Reading Time

4 Things Twitter Could Do to Make Blue Worth Paying For (But Probably Won’t)

Over the past week, I’ve been thinking a lot about why Twitter Blue has not been the success its new owner had hoped for. While subscription numbers are hard to assess, Endgadget reported that as late as mid-January, Blue had only 290,000 subscribers worldwide, which doesn’t come close to Musk’s demand that the company get half of its revenue from subscriptions. Using absurdly basic math, you’d need 15.6mil subscribers at an $8/mo average to meet half of Musk’s revenue expectation of $3bil.

If you’re in podcast circles, you’ll sometimes hear that anywhere from 1 to 10% of your listeners will subscribe to a Patreon (in my case, it’s about 3.7%). For Twitter, the target is around 3.4% (depending on the number you use for active users), which should be achievable in a short time frame. And yet, all data suggests that this isn’t remotely the case.

But why? What is keeping Twitter from reaching a subscription milestone? There are probably dozens of reasons, from violations of the trust thermocline, Doctorow’s enshittification theory, distrust in the ownership of the platform, the consumer conception of value as free things become paid things, and so on. Twitter, in other words, is a mess.

And this has led me to a big question: Does it have to be? If I were to offer a snarky response, I’d say that “this is Twitter, after all.” However, a more serious response would be: no, it doesn’t. Twitter can be better than it is (and has been before Musk’s takeover), and Twitter Blue can be something users might consider paying for. But to get there, I think Twitter has to make some serious changes to its operations.

Here, I’d like to suggest FOUR things Twitter could to do to make Blue worth paying for.

Here goes…

1. Hand the Reins to Someone Else

It’s no secret that Musk’s tenure as head of Twitter has been tumultuous (tempestuous, even). While it’s hard to know the actual truth behind the stories of his behavior within Twitter, such stories are made believable by his antics as a user (appearing to make major decisions based on polls, incessant trolling, and threatening to fire or sue current/former employees over criticism). All of this is amplified by the tendency of Twitter itself to make announcements of drastic changes (or make the changes themselves) without even the appearance of significant forethought, often without consulting anyone who might be affected by those changes or providing coherent information in advance about what those changes will involve (see the Twitter API controversy as a prime example).

When it comes to Twitter Blue, those decision processes have meant siloing security features and platform quality behind a paywall, effectively forcing people to accept a worse Twitter in exchange for the right to pay for the Twitter they already had. None of those decisions have actually made Twitter better; in many cases, they’ve actively made them worse.

Whether Musk is simply out of his depth, overwhelmed by his financial obligations, really is purely incompetent, or all of the above, the results are pretty clear: under his leadership, Twitter is less safe, less functional, and less appealing. Twitter Blue, which existed prior to Musk’s ownership, hasn’t benefited from his leadership either. It solves none of the problems which have plagued Twitter for at least a decade, and, as I’ll discuss elsewhere in this post, doesn’t provide added value for the user. While Blue does have some features that users have asked for (supposedly, it offers an edit button), most of what it offers are things you get for free on other large platforms or things which would only be valuable if they were restricted to a smaller userbase (i.e., if the goal is to get a very large portion of users to pay for Blue, your claim that you’re providing a way to cut through the algorithm quickly falls apart, since any sizable body of individuals will still make finding anyone in the sea nearly impossible at best).

Basically, Twitter under Musk has been a disaster. Worse, his behavior — which I would charitably describe as harmful — has pushed existing users away and convinced potential new users that they were right to stay away. I said on Twitter a while back that almost none of my students use Twitter, and the reasons they give fall into two categories: it’s not their thing OR it’s a hellhole. Essentially, they’d rather be somewhere else. Musk hasn’t fixed the first reason, and he’s actively made the second more accurate. You can only court controversy as a business strategy in this industry space for so long (or, I should say, you can only do that for a platform this large for so long; people of similar character have had more success at smaller scale, but a site of hundreds of millions of potential users is not Truth Social). Arguably, the only reason Twitter hasn’t collapsed yet, despite many apocalyptic assertions over the last 6 or so months, is that there isn’t a viable alternative that is basically Twitter but not (more on that another day)(maybe more on that another day).

And there’s really only one way to solve this: put someone in charge who knows the industry and the technology AND has the disposition to manage the company competently. Someone more invested in making Twitter (especially Blue) a good experience than in being a troll or, worse, a political reactionary who uses their power for their own selfish ends. At the very least, it would reduce the chaos associated with the platform under Musk’s leadership, which might remove or reduce one of the central reasons for user reticence to pay for Blue.

2. Add Value

If you want people to pay for your platform, you have to give them something they can’t get elsewhere OR provide a better service than a competitor OR provide a better version of the thing you’re already giving them (or all three).

Twitter Blue does neither of these things (or does too little of them to justify its asking price — or, worse, to convince people to set aside their moral stances on Twitter).

