There’s a story we tell about advancing in our jobs. We start out inexperienced, kind of dumb or lucky or something. Then something happens to us and we learn from it, and then learn again, and then learn some more, and that all adds up to us being “senior.” It’s a ramp we climb over time and we only get better as we get more experienced. If we learn the right way and we’re naturally talented or whatever, the ramp is steep and goes up high. If we do the wrong stuff then we stay where we are or slide back down.

Of course that’s horseshit. Yes, learning is awesome and there are things we learn that make us better and better. I experience advancement much more as a sawtooth progression, though. Unlearning what you used to know is as or more important than learning.

Early in my life as a manager, I worked at a company that was, to say the least, chaotic. It was early in the life of the Internet industry (1996), and everything was exciting and amazing – who called us? what just happened? is that who I think it is? that’s awesome! – and the waves from all that tossed the internals of the company this way and that.

When a need would arise, people from other groups would go find some engineer they liked and ask them to do a project of unknowable length as a favor. Engineers would stay up all hours or drop the ball or get some other request, and the output of the group was seemingly random. You dropped the project for MassiveCompanyA in order to deal with a request for SomewhatLessMassiveCompanyB? Why? What were you thinking? There was always a more MassiveCompany coming in the door, and yesterday’s hotbutton was today’s backburner.

I became manager for more and more of the engineers, and soon I adopted what would be the style I used for my time there: the Bottleneck Manager. If you need something, come to me, and I’ll assign it and track it and make sure it gets done. If we can’t do it, I’ll tell you “no” and my word is final. If you talk to anyone else, I’ll come yell at you.

And it worked. Things got better. Projects shipped more reliably. Engineers were happier and felt taken care of. Everyone else disliked it but admitted it was better overall. Engineering became a black box with a single interface, and I stuck with that for almost a year, getting promoted every few months. I was miserable (I changed my email signature to “Director of Bottlenecks” after a while, and got chewed out by the CEO for that attitude), and I left the company as soon as I could. The bottleneck approach, though, was probably what that company needed at that time, and what I could provide with the skills I then had.

At other jobs in the future, I’d occasionally reach for that model in response to behavior that felt reminiscent to me. Oh, that marketing person called that engineer directly and threw their work for a loop? BRING ME MY BOTTLENECK! In different environments, without the same pressures, that model completely sucked. If I stand back and give advice to my younger self, I’d never suggest bottleneck management now. Don’t you trust your employees? Don’t you care about the needs of other groups? Don’t you care about your own time management? Isn’t more communication better? And so on.

Bottlenecks go against everything in my style today. Today, when a board member has a complaint about our Android app, my response is, “Tell the developer yourself.” And they do, and the developer loves hearing the feedback, and learns over time not to dismiss it but not to overreact to it, either – a skill every developer needs for all sorts of feedback.

I don’t think I would still be an engineering manager today at all if I’d stuck to my bottleneck approach. But I don’t think I could have been as effective, those many years ago, with another approach. I didn’t have the confidence, nor the ability to teach, nor the ability to rally people around an idea, to do anything else. I chose the approach that was right for me and right for that company and used it well, and fortunately that match worked out okay. But to keep going, I had to leave behind what worked for me there, unlearn the habits and instincts it taught me, and learn a new approach for a new place and a new situation and a new me.

I think the best thing you can do in any situation is start fresh, make yourself a novice again. Look at your style and skills in your last situation and understand them as adaptations to environmental pressures. Ask yourself what is different in the new situation. Ask yourself what worked and didn’t work last time, and name the skills and techniques you want to unlearn. Write them down as a “To Not Do” list. When you see something in the new situation, listen to your instincts for how to react, and then consider them. Are those instincts right here and now?

Don’t wait to change companies to unlearn things. Do it when you get promoted. Do it when your company changes (wow, we went from 50 to 100 in a year! Um…what should I unlearn?). Do it when new peers show up, or a new boss. Of course people get used to who you are, and retraining that within a company or role is harder than when you change companies altogether. It’s okay, though, to tell people you’re doing it. “What we’ve been doing isn’t working. I’m going to try something new. Here it is.”

Advancement is not a collection of skills. Advancement is an awesome ability to adapt to a new situation. Experience is a toolbox to help you do that – but any new job, any new role, any environmental change should make you question whether the tool you know and rely on is still the right tool for the job. Think of your skills as disposable, and actively work on unlearning the ones that were right once but aren’t right now.

[Written for a friend who was recently promoted. Congratulations, friend!]

I’ve moved my blog over to Octopress in advance of Posterous’ shutdown. This is just a test post to make sure the move worked. I’m liking Octopress a lot.

