the above four questions.

Sold? Good.

Fortunately, CPF is dead simple to measure using data you should already have. All you need is a table of posts with post date, clicks, property name, and follower count. Take a look at the below Google Doc example. Feel free to copy straight from this as you set up your own.

Find the full spreadsheet here.

Once you have your table with all of the posts and associated info, just take some basic averages. Average the CPF all of the posts within each property and compare them. Which does best? Average your CPF on a month-by-month basis. What’s the trend? Average your CPF by channel. Any outliers?

All of these calculations have been done in the embedded spreadsheet, so, again, don’t worry–just copy it!

The last question—measuring yourself against your peers—is the topic I tackle in detail in my post on Social Media Explorer. There’s a lot of good performance research in that post, so I’ll let you read it for yourself.