Cross posted from

I was a serial Facebook addict. I used to spend 2+ hours a day on Facebook, most days. Until I worked out how to change my mind.

Let's talk about the news feed. We all have this feeling that the news feed is drivel. Even curated, mine was still full of crud. Even super curated it was dull at best. Eventually I realised, something had to give.

As with many conflicts, indecision feels uncomfortable. Personally, I was super uncomfortable sitting in the cognitive dissonance of two conflicting beliefs:

Belief 1: Facebook is drivel and I want to spend less time on Facebook.
Belief 2: Facebook has good content from my friends that I want to keep up with.

There are three possible ways this can go. Either Facebook is in fact drivel and I will be happy to avoid it at any cost. Or Facebook has good in it and I'm staying around for the good stuff because I know it's worth it. Or Facebook is sometimes bad and sometimes good in some other complicated fashion, and I should check Facebook in some complicated intermittent fashion because of that...

This is how I worked out which belief was right.

You will need:

  • your news feed
  • pen and paper
  • 5-10 minutes
Basic premise: Facebook has some good content and some bad content. But how much of each is ideal, acceptable or tolerable?

  • If there were 10 good posts for every 3 bad posts, I might be willing to accept that. Maybe I can take some rubbish with the good! I should visit more often. 10:3
  • If there was 1 good post to every bad post, I could still accept that. 1:1
  • If there was 1 good post for every 5 bad posts, maybe I could suffer that. After all, not everything is perfect. 1:5
  • But what if there was only 1 good post for every 10, 20, 30 bad posts? I don't think I'd be okay with that. 1:10
  • And if it was worse - 1 good for every 50 bad - that would be enough to leave the platform. There's no point digging for diamonds in a dung heap. I'd rather just read a book. 1:50
Think about the possible ratios and write down your pre-commitment. What ratios do you consider ideal, acceptable, and unacceptable? Don't worry about getting it perfect; you're an adult who can change their mind at any time. The purpose of writing down a ratio is to establish a baseline expectation before testing. I'm not coming to sneak up behind you and see your piece of paper and judge you if you change your mind. It’s fun to try set a belief and then test it. This is doing science.

Once you've decided your ideal, acceptable and unacceptable ratios, write down another number. What do you think the actual ratio is? Try to be honest. No one is looking.

Next, draw this table on your page:

Now comes the part that takes the time. I want you to go down your news feed and I want you to count if you think posts are good or bad. There's also the "maybe" list for if you can't decide. Be honest with yourself. Try not to count posts a particular way. Try not to push the result somewhere. As a scientist, be curious about what it's going to be. No need to bias the results. It won't work as an exercise if you lie to yourself. It's also unnecessary to lie.

What's good and what's bad? I don't know. It’s subjective anyway. I can't tell you how to do that. You might want to think about:

  • Do you want to hear this?
  • Do you like the person it came from, is that enough to make the post good?
  • Is it user generated content or is it shared content from elsewhere?
  • is it "funny"?
  • is it "news"?
  • is it "happy"?
  • is it "political"?
  • Think of your own version of what factors might matter.
  • Is this an ad?

And keep going. You can go to 100 posts, you can go for 5 minutes, go till you get bored. Up to you how you decide when to stop. With a warning: If you don't pre-determine the stopping rule you can bias the numbers a little. What if post 101 is a good one, you might stop at 101 not 100. That means you swayed everything a little more good on the ratios than you would otherwise have measured.

Then what? Count the columns. Then? Do nothing. It's just a ratio. We measured, but we don't have to do anything different.

How does this work? What happened? What did you do to me?

I say do nothing. But I am confident that you are going to allocate less time to Facebook than previously. Just naturally end up on there less often.

The trouble with feelings is that they are based in System 1. I have a hunch that Facebook is boring. But I don't always naturally know what to do with that hunch. The task above takes the feeling and brings that into System 2. We can count and measure the exact quantity of the feeling. Then maybe we can be better informed to act. And act you can.

Depending on what you uncover, you can choose what to do next. You now know exactly how good or bad your news feed was today. Take this information and choose to look at Facebook less, or choose to look at Facebook more. Maybe it's where you find all the good ideas! And any time you like - retest. See how your intuition for the site, mismatches or matches to reality.

Another interesting thing I noticed on Facebook:

If I hit the "like" button, Facebook tries it's valiant and hardest to show me more of that same thing. That means if it's pictures, Facebook will deliver slightly more pictures. If it's from a group, posts from that group. If it's from a friend, I'd be hearing more from that friend over the next week. I kind of stopped hitting the like buttons. Facebook doesn't need to know my likes. I also hate like notifications so much that I got a browser add-on that hides them from view and notification.

Now I wasn't hitting the like button so much. But I would still comment occasionally. I would watch flame wars as they happened. And I started getting interested in arguments. I tried to work out where they started. It seemed that they always started earlier than I expected. Well before a flame war, people are getting aggravated. But while that was interesting to learn, I wasn't commenting - I didn't want to interrupt the arguments that I was seeing, I was trying to be an objective observer. When I would make the occasional comment, same as likes, Facebook would deliver content that was similar to my future browsing.

if you are curious how the news feed will change, you can follow me in stopping likes and commenting. Try it. Also try other experiments. Science is fun!

