The Year Without Pants: WordPress.com and the Future of Work

by Scott Berkun

Programmers are competitive by nature, and someone has to set the pace on every team, constantly demonstrating what can be done. In that first fall and winter, that lead horse was Adams. On a team that has good morale, seeing teammates launch things inspires, and inspiration brings with it new effort and ideas.

Despite millions of blogs and all of WordPress's success, there was still silliness, a humanity, that Mullenweg made sure we and our customers didn't forget. These efforts were corny, but they had positive effects that were hard to measure. Reminders to have a sense of humor gave us room for trying ideas to improve WordPress.com that'd we never pursue in a company afraid to behave in public like a person instead of a machine.

Early in 2011, Mullenweg and I discussed launching Jetpack at SXSW, one of the biggest media events in the world. Automattic rarely did product launches, but I was familiar with them and the work required to do them well. SXSW was in March and the timing just might work, we thought. My biggest concern wasn't the time line: it was the collision of Automattic's culture with a hard public deadline. We discussed it as a team and committed to deliver Jetpack in time for SXSW, which started on March 13, 2011.

In all, five warning bells went off in my mind: 1. Success demanded a different culture. If we missed the date, Automattic would look bad and Jetpack's launch would fail. I imagined Mullenweg up on stage doing magic tricks in front of a huge crowd, killing time instead of announcing Jetpack to the world. 2. Programmers would have to do estimates. I hadn't seen a single work estimate at Automattic. You can't be on time unless you estimate work. The first time that a team that has never done estimates before does them, accuracy is poor, as is how much they care about the consequences. 3. We'd need more people on the team. The rough work map we made for Jetpack implied that three programmers was not enough.1 4. Jetpack needed a simple user interface. Jetpack was a bridge for various blogging features. The technology required to do this was complex, but users' experiences had to be as simple as flipping a switch and forgetting it was there. 5. We'd need release management and compatibility testing. Web programmers work with browser compatibility all the time, but Jetpack worked on different servers. We'd need to test Jetpack on many different hosts. This meant another unfamiliar bag of tricks: bug tracking, test plans, and release candidates (RC), things rarely used at Automattic.

Mullenweg convinced Bluehost, DreamHost, GoDaddy, HostGator, Media Temple, and Network Solutions to include Jetpack in new downloads of WordPress from their services. When SXSW came along, we were ready. We had several days without new issues or new fixes needed, making for a smooth runway to unveil Jetpack to the world.

The distributed nature of the company was never a bargaining plan: the costs of sending employees to team meet-ups balanced out any financial advantages of remote work. But regardless of finances, Automatticians recognized they had more freedom over their time than the rest of the working world, perhaps the most important compensation there is. They call the rate at which people leave a company the attrition rate. It's a fantastic way to examine the health of an organization. In the eighteen months I worked there, fewer than six people left, a very low number.

In the end, every executive judges the organization by who leaves and who stays. Employees can yell and complain, but there is no louder message to management that something is wrong than forcing them to watch a great employee walk out the door.

I've mentioned before that parts of WordPress itself were not designed well, and WordPress.com inherited them. Most users found WordPress easy to use, but only after they overcame a medium-sized learning curve. WordPress was filled with layers of distracting complexity, a classic symptom of engineer-led design. Many of these layers helped WordPress's soaring rise in popularity, but that rise was predicated on appealing to programmers and organizations with technical demands, a different ambition from achieving simplicity for bloggers themselves.

In a culture that emphasized camaraderie and sharing work, it was simply easier for designers, and everyone else, to avoid tough problems like ease of use and stick to the safety of fixing bugs or adding new features, even though new features contributed to the decline of simplicity.

1. Broken windows: good and bad. It's deep in the culture that things should be fixed immediately. This is great. People take pride in keeping things working. But there is definite ADD—people tend to respond to the most recent thing, not the most important thing. Developers definitely stay busy, but I don't think we're good, as a culture, at ensuring the most important work gets done. We should have a real issue priority system (pri 1 = wp.com down, pri 2 = data loss, etc.) since without it, all issue reporting is very subjective, and we default to “fix it now.” It doesn't need to be—it'd make prioritizing for devs/teams easier.

2. Big/Ugly projects we avoid. I've noticed that one of the worst usability issues for wp.org is Media—inserting a picture is still a painful. Why hasn't it been fixed? In part, because it's a messy coding/architecture problem (so I've heard) and no one really wants to take it on. We have similar stories in Automattic (the Store has a similar rep for being a bear to work on). And important things can be broken for weeks or months when they fall off the radar (e.g. LinkedIn Plugin). I'm all for incrementalism but some projects are harder to take on this way. I bet every team has a pile of important things they've basically mothballed that we pretend isn't on the floor. [WordPress 3.5, released in December 2012, redesigned this feature.]

3. P2s have curious side effects. Generally I like P2s. But I'm fascinated by which posts generate responses and which are ignored. There's randomness to it (perhaps generated by the volume of p2 activity). It's hard to know when posting on a particular P2 who has read it and who hasn't, and whether a lack of response = implicit approval. I think some people are afraid to post on P2s since they think it's a megaphone and everyone is reading (including you / Toni / etc.). Not sure what I'd change—I...

4. Conservative ideas. I don't see many people push for big changes, big ideas or crazy thoughts—we're very tactical idea wise—many inspirations are drawn from what competitors are doing. This is ok—we're doing well—but there's not a lot of high powered, crazy ideas kicked around. Not sure why (perhaps #3 above). You seem open to big ideas, but I see it so rarely on P2s. The last big idea I saw was Lenny's post [David Lenehan] on the PollDaddy P2 about going entirely free. My only one so far was on the Writing Helpers (which I think is hugely important) and even that's ...

