NEW POST

I'm curious how various social media services compare in engagement these day. So I compared reactions to a couple of dozen of my recent posts

https://martinfowler.com/articles/2025-social-traffic.html

Social Media Engagement in Early 2025

Comparing engagement on two dozen recent social media posts

martinfowler.com

@mfowler 'I guess that means LinkedIn people are more eager to hit the like button.'

or maybe just more likely to exist?!

@tastapod but's comparing to reposts, which they still need to exist to hit the button

@mfowler I really like the way you broke down the data. The only issue I have with the post is that LinkedIn and X are very heavily botted (and they can be very hard to spot). I don't know about BlueSky. And I don't think Mastodon has had the same targeted bot networks setup yet. (Or, because it's federated, maybe they have a harder time gaining traction.)

I wish there was a way to tease out actual users from bots, because I'm curious if actual thoughtful engagement would be different.

@gregatron5 The question then is to what extent the bots do reposts. Hopefully that's not too bad, but I agree, there isn't really a way to figure that out.
@mfowler BTW, I read *Refactoring* not long after it was published and refactoring has been a bedstone of my career since then. Thank you for such a great book!

@mfowler I suspect that the LinkedIn "likers" are doing it to draw attention to themselves. Their liking of your posts appears in the timelines of their connections, who can then be suitably impressed that John Doe reads posts by the venerable Martin Fowler, and so John Doe must be a clever cookie worth hiring.

Personally I hate it, because those pointless reactions clutter my feed with noise.

@elricofmelnibone Indeed, all the noise in the linkedin timeline makes it easy to ignore

@mfowler @elricofmelnibone

Yea; I've been on LinkedIn for decades, and "follow" you and others there. But not really, as I almost never "use LinkedIn like a social media site" by reading through the "Home feed." Almost only "Notifications" on LinkedIn ever catch my attention.

@elricofmelnibone @mfowler unfollow to your rescue. Twice over the recent years, I have had unfollow sessions, where I unfollowed people in my network who posted stuff I found uninteresting. The next session is soon due…
@mfowler I know it’s far more subjective and harder to track but any thoughts on the quality of engagement when it comes to replies?
With LinkedIn in particular I get the feeling that many of the replies are an attempt to “jump on the bandwagon” of engagement rather than a meaningful attempt to engage with the OP?
@mfowler did you also look at click through rates/ referrers of clicks to your posts?
@mfowler I'm asking because a German publisher (heise, biggest tech publisher in Germany) did that analyses several months ago and noticed that commercial platforms perform badly in driving traffic to external sites (which makes sense given their business model)
@mainec @mfowler I too would wonder. It’s one thing to know how many people are sharing and liking the posts on each site, but I’d be interesting to know how many meaningfully engage with the content, I.e., actually read the post.
@philip @mainec I don't have that data, but have added some data about sources to the overall site
@mfowler How about normalizing the results based on the number of followers on every platform? That would tell us whether your followers on different platforms behave differently.
@malt I care more about the absolute number. Even if all five followers like an article, it's still only five people.
@mfowler I get it. Not saying you should care about it, just that it would be an interesting data point.
@mfowler have you looked at ratios of repost/replies/likes per follower to gauge the "engagement" or effectiveness on each platform ?
@Thomaslissajoux I haven't done that math, although we can get a sense of it by comparing to overall follower count. In any case, the ratio doesn't really matter - what matters is the absolute number.

@mfowler How much do reposts matter if the reposter has significantly fewer followers?

My repost is much less useful than just about anyone else's.

@n3bulous I don't think of it that way. Reposts matter in that I think they correspond to people who think the article was useful enough to them to pass it on. When counting that, I'm not worried about how many they pass it on to.
@mfowler, this is definitely interesting. First, the tribal bits: Mastodon seems much more efficient at bringing technical content to my attention than LinkedIn (LinkedIn’s algorithms seem to favour AI product peddling over content from the tech and academic people I want to hear from).
Now for the non-tribal bits: Would you consider crossposting to Nostr? It's still small and quite incipient in terms of tech content, but Uncle Bob and a few others are there.
@mfowler this matches my experience. Engagement on x dropped significantly in the last couple years and linked in became much more viable for me.
I think a huge confounding factor are the topics we talk about...
@mfowler I visit your websites via RSS. It could be interesting to compare such visits to social media sources.
@banterCZ Indeed it would, but to do that I'd need to exert myself enough to use tracking URLs.