Tweetstorms used to be a hacked-together means of avoiding character limits on Twitter messages. Today, they’re everywhere. Twitter, especially if you follow founders, venture capitalists, and journalists, looks more and more like a long-form blogging platform every day.
How to Write a Tweetstorm
There are a few good reasons for that. For one, a tweetstorm is easier to write (once you’ve done it a few times) than a blog post. It’s more punchy, more conversational and requires less setup.
For another, it’s easier to get eyeballs on the stuff you’re writing. On a blog, starting from nothing is incredibly challenging. On Twitter, just a handful of people liking and retweeting parts of your tweetstorm will sway the algorithm to show your content to more people. And Twitter wants your content to spread—unlike WordPress, which has no stake in your content’s success nor a means to bring it about—which means you’re swimming with the current on Twitter, not against it.
But to take advantage of Twitter’s unique distribution properties and get the most out of your tweetstorm, you need to understand a few things about both the structure of Twitter as a platform and about the psychology of using it.
☝️ The First Tweet Cuts the Deepest
The first tweet is the most important part of any tweetstorm because it is your tweetstorm’s elevator pitch:
- It’s the first part of your tweetstorm that people see in their feeds.
- It’s the hook that pulls them into the rest of the tweets.
- It's the part of your tweetstorm people are (usually) most likely to retweet.
Your tweetstorm’s first tweet has to work as a singular message. That way, people aren’t confused when they come across it, absent of its fellow tweets, in their feed. It must also entice someone to click—either through the promise of some kind of specific, valuable knowledge, the use of a compelling narrative device, or the signaling of some kind of unique perspective.
If you make the first tweet work as a stand-alone message, and you make it effective as an enticement, then you unlock something potentially very powerful on the distribution side. As people like and retweet that first tweet—sending it into the feeds of people you don’t know at all—you expose even more people to your little distribution-optimized thought bomb.
When this works well, it can expose your work to many, many more people than you would ever be likely to reach with a blog post alone.
1️⃣ The Elevator Pitch: Exploiting the Curiosity Gap
In order for your tweetstorm to succeed, your first tweet needs to get people to click through to read the rest of it. Without a first tweet that makes people read on, your tweetstorm is losing a large percentage of its potential audience. This is because Twitter, by default, hides most subsequent messages in long threads. People won’t see your cleverly worded fourth or fifth tweet if they don’t engage with your first.
The most effective way to get people to click on a tweet to read the full thread behind it is to attract their attention by saying something counterintuitive or surprising. There are a lot of opinions on Twitter already, and at this point in the product’s life cycle, it’s a lot harder to compose a hypersuccessful tweetstorm if it’s driven by insights that 90% of your followers already more or less agree with or believe.
Here’s a good example of this done well from Textio CTO Jensen Harris:
There’s no such thing as a “startup inside of a big company.”
— Jensen Harris (@jensenharris) May 8, 2018
This misnomer actively misleads both big company employees working in such teams as well as people toiling in actual startups.
Despite all best efforts to create megacorp “startups”, they can never exist. Here's why:
Okay, maybe most people are rightfully suspicious of corporate “startups” and innovation labs, but this does run counter to the widely accepted wisdom of Clayton Christensen’s “Innovator’s Dilemma.” This tweet makes a strong statement that also hints at some relatively unexplored ground on this topic: the idea that both the big company employees who end up working inside them and the startup employees themselves are “actively misled” by intra-corporate “startups.” The suggestion of misdirection makes you perk your ears—“Is he saying I’m personally being misled?”
The other interesting aspect of this first tweet is the axiomatic certainty offered by the last sentence. It’s one thing to write, “Megacorp startups aren’t real startups.” It’s another thing to suggest the existence of immutable laws of physics that guard against such a phenomenon ever taking place. That kind of cosmic-scale argument will always attract more attention than a simple claim one way or the other.
This example from investor Brianne Kimmel succeeds by deploying a less subtle, more “shock and awe” kind of strategy. It still taps into elements of surprise and counterintuitive curiosity all the same.
Thread: Your startup can’t afford to ignore the competition.
— Brianne Kimmel (@briannekimmel) October 11, 2018
100 million startups are launched every year.
That’s 3 startups per second. How are you going to compete?
You need a systematic way to outsmart your competitors —
The idea that three startups are launched every second jolts you into being interested because it’s such a ridiculous fact to be aware of. It’s not a statistic that gets bandied about like statistics about startup failure rates do. It’s new information. And then, in the next line, Kimmel promises insight into some “systematic,” logical ways to make sense of and conquer that information.
With tweetstorms, you’ll often find that the most compelling stand-alone tweet is buried several tweets deep in the storm. If you have personal experience with writing of any kind, it’s obvious why this happens. We often write in sequential order: you have an idea, you explore it, you eventually get to an interesting point. This is fine for a first draft, but leaving the creative process on the page makes it difficult for readers to sift through your thoughts. This is amplified on Twitter. The algorithm does not reward tweetstorms where the writer clears her throat for a few tweets before getting to the point.