Currently (as of 2/18/23), Blue offers the following:

  • Early access to select features (which mostly don’t exist yet)(not an addition)
  • An Edit Tweet function (legitimate addition)
  • Bookmark Folders (only really valuable for the app because your browser can Bookmark stuff; we’ll call this neutral)
  • Custom App Icons (neutral because it’s not a meaningful addition)
  • NFT Profile Pictures (only meaningful to people who care about NFTs; not a true addition)
  • Themes (only valuable for the app because you can already do this in a browser with a Chrome add-on AND with a third party app (we’ll come back to this); we’ll call this neutral)
  • Custom Navigation (only valuable in the app for the same reasons as Themes; we’ll call this neutral)
  • Spaces (this existed before but has been siloed; it’s now a subtraction)
  • Top Articles (basically, trending but for your followers; I’ll be charitable and call it an addition)
  • Undo Tweet (I actually don’t understand how this feature could possibly work, so it’s a meaningless feature to me; let’s call it neutral just to be nice)
  • Prioritized Rankings in Conversations (this is neutral because it only adds something if Blue isn’t brought up to scale; since Blue is intended to be scaled up, prioritization eventually becomes meaningless)
  • Longer Video Uploads (not a true addition because other platforms can already do this and better; however, it does add a feature, so we’ll count it as such)
  • Half the Ads (not yet in effect; additionally, browser add-ons and third party apps can remove these anyway, so it’s not adding anything meaningful in my eyes)
  • Longer Tweets (not only do many Twitter users not want this, but Twitter is also not set up for this to work; even so, other platforms do this better than Twitter, and so this is only an addition because it is actually adding something)
  • SMS two-factor authentication (this is an obvious subtraction, and a serious one at that)

I’ve taken the liberty of evaluating these based on how I perceive them (I hope I didn’t miss anything).

When you collect all this together, you end up with a handful of additions, slightly fewer subtractions, and a sea of neutral changes. This would suggest that Twitter Blue is, from a consumer standpoint, a neutral choice. However, some of the additions aren’t yet meaningful (poor integration or unwanted), some of the subtractions are actually quite serious (removal of open community access and fewer security options), and many of the neutral choices are either meaningless because you can already do them via other means OR weak additions (lackluster cosmetic features, for example) OR negative depending on your perspective (people really hate the idea of longform tweets).

Blue, in other words, has a serious problem depending on the perspective you hold: either it doesn’t add enough to justify its cost OR it adds nothing of real value and, thus, isn’t worth its cost. Me? Even if I set aside my opinions of Musk’s behavior and politics, I don’t see why anyone should pay for blue. Assuming the $8/mo figure holds, you’re not really getting anything substantial for the price — not for what Twitter does or by comparison to its free competitors. What features it does add are, at best, lackluster, either because they’re extremely basic, extremely replaceable in a browser, or not fully integrated into the platform (either by users or the platform itself).

If Twitter were an early access game on Steam, it would be the equivalent of a subscription-based game promising an open world but only giving you access to one part of it and never telling you exactly when the rest of the game will be completed OR what else will be in the game when those changes are made. While I’m sure a developer has done this at some point, it’s not a common model for soliciting long-term support (and this despite the fact that games don’t necessarily need to survive long term).

Then there’s the big elephant in the room: if you want someone’s money after they’ve been granted access to a thing for free, you have to actually give them something worth paying for. Siloing features they already had or adding things that don’t make substantial adjustments to the experience of Twitter are not suitable additions. You need more.

Blue, however, doesn’t offer more. It’s a cash grab.

3. Better Communication and Decision Making

Before Musk took over Twitter, it wasn’t uncommon for Twitter to solicit feedback from users, float specific ideas to gauge public reaction, or test ideas out in phases. While I’m sure Twitter is still mostly doing this, much of Twitter’s current communication strategy could be described as destructive. Changes which affect developers and users alike have been applied with almost no advanced notice; worse, those changes have lacked sufficient details to make meaningful reactions beyond panic impossible. I’ve seen developers and creators who rely on Twitter’s previous API rules in total panic over how they’re going to keep going — some because of financial limitations and some because they don’t have any information so they can make changes in a suitable timeframe.

Many excuses have been given for these changes, but those excuses often read as dishonest or poorly considered. It’s hard to argue Blue’s verification process in the early days wasn’t a total blunder because it was a blunder so obvious that one wonders how Musk didn’t see it coming even though he was almost certainly warned by Twitter users (whether an employee warned him is a different matter given his tendency to fire people who disagree with him). There have also been petty changes, including the almost transparently personal decision to ban Mastodon, banning journalists for criticizing his decisions, and so on.

None of this is really surprising, though. Twitter is being run the way one runs a top-down company, which probably works more effectively in industries where decisions have fewer consequences. As a result, Twitter’s communication and decision making processes feel more like the site is under the hand of a dictator.

