karl taylor

3 minute read

and that’s “Talking About This.” (PTAT)

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/90632-1oyub4gxswh12lbkzaai6eg.png

If you’ve been working with Facebook longer than, oh, maybe 2011 or so, you probably already know this, and you can go ahead and skip this one.

If you don’t, you’re in for a treat.

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/d78b5-1z0cp1c0vxijzq_tfsyp6qg.jpeg

I don’t know that anyone has taken the time to take a step back and write a little about Graph (Theory) By Zuck. Still, if you’ve spent any amount of time around the Facebook platform, you’re likely familiar with all of the different ways a user may engage with a piece of content.

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/fb871-1rdmttkrmziitgnwogwgc4a.jpeg

You might react (like, or sad or angry a post,), or you might comment. You could share a page with a friend in a message or by tagging them on a post. You could click through a stack of images and download a few. I’m not sure I’ve seen a complete list of the inputs in the current calculation, but AdWeek featured a look back in 2012.

The weighting of each of these activities takes place behind the scenes and is probably more nuanced than one could reasonably expect to cover in a blog article.

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/0dc9b-1pwlcxjojq5sstyofl9fr6q.jpeg

But the total of these interactions is quite insightful.

If you’re running a page with 30,000 likes, but you only have a Talking About This score of 300? You’re in trouble.

If you’re running a page with 100 likes and 250 PTAT? You’re growing at a remarkable clip.

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/3804a-11w_ra1cfalqqmwldzzkjgg.jpeg

To encourage Advertisers to discover how changing each element of the total could affect performance, the fine folks at Facebook have made this number a little less visible. However, you can still access the number for any page with a simple Google search.

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/5924c-1elndirhroslelxgtix5yea.png

Relying on this number isn’t authoritative. I happen to know that for this week, we’ll trend a little off from the screenshot.

I’m not guessing that or feeling it; I know it to be the case.

How?

I took the time to look at each of the metrics and how they influenced the score.

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/17845-1vsujtgoced3yytwlgsmtlg.jpeg

You can do this too.

There isn’t any great insight or single trick or magic widget you need to buy that’ll help you sort out how to manage the winds of which feature is weighted how by the platform. But by paying attention to the metrics that the platform pays attention to when it decides which posts should be prioritized in a feed, is the only way to know what you’re doing.

You might be doing work that you think is cool. You might be doing work that you feel that the audience you read about in a report will enjoy, but one of the best things about buying media on Facebook is that you get to talk to the people directly.

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/e4d73-17lrpbqcgez7tnn78ritprq.jpeg

When they like a post, they’ll like it.

When they have something to say, they’ll say it.

If they see something you’ve read and want to know more about you, they’re going to Google.

You can see these things happen in real-time, and by taking the extra bit of effort to be humble enough to accept that dynamic feedback, you can accomplish remarkable feats.

https://raw.githubusercontent.com/karljtaylor/kjt/blog/content/assets/1a092-14zdze_yu0qy0sp3yrewnhq.jpeg

Running your digital presence this way isn’t always easy because it requires you to check your preconceived notions about what makes something “good,” at the door.

Content made by a kid with an iPhone gets consumed at just about the same clip as the stuff we painstakingly antagonize over. With more and more people looking to networks for “authentic experiences,” you can bet that’s code for content consumed by real people.

The best way to do that is to listen to some.

comments powered by Disqus