I Tested Twitter's Competitors for You
There are at least five major sites attempting to dethrone Twitter as the king of microblogging sites. Which, if any, are ready to do so?
If you hadn’t noticed by now, Twitter’s got problems. The microblogging site that has been popular with journalists, activists, and political junkies for over a decade was bought by Elon Musk last year, and things have gone badly from the start. The problems are too numerous to name every one of them here, but firing 75% of the DevOps team (and most of the rest of the company’s personnel), and replacing the “blue check” verification system with a “pay for play” model were two of the biggest mis-steps. After a weekend where Twitter effectively ceased to function, the rush to leave the site began in earnest again.
Since it became clear that Twitter was taking a turn for the worse under Musk, a number of competitor sites have sprung up, waiting like vultures for the inevitable. This includes Post, BlueSky, Spoutible, and most recently Meta’s (formerly Facebook’s) offering called Threads. There’s also Mastodon, which has been around since 2016.
Despite coming from a long line of late adopters (my father had a Betamax until the mid-90s) I decided I needed to make sure I got in on the ground floor of whatever might one day supplant the gravely wounded Twitter, which appears to be having its MySpace moment.
Also, I didn’t want anyone else claiming my usual username. I spent a lot of time and effort building up 125k followers on the dead bird site.
There were several sites I didn’t evaluate. Spill, though promising, is still too new, and likely is too left leaning to gain a wider audience. Telegram is Russian owned, and absolutely crawling with Nazis, Russians, Russian disinformation, and Russian Nazis who post videos of themselves committing acts of unspeakable horror. This doesn’t seem like a great business model. Tribel aims for a left-wing user base, and suffers from having a relatively small number of users. Gettr, Gab, and Truth Social are all far right sites with little moderation. People who don’t like hanging around with deranged right-wingers spouting conspiracy theories and disinformation are unlikely to adopt these platforms. Other sites aren’t really for microblogging, including Reddit, LinkedIn, and Facebook.
I went about creating accounts with the five Twitter competitors, with varying degrees of difficulty doing so. The first thing to do was evaluate the basic functionality of each in easily measurable ways where possible (measures of performance). The metrics are for the “vanilla” version of each: no browser extensions or quality of life apps. A short description of each of these measures of performance is listed below. (Note that I’m not going to evaluate the security of these sites and the data contained therein; I’m not a cyber-security expert or a pen tester.)
IOS: Is there an Apple IOS app for this platform?
Android: Is there an Android app for this platform?
Web-browser: Can you use this platform on your PC or MAC?
DMs: Can you send individuals direct messages?
Group DMs: Can you send direct messages to groups of people?
Hashtags: Does the platform support hashtags to identify and locate posts about a particular topic?
Following Only: Does this platform allow you to look at a feed that only includes people you follow?
Threads: How well does this site support posting multiple posts in a row to tell a story or include multiple supporting links (like citations or footnotes)
Search Engine: How well do the search functions for people and topics work? (This is somewhat subjective)
"For You" Quality: If the site has a feed of posts it picks out for you, how relevant are those posts?
Images: Can you load images in your posts?
GIFs: Does the platform support loading GIFs into your posts?
Videos: Does the platform support loading videos into your posts?
Article Thumbnails: If you include a hyperlink to an article does the platform properly display a thumbnail and link to it?
Billionaire Owner: Is the platform owned by a billionaire?
Block: Does the site allow you to block accounts?
Edit Posts: Can you edit posts after sending them?
Lists: Does the platform support making lists of people whose content you want to see in one feed?
Bookmarks: Can you bookmark posts so that you can find them again later?
Quote Reposts: Can you re-post someone else’s content, and include your own comment at the top?
Total Engagement: I posted the same image on all the sites (the matrix below), and observed how much traffic (comments, re-posts, and likes) each generated.
Engagement Ratio: Number of reactions to my post divided by how many followers I have on that platform
Moderation: Subjective measure of how well a site does at taking down hateful contents and removing users guilty of harassment.
Account Verification: Can you get a blue check (or similar) by proving you are who you say you are?
User Base: Estimation of how many users the platform has.
Waitlist: If the site is still in beta testing, how many people are waiting to get in?
I went through and evaluated each of these sites based on these criteria, and produced this color coded matrix.
