The Trenderprise

Software, so hot right now.

Why does everyone hate Jira?

There are lots of answers. Because the interface is clunky and slow and makes people angry. Because they need to watch a few hours of tutorials to be able to understand this piece of crap. Because it’s nothing more than a collection of hacks. Because it’s full of mountains of feature bloat that get in the way. Because it requires a 747 cockpit worth of buttons and links just to do something that should take two seconds. Because it is a blight on humanity.

Ok, but why do people use Jira? They would rather use Trello. They would rather use a text editor. They would rather stab themselves in the eyes. They would rather peel off their own skin and bathe their weeping raw flesh in a bath of vinegar.

Alas. They still use Jira, because these people—the people who hate Jira—aren’t the people who buy Jira. Management, who doesn’t have to actually resolve tickets in Jira, buys it, forces everyone else to adopt it, and nobody can use anything else. If you have to use Jira, you hate it. If you don’t have to use Jira, you buy it.

For decades, this was the paradox of enterprise software: The buyers and users were often different people. IT departments bought software for everyone else, and the things that IT wants—options to customize and control how software gets used, long lists of security certifications, features that check the boxes on their RFPs, to be invited to fancy dinners at Gartner conferences—are not the things that people who regularly use Jira want. So Jira, like nearly every other enterprise software product of its generation, wasn’t built just for its users; it was also built for its buyers. And as long as those buyers are happy, it doesn’t matter if people who use Jira hate it, if they want to claw their eyes out when they open it, if it makes them pull out their hair, or if it makes their souls die. Because they are trapped in the belly of the horrible machine, and even if the machine is bleeding to death, it will not let them out.

But about fifteen years ago, the trend reversed. Startups began marketing their products directly to Jira’s condemned users. Jira is bad for you, they said. It’s what makes you hate your job. Its loading spinner isn’t just a five-second delay; it is a corporate Pavlovian trigger. It is your miserable Monday morning. It is your incompetent boss. It is your office’s humming fluorescent lights, the chemical stink of Melissa’s cheap perfume, the whine of Alex’s dumb laugh, the nauseating air in the third floor kitchen. Don’t buy Jira, which was made for Gilbert Huph, and ruins your incredible potential. Buy Asana, which was made for you:

“I know enterprise isn’t usually your beat,” [Asana founder Justin] Rosenstein remarked when I mentioned that I had heard about and was interested in Asana’s popularity, “But I’m guessing that’s related to the fact that most enterprise services are just bad products, they’re made to appease the CIO instead of the end-user… I’d like to think one of the biggest things we’re doing differently is that we come from a deep product user-experience-above-all-else background.”

For companies like Asana, the pitch was product delight, and the sales motion was subversion—or, in some cases, light blackmail. These startups¹ encouraged people to sign up on their own, without asking IT for permission first. Then, once people started to build habits around it, Asana’s sales team would call up the IT department, and make them an offer they couldn’t refuse. Asana, and its generation of enterprise software, didn’t want to force people to use bad software by selling to IT; they sold good software to people, and forced IT to buy it.

The whole thing had a name: The consumerization of IT.² Consumer technology products like Snapchat and Netflix and Apple AirPods are typically bought by the people who use them, so apps that are slow, confusing, or missing important features won’t get bought. The better product wins, not the one with the best sales mechanics or the longest list of compliance policies.³

There was something healthy about this, and something final. Products being bought and sold by backroom negotiation is a form of market failure. Products being chosen on their own merits is pure and efficient. It is the invisible hand exerting its will; it is the market reaching its unavoidable equilibrium.

Or so it seems.

Fame and fortune

Last week, I mentioned an old survey I ran in which I asked people which type of founder they would invest in, between an industry veteran, a ruthless operator, a talented prodigy, someone who is decisive and fast, or a data expert. Though experience and analytical expertise won, I may have left out the best option: A celebrity.

Because, in a world of social media, copycat trends, and flattened global fame, this is how we buy stuff now. We used to buy roughly commoditized products—clothes, home goods, food, liquor, whatever—based on long-built reputations of product quality and reliability; then, we bought them because brands hired celebrities to pitch us their products in TV commercials; and now, we buy them because influencers make them. Emma Chamberlain doesn’t promote Warby Parker glasses; she designs them. We buy Kylie’s cosmetics, Mr. Beast’s chocolate bars, Logan Paul’s energy drinks, Rihanna’s beauty products, 63 different celebrities’ tequilas and bourbons and gins, and Gwenth Paltrow’s goop. Though quality still matters—Dave Portnoy found the floor—it’s hardly the only thing that matters, and probably not even the most important thing.⁴ What matters is winning a global popularity contest.

Nothing illustrates this better than the Stanley car fire. About a year ago, a woman’s car caught on fire. She posted a TikTok of her pulling a charred Stanley tumbler out of its melted skeleton—because every emergency is an opportunity for some good content, I guess—and shook it. “There was fire yesterday,” she said, “and it still has ice in it.”⁵ The video went viral, Stanley saw it, and offered to replace her car in a response video. She got a car, Stanley won some marketing awards, and people bought 750 million dollars worth of Stanley cups.⁶

Ostensibly, the original video showed how good Stanley’s tumblers are—and how, according to their response video, they are “built for life.” But that’s obviously not why people bought them. They bought them because Stanleys were cool, because surviving a fire was cool, and because giving a woman a car was cool. Nobody wants their Stanley for life. They want to buy 15 of them, and use them all once. They want their Stanley for a season, and then they want it to biodegrade into a 20-year-old digital camera.