But Twitter should be run as a user-centered business. That means returning to the less opaque communication strategies of pre-Musk Twitter (and improving upon them) and putting together a formal decision-making process that bases those decisions on well-researched data. Doing so would, one hopes, give developers and users access to the conversation about decisions AND give them plenty of leeway to adjust to forthcoming changes. That means relinquishing control (at least a little bit) in the same way as Patreon and other sites have done.

While doing this probably won’t end the panic that comes with change — after all, users have revolted against Twitter changes since the site was born — it would mean that reactions will be (sometimes) tempered by giving people time to adjust.

Of course, this also means those changes need to be things that people can reasonably adjust to. And this is where Blue fits into this item. The idea that Blue could be a test platform for all of the neat features Twitter wants to play with is not a horrible idea (though they actually have to test things). An even better idea would be to integrate your user research into Blue. Here, you have an audience of people who want new and better features, which means you’ve got a group of invested users for soliciting feedback and testing new things.

And with better communication and test processes in place, this could be used as an advertising tool for Blue: join Blue, and you’ll get first run at all new features AND be included in focus groups to help us build a better Twitter. Some of those features might stay in Blue while others might go to classic Twitter. Either way, you’ll have shifted Twitter more explicitly to a user-centered platform, one where what we say has real meaning and value because we’re literally invested in the platform with our money.

But to make that work, you need clearly formed plans, clear communication strategies, and clear steps for integration of changes so everyone knows what’s going on and when. Without that, it’s just more chaos.

4. Leave Regular Twitter Alone

Or, at the very least, keep it close enough to what people expect from Twitter as possible. Change is inevitable, but restricting access to features and trying to force users to engage with Twitter according to corporate dictates hasn’t improved the perception of Twitter among users and the public at large.

One of the reasons for this are the changes we’ve seen to the everyday Twitter user’s experience, from removing security features to trying to nerf third party apps to other visual changes which affect how people use the platform. Most damning of all has been Musk’s decision to reinstate users who were previously banned, which, at best, has led to a lot of vulnerable people leaving the platform and a lot of others having to deal with harassment. Again: none of this has improved Twitter for anyone. We’re not having better conversations. We’re not engaging more meaningfully. Twitter is still a shit show.

Blue, as I’ve said above, has to be a dramatic improvement upon the Twitter experience to convince enough users to actually pay for the platform. Restricting the classic experience doesn’t make Blue a more enticing feature; it makes users aware of what is being taken away from them. You’re basically trying to convince people to spend money by making them feel bad about not having something they used to have. That’s not a great motivator for users, especially when other platforms exist and offer most of their features for free.

Twitter could reverse course on this pretty easily by only updating the classic experience of Twitter to help it keep up with competitors while putting more energy into building a better premium experience for anyone who wants it.


Having said all of this, I think it’s important to note that I have no confidence that any of these things will happen, nor do I think they will solve Twitter’s bigger problems. It has been well reported that Twitter has a money problem. A big money problem. That includes Musk’s loan payments, expenses to run Twitter, the cost for development, and more. I don’t know if there is any actual money in the Twitter coffers to do anything substantial to improve the platform. This is one of the reasons, I think, behind the weekly declaration that Twitter is dying, which we’ve been reading for at least 6 months now — the platform simply can’t fix itself because its hemorrhaging money.

I don’t feel particularly sympathetic to Musk. Even if I set aside his behavior on and off Twitter and some of the decisions he’s made which I think are antithetical to a functional social media space, he ultimately played billionaire chicken and, thus far, seems to have lost poorly. From what I’ve read, he put himself into a position to be forced into the purchase; whether he did that because he thought he was being clever or for some other reason, I think he ultimately went through with the purchase without any real plan for the platform. At the very least, if he has a plan, nothing he’s done thus far seems to suggest it’s a coherent one.

And so Twitter is stuck between a rock and a massive drill bit that is slowly grinding its way down. Do I think Twitter will eventually die? If it keeps going like this, yes. I don’t know how it survives under its present conditions. The only way it will get through this is with change. That means some egos need to step aside and some processes need to change so they can either build a better Blue experience worth paying for OR pursue the comparatively easier task of bringing back major advertisers by returning to a less toxic Twitter.

But I will say this: literally everything I had planned about Twitter for the next cycle of my Digital Writing courses (specifically, Digital Rhetoric and the Rhetoric of Social Media) has to go out the window now. Hell, I might not be able to plan anything until the last minute given Twitter’s cyclical chaos. From the researcher side, it’s been an interesting near-year; however, from a user’s perspective, it’s been remarkably disruptive in all the wrong ways. Hopefully, things will turn for the better…

Email
Facebook
Twitter
LinkedIn
Digg
Reddit
LinkedIn

Get My Newsletter!

Subscribe (RSS)

Support Me

Recent Posts

Top Posts

Archives