5. Talent, camaraderie and morale are high. I've told you before but this is the secret sauce that makes everything about Automattic work. I can't emphasize enough how critical this is—you've done very well in this regard. Most of the sy...

6. Some things are opaque. Transparency is very high but there are things no one seems to know. 1) How people are compensated 2) How people are hired (criteria for picking people for trial offers / full time offers). #1 is tricky, but #2 isn't. Leads, at minimum, should know more about the p...

7. Lack of usability methods. I don't think wordpress.com is as easy to use as we think it is. There are some quick and dirty techniques for evaluating UI, existing code as well as in mockups. I've yet to see anyone propose them or talk about them. Example: I think our dashboard is pretty hard to navigate—it's huge and confusing now. But we have no way of discovering this is a problem (it won't appear as a ticket really), or identifying where th...

The initial big ideas come from one person, which, if they are good, is fantastic for early growth. But as the company matures, the need for more people with similar courage increases.

A good sign as a leader is when output is high and meetings are short. This means all pistons are firing, there are few roadblocks, and things are on track with just a soft touch. It also means meetings can be used to stay ahead, flagging issues before they become blockers. Managers often wrap their egos around meetings, and long meetings ensure they always feel that they're the center of attention, even if the meeting is a waste of time for everyone else.

It was a combination of programmers, designers, and happiness engineers, some local, some remote, all working together to deliver on a single grand idea.

While it was hard to find pubs open as late as we'd found in Athens, the kiosks became our second home. And instead of our masochistic ouzo rituals Hugo took us out for ginjinha, a bittersweet liquor made from cherries, at the oldest hole-in-the-wall tavern in town that served it, called, predictably, Café A Ginjinha.

We all make judgments of ability at the most superficial levels. If the results are good, we give praise. If the results are poor, we criticize. We rarely give credence to the feeling in the back of our minds that the winner or loser doesn't quite fit the part. We know in our careers people who were shafted, taking the fall for incompetence that wasn't theirs, and also people who slide through organizations as if coated with Teflon, causing misery and frustration at every turn, yet they move into promotions unscathed.

A leader who shields others from things that get in the way inspires everyone to do the same. It's small habits like these that shift a culture away from the pointless exercises of finger pointing and dodging blame and toward a contagious confidence that the best work of your career is possible right now. The feeling that there's nothing in your way is something few feel often in their careers, if they ever feel it at all.

What made work at Automattic fun and challenging was how often things changed. You never knew what the week was going to throw at you. While our goals, our support from Matt and Toni, and the basic mechanics of P2s, Skype, and continuous deployment were steady, many things were in constant motion.

Often Adams and Beau were called in as experts to help the Happiness team fix a difficult problem or meet with VIP clients to teach. This was part of the way Automattic worked, and you had to roll with it. What you put in, you generally got back, if not more, as Team Social often benefited from help from the NUX, Data, Theme, and Happiness teams.

The reality check is to consider how many things you've wanted to say to people you've worked with that if they were open to it, could have helped them do better work but that you've been afraid to mention.

Creators love control over their pixels and bits and to share work before it's done is to give up all sense of control. But the value of having a designer on any project at any company is involving them early. They can try out ideas in sketches faster and cheaper than any other profession can.

1. Everyone worked on new teams formed just for the meet-up. 2. Projects were picked by Mullenweg based on suggestions. 3. Every team would have a new lead, someone who'd never led before. 4. Each team had to present to the company on the last day. 5. The goal was to ship something before the presentation.

The first day of a new team on a short deadline is fascinating anthropology. Everyone tries to figure everyone else out—who is talented, who has the same taste, who is easy or hard to work with, who has status—all at the same time they're trying to figure out the project itself. And at Automattic there's the wrinkle of everyone relearning how to work in the same physical place, where verbal persuasion returns to the mix of influences.

Of the many dreams I've had in my life, making great software is one. But the biggest, craziest, and most rewarding dream to chase has been the writing life. If what I wanted most from my time at work was to make software, I'd have stayed. But that wasn't my biggest dream. I'd already stayed longer than I'd planned.

The last act of good leaders is to ensure things go well when they're gone. Many legendary leaders failed at this: Alexander the Great, Julius Caesar, Napoleon, and nearly every monarch in history. As much good as they did in their reigns, most of it was undone by those who followed. The same ego that drives grand leaders defeats them in the end because they can't accept the notion that someone will replace them. Succession planning must be part of any long-term leader's thinking, and it has to be done now.

The most dangerous tradition we hold about work is that it must be serious and meaningless. We believe that we're paid money to compensate us for work not worthwhile on its own. People who are paid the most are often the most confused, for they know in their hearts how little meaning there is in what they do, for others and for themselves. While money provides status, status doesn't guarantee meaning. They're paid well because of how poorly work compensates their souls. Some people don't have souls, of course, but they're beyond the scope of this book. Among those with souls and high-paying but empty jobs, there's a denial of how what they seek is hard to get in the way they're trying to get it.

Meaning is personal. There is no singular meaning of life; instead there are multitudes, and they're different for everyone. Emotional words like meaning, passion, and soul are scary to people who believe everything in life hinges on pure rationality.

Humor, storytelling, and songs are social skills we developed thousands of years ago around fires while we did the critical work of staying warm and cooking food to survive. It's a shockingly recent notion that work and play should be mutually exclusive things. We learn about ourselves and each other through play, which helps us work together. Not everyone believes this, of course, but I do.