Adding one extra step into your Twitter unfollow process will allow you to see which "inactives" really are inactive.
Did you know that most of the users Twitter unfollow apps report as "inactive" have definitely accessed their Twitter account within the past few days, and probably log in every single day? Well, you do now.
But how did I know that? Or perhaps more intriguingly, why don't the apps know that?
Having checked the Likes pages of everyone on the app's "inactives" list, I was astounded with the results.
I'm going to explain how I found out in the course of this post, but the reason the apps don't know, is that app developers always use the date of the most recent tweet or retweet as an indicator of when an account holder was last on Twitter. The problem is, most people don't tweet or retweet every single time they log into Twitter. Which means the date of the last tweet/RT is near useless as a guide to an account holder's most recent activity.
If you're a publisher on Twitter, your concern is not who's talking, but who's reading. And one of the greatest dangers in unfollowing supposed "inactives" as a publisher, is that you might unfollow someone who is not only still using Twitter, but is actually using Twitter to read your tweets. And is not only using Twitter to read your tweets, but is also actively engaging with those tweets, via the Like button.
With one of my accounts I recently chose to unfollow some inactive profiles. I used the Unfollowerstats app, which presents a list of supposedly inactive users within your Friend list. The idea is that you just hit the Unfollow button for every “inactive” in the list, and it’s job done. But I couldn’t be sure whether any of the people in the app’s list had taken an interest in my tweets, and I absolutely didn’t want to unfollow anyone who had.
This is actually very telling, and would suggest the user does not read their timeline.
Not only would unfollowing an obvious fan of your work be a pretty scummy thing to do - it would also potentially be throwing future Likes in the bin. Someone with an interest in your work is going to notice you've unfollowed them. How are they going to feel about that? They're not exactly going to become a bigger fan, are they?
So rather than unfollowing "inactives" from an app, I decided to use the app's "inactives" function only to make a list of people who hadn't tweeted for a long period. I would then check each user's Likes page manually, to make sure they hadn’t been Liking tweets with a similar theme to mine.
OMG, was that process a revelation…
The second user I checked - someone who hadn't tweeted for months and had shown up as “inactive” - had not only continued to Like tweets, but a tweet of mine was right at the top of his Likes list. This was the exact person I didn't want to unfollow, and thanks to the extra step of checking his Likes page, I'd managed to avoid doing so. Even better, I now know who he is. He’s no longer one of the people whose thanks get lost in the Notifications page noise.
It's pretty easy to see who's an engaged reader and who isn't.
I realised that this was no longer an experiment - it was going to be my default practise for all future unfollowing of supposed "inactive" accounts.
Having checked the Likes pages of everyone on the app's "inactives" list, I was astounded with the results. Most people who hadn't tweeted in more than a week, had still continued to use the Like button after their last tweet. Meaning they were not inactive for the periods denoted by the app – which works in the same way as all similar apps I’ve used in the past. The headline, then, MOST “INACTIVE” TWITTER USERS ARE PROBABLY STILL LOGGING IN. They’re just not tweeting.
But that was only the starter. Checking each "inactive" user's Likes page proved to be a fantastic window into their interests and motivations. Even if they're not liking your tweets, you get to see if they're Liking tweets with a similar theme to yours.
On the Likes page, you can normally tell at a glance whether the user followed you out of interest in your subject matter, or whether they were just looking for a followback. If their Likes page is full of random spam, it’s doubtful that they’re reading their timeline. The probability is that they have an app Liking tweets for them, solely in a bid to invade people's notifications. It's pretty easy to see who’s an engaged reader and who isn't. You don’t have to do much scrolling.
You can normally tell at a glance whether the user followed you out of interest in your subject matter, or whether they were just looking for a followback.
But there's more. You can additionally see by inspecting a user's Likes page, if they only Like tweets into which they've been @ tagged. This is actually very telling, and would suggest the user does not read their timeline. How so? Well, because if a user solely Likes tweets into which they've been tagged, they have no qualms about Liking tweets per se. But they're confining their Likes to tweets they can access from the Notifications page. Therefore, it's highly doubtful that they regularly look at their Home page. The takeaway? Unless you're tagging them, these users are never going to see you. It doesn't really matter whether they follow you or not.
Then there's the frequency of a user's Likes. This can give you a sense of how often they're on Twitter, but more than that, whether there's been any breakdown in their usage pattern.
For example, If someone has 35K Likes in their list, but they currently only Like a tweet once every five or six weeks, the amount of Liking they do has taken a nosedive. Have they lost interest in Twitter? Moved to another platform such as Facebook or Instagram, maybe? People's behavioural habits tend to remain pretty constant. So a high Like total coupled with a very modest amount of recent Liking, probably does mean they very rarely now look at their follow timeline.
It may be the Likes page, rather than the main tweet feed, that reveals the most about whether the account is run by a real person, or a bot.
Another common picture is that of a user Liking regularly, but confining their Likes to the same one or two users. This is a potential indicator of low engagement, since in a lot of cases the one or two profiles the account holder repeatedly Likes are their own. If you're following a thousand profiles, but you always end up Liking the same one or two, you're probably Liking those tweets from somewhere other than your follow timeline.
Few people consider the idea of skimming down a Twitter user's Likes page before unfollowing them, but the process is quick, and it provides a lot of key information in seconds. I now regard it as essential.
In fact, once you've got used to doing this, you may want to try it with some profiles that don't show up as inactive. Just because an account is tweeting, it doesn't mean there's someone there. And it may be the Likes page, rather than the main tweet feed, that reveals the most about whether the account is run by a real person, or a bot.