However, looking at the results of this, I realized that there was only so much you can get out of the above matrix beyond how well developed the functionality of each of these sites is. (Finding out that Twitter, which has been around since 2006, has the most implemented features is not particularly revelatory.) Quite a few of these functions have asterisks that need explaining. There are qualitative pros and cons that don’t fit neatly within the matrix. The matrix doesn’t capture the “feel” of each site, nor does it capture some of the biggest reasons why you wouldn’t use some of them (like how the algorithm handles content, or a janky and dated User Interface (UI)).
So, I’m going to give a quick description of the biggest pros and cons of each platform, along with a qualitative assessment of my own experience with each.
Pros: Giant user base. Has the most features, most of which are better implemented than competitors (and also are pre-Musk legacy code).
Cons: Musk. Almost total lack of moderation. Musk. Pay-for-play model. Elon Musk. DMs are full of crypto scams and “kill yourself commie tr*nny c**t before I do” death threats. Elon. Site reliability is degrading quickly. Disinformation and people agitating for stochastic violence are given favored status by ownership (<ahem> Libs of TikTok <ahem>). And did I mention Elon Musk?
Qualitative Factors: The Twitter experience has been in sharp decline since Musk took over and basically ended content moderation, or biased it in favor of right wing trolls (like Andy Ngo). It increasingly feels like 8Chan with better production values. The only reason it’s still even remotely usable is that I ran a massive block algorithm a while back, which now can’t be done because Musk blocked user access to the Application Programming Interface (API). There’s been talk of making the block feature available only to blue checks (i.e. paying customers).
Verification is a farce: anyone can pay money to claim to be anyone, which backfired hilariously. Most blue checks are conservative people who post frequently but have few followers. The result is the site promotes responses from people who generally have nothing interesting to say. The result is a platform which has the best functionality, lots of people I want to interact with, but makes the experience as unpleasant as possible by promoting the people who send me the lovely “kill yourself commie tr*nny c**t before I do” DMs.
But, with 125k followers, it still functions well for generating engagement (for me at least). When I ran the engagement test, it generated 53 comments, 133 re-tweets, and 379 likes. This was more than the other sites. I doubt it would work as well for someone with fewer followers however: the follower to engagement ratio isn’t great.
Unfortunately, until something else emerges as the winner, I’m kind of stuck with it if I want to interact with my favorite people online in one place.
Threads
Pros: Linked to your Instagram account to import followers. Big user base that’s growing quickly. Having the Meta Corporation funding it means there will be continual improvements to functionality.
Cons: Very limited functionality. Search engine is bad at finding people of interest. No PC/MAC web browser support. What functions it has are buggy. No “following only” feed. The “suggested for you feed” is stuff I don’t care about. Not available in the European Union. Can’t delete your threads account without deleting your Instagram account too.
Qualitative Factors: Bottom Line Up Front (BLUF): Threads as it stands now is completely unusable for the sorts of things I use Twitter for. I can’t easily find the people I want. I can’t filter what I see. It’s difficult to create long, interesting threads for people to read, and the algorithm will de-boost them anyway because of the sorts of things I write about.
Adam Mosseri, the head of Instagram and Threads, has outright said that Threads is there to be light entertainment, and will avoid controversial things, like hard news and politics. There’s no point to me, or other writers, posting there.
When I ran my engagement test, the effect of the algorithm showed. Despite 1000+ followers, the matrix generated 13 replies, and 9 likes, and no shares. Most of the engagement centered on how I shouldn’t be measuring functionality right now. It didn’t feel like a great conversation. Even the response to follower ratio was the second worst, behind only Twitter.
The Threads verification is a blend of old twitter and new Twitter. Which is to say you can get a verified account if you’re famous using the legacy Instagram process, or you can get a “verified” account for $15 a month without providing any sort of proof that you say you are who you say you are. This creates the same problems as Twitter with its verification process: even if you see someone with a bluecheck, nothing has been done to actually determine if they are who they say they are.
Threads is missing a lot of key features, and some that it has are still very buggy. I tried loading videos from both my phone and PC, and they universally had the sound stripped out of them (both videos worked fine on Instagram). Threads very much feels like something that was released way too early: we are looking at a product with 75 million users that should still be in closed beta, if not alpha given the bugs in the very limited core features.