This thread from Mixpanel founder Suhail Doshi, for example, starts like this:
1/ Every startup goes through the same repetitive bottlenecks: (1) you need to improve the product, (2) you need awareness to test the product, (3) you have scaling problems because people like the product.
— Suhail (@Suhail) July 10, 2018
It’s an interesting thread, but the most interesting tweet to start with would have been the ninth one:
9/ When times are dire, the worst thing you can do during this moment is to be overly patient for numbers to rebound. Freak out, don’t zen it out: https://t.co/2roqqzsDjS
— Suhail (@Suhail) July 10, 2018
The concept that a startup facing death should freak out and go into crisis mode is a refreshing and realistic narrative. As a reader, I’m curious to hear more about the argument and thinking around that. The idea that “every startup goes through the same repetitive bottlenecks” is not as intriguing for one simple reason: if every startup goes through these same bottlenecks, do I really care to read about them? Is it not a story I’ve already heard a million times?
Starting from this later tweet would have created a better hook. You can still talk about all the same stuff and have the tweetstorm make sense—it’s just choosing a more interesting place to mark your entrance into the argument. Often, you will also see popular threads that have succeeded despite not following any of the ideas in this article, such as this one from Rainway cofounder Andrew Sampson:
As the CEO of a game-streaming company, I thought I'd give my thoughts on #GoogleGDC19 and Stadia. Here's a thread.
— Andrew Sampson @ PAX South (@Andrewmd5) March 19, 2019
The tweet here doesn’t give the reader anything to go on except the implicit appeal to authority. It gives no hint of the insight contained therein. It offers no surprise or counterintuitive wisdom. In this case, it still performed well because the tweet was timely, it was from someone with legitimate expertise in the space, and the content inside is good.
It could have done even better, though.
While I don’t think it’s necessary to always “craft” the first tweet in your tweetstorm, doing so a little more mindfully—and with the recognition that this is your first and best opportunity to truly grab your reader’s attention—will help you routinely get more readers and interest in the content you post on Twitter.
2️⃣ The Most Memeable Asset: Creating a Signal Worth Boosting
The best tweetstorms aren’t just well framed and full of high-quality writing and insight. They also give readers something that they want to retweet.
In doing so, they can gain some kind of social capital or status with their readers. As Eugene Wei explains, social capital is the currency of social media. On Twitter, one way to get that social capital is to share content that makes you look good, or smart, or prescient.
Think about why you retweet anything. Except in cases where you’re retweeting a friend or colleague (perhaps out of an expectation of some kind of reciprocity), you usually retweet tweets you want to endorse to your followers. Instead of speaking, you’re letting the other person’s tweet speak for you—and, in doing so, you’re co-opting its messaging. Just look at the RTs on this tweet for proof:
In my whole life, I have known no wise people (over a broad subject matter area) who didn’t read all the time—none. Zero. - Charlie Munger
— David Cancel (@dcancel) June 21, 2015
You’re not so much trying to be funny as you are trying to look like you’re in on the joke. You’re not so much trying to appear knowledgeable as you are trying to show that you’re aware of something other people are not. And this dynamic holds the key to consistently cultivating distribution on Twitter.
The concept of retweeting to “let a tweet speak for you” is most obviously and often used in political tweetstorms, like this one from consultant Miqdaad Versi:
Some people doubt the scale of Islamophobia in the media, claim it is limited to certain views of Katie Hopkins, Rod Liddle and Melanie Phillips, and believe the far-right attitudes come from extreme rather than mainstream sources. This thread aims to challenge such assumptions.
— Miqdaad Versi (@miqdaad) March 18, 2019
You could imagine this tweetstorm with a different, more coded first tweet, something more subtle. But because this first tweet contains within it a distilled version of the entire tweetstorm’s argument—and because that distillation is a potent “point” in the contemporary discourse about Islamophobia—this first tweet becomes a memeable asset.
For the retweeter, retweeting this tweet is semiotically equivalent to organically tweeting out a message like: “Some doubt the scale of Islamophobia in the media, claim it is limited to certain views of Katie Hopkins, Rod Liddle and Melanie Phillips, and believe the far-right attitudes come from extreme rather than mainstream sources, and that is not true.”
The only difference is that by retweeting a thread, the retweeters can broadcast that sentiment to their followers backed up by the knowledge that their assertions have actual evidence behind them (and they get the display of authority that linking to a thread brings)—it gives them a de-risked way to signal an idea that might otherwise be less powerful.
The best tweetstorms all operate on this basis—they give their retweeters a signal that it is valuable for them to boost.
On startup Twitter, venture capital is a common conversational flashpoint. Many discussions about the merits of bootstrapping versus venture funding devolve into tribalistic bickering and apologia for one or the other. Because it is a topic that inspires heated arguments on both sides, it is exactly the kind of context that rewards and promotes the creation of memeable assets:
1/ A few years ago, entrepreneurs and investors would frequently advise me that Wistia was “too profitable.”