Meta: I did this 6+ months ago and the results stuck. I spend a lot less time on Facebook than I used to. But the amazing thing is that the effect was almost overnight. It was obvious. My ratio was worse than 1:20. That was unacceptable to me.

When you try this, make sure you put down some predictions. Part of the scientific method is to make predictions and then test those predictions. In the process of modelling the world, predicting the future and generally being awesome, it’s okay to be wrong, it’s okay to be right - that’s why we test. It’s not okay to fake the results and lie to yourself. Wouldn’t it be something if you were surprised. Or confused. Or you changed your mind in response to the evidence you found.

Facebook and social media are becoming an entrenched part of our lives. Hopefully you ask the question about who's in control. Facebook is out to get you after all.

New Comment
11 comments, sorted by Click to highlight new comments since: Today at 7:27 AM

So, first of all, most people aren't unfollowing anywhere near as many people as they could be on Facebook, so if anyone hasn't tried just unfollowing anyone who posts a single thing that isn't worth your time I would highly recommend it.

But if you'd rather whitelist than blacklist, and also you're annoyed at how inscrutable the newsfeed algorithm is, consider the following alternative from Robby Bensinger: create a custom friends list of all the people whose Facebook posts you actually want to see, and look at that friends list instead of your newsfeed. This will display all of your friends-who-are-on-the-list's posts, in chronological order. You get way more control over what you see this way.

I don't use Facebook, but I should try something similar with my RSS feeds.

I find it interesting that facebook responds to commenting by showing you more of the same. IIRC, posts that aggravate people are also the most likely to inspire them to comment. That suggests Facebook is effectively rigged to piss you off.

Does that match people's experience? It matches my priors, but they're weak priors, since I won't touch the service with a ten foot pole.

That suggests Facebook is effectively rigged to piss you off.

Facebook is rigged to keep you checking Facebook as often as possible, regardless of the emotions that may give you.

I actually organize my RSS feed by signal:noise ratio. So I do basically this process with each feed, or perhaps just rough-estimate it from the first 10 items or so. Once I have the rough signal:noise I put it in the appropriate folder. I found that this pretty much corrected the "giant wall of crap" problem immediately, so that it was easy to find the things which were important and leave the high volume slim pickings feeds for last. (Or better yet, in many cases not at all.)

Facebook is effectively rigged to piss you off.

Facebook is agnostic towards content. So yes. Sometimes it can be an outrage machine, sometimes it can be a happy-joy machine. I suspect the extra Emojis help to classify and filter posts too. If you angry emoji something you can react in a way that maybe it will show you less of that. But I am unsure if that's actually what happens.

If you actually decide to use Facebook I think it's key to use your powers to shape it to be the way you want it to be.

1) Unfollow people who write posts that you don't find valuable to read.

2) Use plugins to remove elements from the side you don't find valuable to see. I use Facebook usability suite and remove further content with UBlock.

3) I find the Lite app to be more to my preferences than the regular Facebook app.

My good:bad ratio was 1:2.75, which is a reason to use Facebook less for me (1:3 was "I'd rather not put up with that", 1:1 was "would be okay"). This is after unfollowing lots of people. (In this exercise I also found two more people to unfollow.) I've hidden my newsfeed using F.B. Purity for now.

Great! Well done. Now you have a tool to use any time you want to check in on those mixed feelings and compare qualities of the inputs you are getting.

1. I think this technique can, and probably should be applied to lots of other stuff.

2. I wonder how much of the benefit comes from just writing down the tallies, versus from looking back at the tallies and using that data to think "I enjoy Facebook more/less than I thought".

3. From what I understand, in the field of positive psychology, they have the subjects do stuff like this, rather than asking them to think back and answer questions like "How happy are you when you're using Facebook?". From that, I assume that if positive psychologists value this technique, there must be something different and useful about in the moment data versus looking back and thinking about your past feelings data.

  1. I think this technique can, and probably should be applied to lots of other stuff.

Yes definitely. Any kind of feed or regular sequence of events. For example details of your work day. A birthday party event. running an event.

.2. I wonder how much of the benefit comes from just writing down the tallies, versus from looking back at the tallies and using that data to think “I enjoy Facebook more/​less than I thought”.

yes. There is a lot of benefit to being able to be clear on what was a cloudy feeling before.

.3. in the field of positive psychology, they have the subjects do stuff like this, rather than asking them to think back and answer questions

subjective ongoing experience is often recorded differently to retrospectives. Dan Arieli has a lot of work on how to manipulate people into remembering slightly more positive experience.

Like someone else mentioned, this technique could work for other things. I don't use Facebook. I used a slightly modified version of this technique on my Youtube watch history and came to a ratio of about 1:20 (bad videos do tend to be shorter, so time wise it's probably 1:10 or so).

I've been trying to reduce my usage for a while, with moderate success. According to RescueTime I spend 32 hours on in the past 30 days, and 49 hours in the 30 days before that. I'll keep in mind the poor good:bad ratio from now on and I'll report back in another 30 days with the new figure from RescueTime.

If it's under 25 hours I consider it a moderate success, under 20 a huge success. 



Results are in, just 15 hours for the past 30 days. It seems to have worked quite well for me.