Other users rightly point out that Threads is brand new, and that the code and functionality will improve over time. True, but I can only evaluate it as it stands now, not based on what might be in the future. Additionally, it says something about Meta’s commitment to quality that they’re a multi-billion dollar corporation, yet still rolled out a product that should probably still be in closed beta testing (based on limited functionality and glaring bugs) out to 75 million users.
The best way to summarize my Threads experience is that it’s great if you want to read about what a celebrity had for lunch instead of seeing a video of what they ate on Instagram. It’s terrible at providing or disseminating the sorts of content I get at Twitter.
Post.
Pros: Great for news aggregation. Permanent hashtags instead of lists is interesting. Focus on hard news means very little disinformation and trolling.
Cons: Small user base means lack of interaction. Pay-for-access aspect is novel.
Qualitative Factors: Post spent months in closed beta, and I was one of the people on the waiting list. It has been open access since April 2023. The concept behind Post is that you can pay something around 25 cents to look at any article without having to get a subscription to the NYT, WSJ, or other news organizations that put their content behind a paywall. When I got in, there’s plenty of news in my feed. Finding articles using the hashtag list system is relatively easy. On the plus side, it has decent functionality compared to some other alternatives. Post fared somewhat poorly on the engagement test: 6 comments, 6 re-posts, and 17 likes. However, I only have 430 followers there, so the ratio is about average. On the downside, there’s so few people there that content engagement is minimal: it sometimes feels like a ghost town that has a world class news stand operated by a robot.
Still, there’s value in visiting that newsstand.
BlueSky
Pros: Good moderation. Invite systems during closed beta testing has kept the worst of the trolls out.
Cons: Second worst functionality among platforms tested. Still very much a closed beta product. Small user base limits interaction rate with your posts.
Qualitative Factors: Getting an invite to BlueSky was the hot thing a couple of times when Twitter stumbled. It definitely has a “Twitter-circa-2010” feel to it, which is both good and bad. On the downside, it’s a Jack Dorsey product, and that screams “haven for libertarian tech-bros” who have no understanding of what it’s like trying to exist as a woman, person of color, or sexual minority on social media. While Dorsey isn’t the owner, he’s on the board of directors, and it’s an open question how “hands off” he’s willing to be in the long run.
Threads sort of work, but other users can’t see that it’s a thread without clicking on the original post. This can be mitigated by including 1/n or similar to clue readers in that there’s more content. Engagement is right in the middle of the pack, both in terms of total engagement and engagement rate.
Moderation hasn’t generally been needed so far, but then again, there’s hardly anyone there yet. The search algorithms for other users or content is okay, but not at the level of the more mature platform of Twitter.
Mastodon
Pros: Been around since 2016, so there’s considerable functionality. Dedicated user base. Open-source code, and no billionaire owner. Self-policing and robust (sometimes over-zealous) moderation.
Cons: Janky and unpleasant UI. DMs aren’t private. Search algorithm is so-so. Missing several crucial functions that I need for the types of posts I do.
Qualitative Factors: I would describe my experience with Mastodon’s UI as “relentlessly unpleasant and needlessly complex”. The UI differs significantly from the other platforms. It is difficult to look other people up, and communities are siloed. Even when you find someone you want to follow, you cannot see who they are following to help you find other interesting people. You cannot quote-RT other people’s posts. You cannot post threads. When I post hyperlinks to articles it doesn’t bring up thumbnails: it just leaves a “naked” link where people cannot see what the article was. The federated feed goes by too quickly to read, and the local feed is 50 posts in a row by the same guy, most of which aren’t worth reading.
Even evaluating engagement took more effort than it should. It doesn’t the show number of replies, and you can only see re-posts and likes by clicking on the post itself. I ended up hand counting unique replies.
The explanations on how to fix or work around these issues by fans of Mastodon reminded me of Jordan Peterson fans explaining how I needed to read 30 other things to get the context of the nutso things he was saying. If it takes one or more obscure apps and browser extensions to simply see a thread posted in the correct order, it’s not going to catch on with the general public.
Mastodon seems like something people who drive manual transmissions and program in Unix would love. For the general public, it requires too much effort and technical experience for it to catch on. Thus, I don’t see Mastodon as a long-term solution to Twitter’s implosion.
