
Trello: the boss who couldn't see the work
Joel Spolsky couldn't tell what his own team was working on. A list of five things didn't fix it. A wall of sticky notes did, and then the team refused to let it become a tool only programmers could use.
By Matt Wagner7 min read
Trello cofounder Michael Pryor, then its CEO, on stage at TechCrunch Disrupt NY in 2015. Photo: Noam Galai/Getty Images for TechCrunch, CC BY 2.0
In this story8 sections
Most people know Trello as the board where teams drag cards from "To Do" to "Done." Fewer know it began with a boss who couldn't see what his own team was doing, and a first fix that didn't work. This is how a wall of sticky notes became a product used far beyond software, why it now belongs to Atlassian, and what that history tells you if you are deciding whether Trello fits your team.
2010
The boss couldn't see the work
In 2000, Joel Spolsky and Michael Pryor started Fog Creek Software. Around the company's tenth birthday, Joel wanted new things for his programmers to build, and he realized Fog Creek "could easily afford to make four little two-person teams to launch four new products."
He also had a problem of his own. "After ten years in management I still never knew what anyone was supposed to be working on," he wrote. Once in a while he walked around asking people what they were doing, and half the time he couldn't see why they were working on it at all.
“I still never knew what anyone was supposed to be working on.”
So one of the small teams took on his problem: keeping track of who was working on what. It had to be "super simple and friction-free so that everyone would use it, but it had to be powerful, too."
See how Trello fits your team
Tell Otis what your team needs and Juukbox lists matching products in order of best fit.
2010 to 2011
First try: a list of only five things
The first idea was called Five Things. Everyone would have a list of exactly five things they were allowed to work on, and nobody could ever have six. The reason was simple: "If you have too many things on your to-do list, your motivation tends to sag."
It was a hard cap on how much work one person could hold at once. It didn't stick. "Five Things wasn't the right idea," Joel wrote, "but it led us to the idea that became Trello."
2011
The answer was already on the wall
The better idea was hanging in plain sight. At any software company, Michael said, "you would look where the developers were and they would all have a whiteboard and it would have Post-It notes on it and they would be in columns." Teams moved the notes across the board as the work got done. It was "very familiar to developers," he said, "but not really known outside of that."
So the team put that wall on a screen, for everyone. In January 2011 the prototype was pitched inside Fog Creek under a code name, Trellis, because the software "gives structure to what you're doing." They couldn't get the domain. According to Trello's own history, they also narrowly avoided names like Cardvark and Planatee before going with Trello.
It started as an HTML mockup that the two-person design team put together in a week. Soon four programmers and two summer interns were building it, and they began running their own work on it when it was only 700 lines of code. "Even in that super-simple form," Joel wrote, "we found it incredibly useful."
A board, lists and cards
Trello has stayed close to that wall ever since. A board is one project, like a product launch or an office party. Lists are the steps, as simple as To Do, Doing and Done. Cards are the pieces of work, and you drag each card from list to list as it moves along. Open a card and it holds the details: who it's assigned to, the due date, the files.
Michael believed the picture did the teaching. "You don't have to learn a vocabulary when you get there," he said. He was careful about one word, too: "the cards in Trello are cards. They're not tasks."
“You don't have to learn a vocabulary when you get there.”
Even the dragging was a choice. Picking from a drop-down menu would be quicker, Michael admitted, "but the tactile nature is important." And people tended not to archive finished cards. They moved them to a done list instead. "It's almost like a trophy wall," he said, "to show you progress."
Their own fans wanted more. They said no.
Fog Creek's audience was programmers, and Michael knew they would ask for "swim lanes and Gantt charts," the heavy planning tools software teams love. The team said no. "When you're building a tool for the mass market, those things make no sense," he said.
“When you're building a tool for the mass market, those things make no sense.”
Joel put it another way in 2012. Some people looked at Trello and saw a tool for building software. "Yeah, it's that," he wrote, "but it's also for planning a wedding, for making a list of potential vacation spots to share with your family, for keeping track of applicants to open job positions, and for a billion other things."
The line held for years. In 2020 Michael was still warning about tools that keep adding features for power users, until "those people that were using it, you've now turned those people off."
That refusal is the most important thing to know about Trello. It is why anyone can read a board in seconds. It is also why teams that plan in fine detail sometimes want more.
