Facebook post outlier checker

Each post gets one number. A reaction counts once and a comment counts four times, because writing a reply takes more effort than tapping a button. A share counts six times. Passing a post along to friends is about the strongest signal a reader can give. SwipeKeeper uses the same weights inside the app.

Each post is then compared with the median of the other posts you typed in. The median is the middle value, so one huge post can't drag it up the way it would drag up an average. A post at two and a half times the median or more is marked as an outlier. Anything from about one and a half times up to that line is above usual, and a post under 0.7 times is marked quiet.

Three hundred reactions is a flop for a page that usually gets three thousand. For a page that usually gets forty, it's the best week of the year. Measuring a post against its own page's ordinary day is a much fairer way to tell what worked for that audience.

Use at least five posts, and ten or more if you can. Pick them from the same few months. A page that grew a lot last year will make its older posts look quiet.

Add the first line of each post if you want to see what kind of opener it used. The checker tags each one, for example as a question or a first-person scene, so you can see whether the outliers share an opening.

It only knows the numbers you give it. It can't tell how many people saw a post or whether money was spent to boost it. A boosted post can look like an outlier when it simply had a budget behind it, so leave those out if you know which ones they are.

The counts under a public post are fine to note down for research. Copying someone else's post word for word is not. Use the outliers to learn the shape of what worked, then write your own version in your own voice.