Spoutible
Pros: Not billionaire owned. Some unique features, like the embedded bot rating and automatically numbering the posts in a thread. Better functionality than one would expect.
Cons: Small user base. Controversial owner and creator.
Qualitative Assessment: Spoutible was ostensibly created by Christopher Bouzy (creator of Bot Sentinel) with the intent of cutting down on misinformation, bots, and harassment. While not a billionaire, Bouzy is also no stranger to controversy. On the whole, I was pleasantly surprised by the functionality, engagement, and interface on Spoutible. The built in bot detector isn’t all that necessary now on a site where the user base is there by choice, but would be amazing on a big site. The “Help” feature doesn’t have a huge library, but it’s better than expected and readily accessible.
Engagement rate was excellent: the user base is engaged and passionate about the platform. Transparency on upcoming features is good: leadership is working on an iOS app, an enhanced version of lists, and a capability to do something akin to Twitter spaces. Spoutible has a verification system in place, but it is not taking new applications until Q3 of calendar year 2023.
On the down side, the threads features don’t make it easy to track who is responding to what. Replies aren’t shown directly beneath what specific quote they are responding to, and aren’t indented to show how far down the tree you are. Additionally, sometimes the reply Graphical User Interface (GUI) Box for replies doesn’t work properly, and the things that I typed disappeared into the ether.
However, It’s rather damning that Bouzy and a small team could build something with limited cash reserves that’s far better that what Meta could do with an entire corporate empire behind it. The biggest problem with Spoutible is that it has a small user base. Putting aside all the other factors, of all the Twitter competitors Spoutible has the best desktop UI by far, and its total functionality is on a par with platforms (Mastodon) that have been around far longer, and its ease of use is very good.
If Threads had rolled out Spoutible as a platform, I’d be willing to pronounce Twitter’s time of death. Unfortunately, the small user base, and limited resources of the development team, are going to be a huge issue going forward. I wonder whether Sputible has the resources, architecture, and operating model to massively scale up its platform. In many ways, this is Betamax vs. VHS all over: the superior technical product is in danger of losing the war to an inferior one due to the ubiquity of the latter.
Conclusions
Twitter is devolving into an unusable hell-site due to its increasingly deranged, out of touch, and fascist owner, whose online radicalization is being watched by millions in real-time. Unfortunately, none of the competitors are ready for prime time. Most simply don’t have a big enough user base to support the sprawling world community we had on Twitter. The only one that does have a huge base (Threads) should still be in closed beta, and it doesn’t provide the news and political information that a lot of us want and need. Some lack the functionality and usability necessary to compete in the long run (Mastodon).
Each of the competitors is deeply flawed in some way. Threads would be in the best position to challenge Twitter, except that its core concept (entertainment) is unlikely to lure away Twitter’s core users (news and politics). Spoutible provided the best experience in terms of functionality and usability, but was also the platform with the least amount of resources of the six.
Thus, without any suitable challengers, the decline of Twitter is likely to remain slow and painful, rather than a sudden collapse. If they keep failing to pay their bills at HQ, maybe we’ll just wake up one day with their furniture on the lawn, the cops evicting everyone (including Musk), Twitter dead, and everyone else forced to scramble to pick a site.
"Mastodon seems like something people who drive manual transmissions and program in Unix would love." -- as I've commented on Mastodon, I miss my manual transmission car. Thank you for this analysis, Brynn.
Spoutible is built on a readily available PHP script called "ColibriSM", which just cost a few bucks - you can check @doubtible on Twitter, they use the same one and debunk most of Spoutible CEO's bogus claims. The developer of ColibriSM, MansurTL, also confirmed that Spoutible is using his script.
Upon release of the platform, Spoutible's CEO forgot to change a default setting for the API, which led to tens of thousands of users' email- and IP-addresses being publicly available for a while.
The userbase is even smaller than he officially claimed (e.g. in the WIRED article). There are still less than 200,000 registered users on the platform and it's already gaining less and less new users each day.
Oh, and Spoutible's CEO is known for being a scammer in the crypto space, where he used the alias "Iconic Expert". He regularly harasses and doxes people on Twitter.