September 2011
Launch week nearly broke it
After a closed beta, Trello launched at TechCrunch Disrupt in September 2011, on the web and the iPhone. The big stage brought less press than you might think. Disrupt had about 200 journalists, Joel wrote, and Trello got "four stories," every one from a reporter he already knew.
The users came anyway, faster than the servers were ready for. Trello shows changes to everyone at once: when a teammate moves a card, you see it move. Under the launch crowd, the part that did that "started behaving very strangely." The team switched to a plainer, slower method that keeps asking the server for changes, and Trello kept working as it grew "from 300 to 50,000 users in under a week." By January 2012 they were back on the faster system.
2012 to 2017
Then it spread far beyond software
Trello was free to start, on purpose. "Our entire marketing budget was just making it free," Michael said. "The hypothesis was that it would spread throughout an organization."
It did. By the summer of 2012 Trello had 500,000 members, and Joel's dog Taco had become its official spokes-husky. People found uses nobody planned. Michael once came across "a long article with 18 screenshots of how to use Trello to plan a wedding," written for a wedding blog.
In July 2014 Trello split off from Fog Creek as its own company, Trello, Inc., with Michael, who had cofounded Trello with Joel, as CEO. Looking back that year, Joel listed what had surprised him, starting with how well it worked for people outside software. "We've successfully made a non-developer product that actually appeals to civilians," he wrote. "The board / card metaphor makes every board instantly understandable."
2017 to today
Now Trello belongs to Atlassian
On January 9, 2017, Atlassian, the company that makes Jira and Confluence, announced it would buy Trello, in a deal it valued at about $425 million. By then Trello had more than 19 million registered users, nearly double the year before, and Atlassian said more than half of them worked outside technical jobs. Atlassian said Trello would "remain as a standalone service," and Trello's own history says the deal closed in early 2017.
Atlassian's cofounder Mike Cannon-Brookes explained the deal in words any buyer can use: "There is no silver bullet for teamwork. Teams have diverse ways of working and require a broad set of tools that adapt to their specific needs."
“There is no silver bullet for teamwork.”
Trello is still sold on its own, with its own site and its own plans, and its pricing page says it plainly: "Trello is an Atlassian product." The top plan, Enterprise, includes Atlassian Guard Standard, Atlassian's security and sign-in service, which a company can turn on across all its Atlassian products.
Joel's question was who is working on what. Today, one look at a board answers it.
The tool in this story
More stories
Questions to ask before choosing Trello
Does your work move in steps?
Trello works best when tasks move step by step from list to list. It's simple on purpose, so if you plan around linked deadlines or team workloads, test those first. Calendar, Timeline, Table, Dashboard and Map views need Premium, at $10 a person a month billed yearly, or $12.50 billed monthly.
Do you need to see work across many projects at once?
Each board is one project. Trello's Workspace views put cards from many boards into one table or calendar, filtered by label, list or member, and they come only with Premium or Enterprise. If your team runs many projects side by side, budget for Premium.
How many people will share it?
The Free plan covers up to 10 collaborators and up to 10 boards per Workspace. If a free Workspace goes over 10 collaborators, its boards turn view-only and its automations stop until you upgrade. Standard removes both limits at $5 a person a month billed yearly ($6 billed monthly).
See how Trello fits your team
Tell Otis about your team and Juukbox builds a space with matching products in order of best fit, Trello included.
Trello isn't a sponsor of Juukbox and didn't pay for this story.
Sources
- Announcing Trello, Joel on Software (Joel Spolsky)
- How Trello is different, Joel on Software (Joel Spolsky)
- Trello, Inc., Joel on Software (Joel Spolsky)
- Should you launch at a conference?, Joel on Software (Joel Spolsky)
- About Trello, Trello
- Trello pricing, Trello
- Trello 101, Trello
- Create a board, Trello
- Trello tour, Trello
- The Trello tech stack, Atlassian
- Atlassian to acquire Trello (press release, January 9, 2017), Atlassian via the SEC
- Collaborator limit for free Workspaces, Atlassian Support
- Workspace views, Atlassian Support
- Interview with Michael Pryor, Mixergy
- Trello's CEO on the Surprising Benefit Cloud-Based Tools Bring, Entrepreneur
- Podcast interview with Michael Pryor, Intercom
- Interview with Michael Pryor, The Register
- Cover photo: TechCrunch Disrupt NY 2015, Day 3, Wikimedia Commons