— Chris Savage (@csavage) December 21, 2018
One of the biggest mistakes I've made was believing them.
This tweet has been one of Chris Savage’s most successful ever, due in no small part to the fact that it provides people who are invested in one particular side of the VC/bootstrapping “debate” with the perfect piece of ammunition: primary evidence.
Write your tweetstorm with the social capital and beliefs of your potential followers and retweeters in mind. You’ll find that you get retweets and shares far more freely.
📝 Use a Simple Narrative Arc
Every tweetstorm needs some kind of narrative arc to propel it along. Without some kind of arc, a tweetstorm just becomes a bunch of connected tweets rather than forming a cohesive experience—like a bunch of loosely grouped non sequiturs.
While such a tweetstorm can succeed, you’re much more likely to find success with your tweetstorms when you can get people to read all the way through. The more people are engaged by each tweet in your thread, the more likely they are to like posts (thereby exposing them to their followers) and to retweet the thread on the basis of its quality alone. A great narrative, in this way, can help mitigate the effects of a non-enticing first tweet or amplify the effects of a great one.
In that last example from Chris, he tells the story of how eschewing profits affected Wistia—the struggles it brought the company. It is a story with a meaning, with characters to root for, and with progression—in short, everything that a story needs. Read the full tweetstorm here.
There are a few nuanced narrative tricks this tweetstorm uses that you can easily apply to your own content:
- Painting with particulars: By beginning the tweetstorm with something that happened to him and to Wistia, Chris immediately situates the story in the personal, and this makes us far more likely to get invested in the story and want to read further. Each subsequent tweet uses vivid details of things they did and tried at Wistia to keep us interested.
- Contextualizing memories: While people care about the personal, they don’t inherently care about whatever happened at your company years ago. If you want to talk about what your company did in the past, start by establishing context that helps your reader understand why they’re getting this information and how it is going to pay off.
- End on the universal: The end of your tweetstorm is your opportunity to circle back and drill home the point that you’re trying to make. Skip this part and you’re relying on your reader to draw their own conclusions.
Chris’s narrative also keys into what the University of Vermont’s Computational Story Lab has identified as one of the seven basic narratives in the English canon: specifically, the “Cinderella,” in which the protagonist experiences a rise, a fall, and then another rise.
- Rise: Wistia’s implied success in becoming profitable
- Fall: Wistia’s focus switching to growth and breaking their culture
- Rise: Wistia reinvesting in their culture and becoming profitable
Whether you’re writing a Cinderella story, an Icarus (a rise and then fall), or an Oedipal tragedy (fall, then rise, then fall again), fitting your tweetstorm to a narrative form is going to give you access to the oldest form of memetic distribution of all: the power of the tale.
🔗 Tag and Link to Increase Your Distribution Surface Area
The last point concerns an aspect of Twitter that some people utilize well but which offers another sneakily effective distribution mechanism: tagging.
Twitter has a vested interest in content getting distributed, so use that to your advantage when writing a storm, and tag people and organizations relevant to what you’re talking about.
Every tweet that you put out has a chance of being retweeted or liked. Each retweet or like puts that tweet into other people’s feeds and produces another percentage point increase to the chance of it being retweeted or liked again. In other words, the more a tweet is shared, the more likely it is to continue being shared.
One of the simplest ways to gain momentum and increase the chances your tweetstorm will be highly successful is by tagging other people who might like the tweetstorm. Better you, you can tag people who might benefit from boosting the signal you provide them.
1/ Networks built on group messaging (@telegram, @discordapp, @findyourislands, @Nextdoor, etc.) avoid a lot of the problems with social media outlined in this thread:
— Jan-Erik Asplund (@janerikasplund) March 4, 2019
- no need for more complex PoW
- old money can't hog the social capital
- no tension around the mix of users https://t.co/eLV2plqixk
This tweet of mine could have been retweeted by Wei, or any of the companies or founders tagged in that first tweet. There were a few other tags throughout the thread to increase my potential RT surface area, too.
In the end, this wound up being one of my better-performing threads thus far, and solely because one person (@gregisenberg) whose company was mentioned retweeted it.
If you can give other people something that they want to share, it’s going to increase the chances of your tweetstorm succeeding. Most of the time, if you want that to happen, you’ll have tag them first, so don’t be shy.
/thread
The benefits of being able to quickly spin up a good tweetstorm go well beyond getting followers and retweets.
Once you’re comfortable pushing tweetstorms out on a regular basis, you’ll have a reliable mechanism for using Twitter to test complex ideas, clarify your thinking, and strengthen your own skills as a writer and communicator.
Writing fortune-cookie aphorisms might be a good source of short-term satisfaction. But an intellect great enough to win the prize needs other food besides success.