Of course, none of this is entirely new—trends have existed forever.⁷ But that’s the point. This is the natural evolution of how categories of things are bought and sold. First, by salesmen; then, by product reputation; and finally, by fashion and fad.

And why would software markets be different? If IT was consumerized, why wouldn’t it get memeified next?

Fad not function

In some ways, it’s already happening. Silicon Valley has already rotated past product quality alone, and is entering an era of “taste:”

In a world of scarcity, we treasure tools. In a world of abundance, we treasure taste. The barriers to entry are low, competition is fierce, and so much of the focus has shifted — from tech to distribution, and now, to something else too: taste.

Taste is eating software. Taste is the new weapon.

Whether expressed via product design, brand, or user experience, taste now defines how a product is perceived and felt as well as how it is adopted, i.e. distributed — whether it’s software or hardware or both.

Fair, though taste seems like too substantive of a term. There’s a nobility to it; it has echoes of craft and quality; it is suggestive of the opinions of the discerning.

I’m more cynical. We do not like things from first principles. We like them because other people like them. Our desires are memetic; our social proof is gossip; our interests are compelled. Is ChatGPT the leading enterprise chatbot because it’s the best product, because it has the most agreeable aesthetic, or because it’s the trendy chatbot? Or, more pointedly, if you were funding an OpenAI competitor, who would you bet on to dethrone them? A marginally better model, or a celebrity?

The market seems to prefer the latter. Bret Taylor’s AI startup Sierra was valued at a billion dollars within a year of its founding, and 4.5 billion dollars nine months later. OpenAI cofounder Ilya Sutskever raised a billion dollars for Safe Superintelligence before his company had a dozen employees and before its website used JavaScript. And Elon Musk raised 12 billion dollars for his AI hobby, because he’s the President of the United States.

The whiplash between AI coding assistants has followed a similar pattern. GitHub’s Copilot was the original leader. Then Devin engineered a viral launch that made them the model of the moment—until they were upended by Cursor, so hot right now, Cursor.

Conceivably, this hot potato-ing of attention is a reflection of the dynamic AI market in which these products are operating. Demand is vertical; technology is changing; with every new state-of-the-art release, a new leader rockets to the top.

Perhaps. But that seems like an overly wholesome explanation. Engineers are just as susceptible to social pressures and popularity contests as teenagers on Twitch. Nobody gets fired for buying IBM, nobody gets made fun of for wearing a JanSport, and those two things are related. Because, for as much as Silicon Valley celebrates the builders, our economy is content and our opinions come from others' upvotes. And in that collapsed context, attention is all you need.

---

FTP over Feastables

Mr. Beast’s chocolate bars are called Feastables. As far as I can tell, that pun is the only reason Mr. Beast chose to make chocolate bars instead of sandals, or soda, or collectable figurines. He was famous, he needed something to sell, chocolate bars are all kind of the same, and feast rhymes with beast. So he sold chocolate.

What else is all kind of the same? To-do apps. Email clients. Enterprise applicant tracking systems. (BI tools.) Someone releases a new one on Product Hunt every week, attempts to astroturf some buzz, and then slogs away on LinkedIn for a decade, trying to get anyone to pay attention to them.

So…if you’re a celebrity, why not start a software company? Steph Curry has 58 million followers on Instagram. How is some 25-year-old PM from Ramp supposed to compete with that? Who would you rather work for, the greatest shooter in basketball history, or Jeff? Whose brand would you rather create, the one that weaves together stories about parties with LeBron James and Kendrick Lamar lyrics, or the one that was inspired by Jeff’s frustrations from managing a remote team? Whose email provider would you rather use, yet another Superhuman close—dark mode! Keyboard shortcuts! AI summaries!—or the one that half the NBA uses?

Media companies are dying. Restaurants are low-margin. Don’t just invest in tech companies, Steph. Make a ChefTP, the leading FTP server for Silicon Valley, and claim you rightful place as the true king of San Francisco.

1. Which included Yammer, where I worked for 18 months.

2. From that 2012 TechCrunch article about Asana: “One might say that Asana is the poster-child for the consumerization of the IT industry.”

3. Although, there is the occasional enterprisification of consumer products, in which people buy stuff because it’s bundled together with other services. A lot of people probably bought Apple Music because Apple gave them a free subscription with their iPhone. This can be tricky though: Bundle too many consumer products, and you’ll probably get sued.

4. Case in point: Dupes and knockoffs are sometimes more popular than the real thing.

5. I mean, that is impressive.

6. This was in 2023, and the fire was towards the end of 2023, so it’s not clear how much the fire contributed to that number. But searches for “stanley for sale” hit a peak the week of the story, and Stanley kept growing for another six to nine months after the fire. So it clearly helped.

7. We can manufacture them much faster now though. You can be a nobody one week, make a viral TikTok the next, and create a memecoin to directly monetize your fame in the third. Influence and popularity are fundamentally the same, but they have all collapsed into a single hurricane of noise, and a single marketplace for attention.