1. Work with people you like and respect who think much like you do. Disagreements are okay and can be productive, but I’ve had a better time working with people where we largely agree on the basics.
2. At the end of each day, write the people who gave you more time than usual that day and helped you as a result, and thank them for it. It’s a small thing, but it signals the right thing about other people’s time and yours. Time is valuable and you should pay for it with respect and gratitude.
3. Follow this advice. Go find people in your company you think are awesome and ask them, “How can I help you?” Focus on the happiness of people around you and you will be happier than if you focus on your own happiness. Be very generous with your time for the people you like.
4. Be willing to ask anyone for help, respectfully. Make a short, clear, reasonable request and send it to anyone on the Internet who might be able to directly help you. Many of them will say no or not respond, but a surprising number will say yes.
5. Learn how to say “no” and say it a lot. Be respectful, be considerate. But say no to things that are going to make it harder for you to do your job or lead your life.
6. Wait until you know what needs doing, then do it all the way, right away. Don’t dilly-dally and don’t be wishy-washy.
7. If you are a leader, listen closely for ideas from the people you work with, and shine a bright light on the best ideas. Give people a chance, give them your authority, give them a problem to solve and the power to solve it. Let them step up and surprise you. If you are not yet a leader, offer your ideas freely and often, and don’t be discouraged if you hear ‘no’ a lot. Push for what you think is right and when given a shot to make it right, make the most of that shot.
8. Lead a full and happy life outside of work, and never neglect it. You will be happier at work if you are happier in life. Go do that most of all.

I live in lower Manhattan, so since Thursday I’ve been consumed with Hurricane Irene and what the best way to deal with it might be. A lot has already been written on this, so I’m going to confine my thoughts to the geek’s-eye view of the weekend.

• The single most important tool that helped New Yorkers make plans this weekend was the evacuation zone map and online map search tool, posted by the Mayor’s office on Thursday. This bucketed all of New York City into four regions: Zones A, B, and C, and “No Zone” areas that were unlikely to have severe effects. I’d love to see the NY Times or someone assess the map’s eventual accuracy, but in a sense it doesn’t matter, since I fully believe it was the best information available at the time. Knowing your Zone meant you had a sense of direction. Zone A was under a mandatory evacuation order. I live in Zone B, but because I live in a basement apartment near a river, I made the decision to evacuate to a No Zone area (my friend Sarah’s apartment in Brooklyn). The New York City Mayor’s office and OEM deserve HUGE thanks for making this information available as early and as clearly as they did. Other city governments should use this as a model in similar situations. Maps save lives.
• For me, the second most important tool was Twitter. Kathryn Yu noted that Twitter makes it easy to have tweets for particular accounts sent to your phone as SMS messages, and setting that up for @NYCMayorsOffice meant I could immediately have the most critical information we needed, possibly even if power, internet, and cell services all went down (since SMS tends to be more hardy than any of those). Twitter posts quickly spread information and reacted against false information. I kept my family and friends informed about what was going on by telling them to follow me on Twitter. I got local reports from people who had stayed near the area I’d evacuated, which helped me be less stressed out about the state of my apartment. This sounds like an ad for Twitter, and I do have friends who work there, but I can say without bias that Twitter was invaluable.
• While it is a UI and community management disaster area, Wunderground, the weather geek’s site, provided a lot of information and analysis that an overconfident “if I just have the right data” geek (ahem) could use to make decisions about the weather forecasts. In case this isn’t already obvious, journalism as an industry does not provide useful planning information about impending disasters. It’s a short distance from “if it bleeds, it leads” to the conclusion that fear sells papers and boosts ratings, and every professional journalistic source I tried to use either overhyped or reacted against overhyping, both of which are useless. Wunderground isn’t in that industry and isn’t reacting against it (they certainly make fun of The Weather Channel and CNN coverage, but as outsiders, and choose their own path instead). The site was basically a constant stream of comment posts from all kinds of sources – everyone from scientists to victims of the storm to straight-up internet trolls. In my experience it most closely resembled a stock discussion forum, with “barometric pressure” substituting for “P/E ratio” and so on. But, I’m used to that kind of community, it was easy for me to spot the trolls (I think), and the people who got a lot of positive feedback on their comments often turned out to be right. Wunderground did a good job of both raising concerns early, and then reasonably tempering concerns as the storm progressed; professional journalism, as an industry generalization, did exactly the opposite.
• Because we had enough warning of what might come, Amazon was useful. While UPS made yeoman efforts to be useful too, for me they weren’t, while for others they got the job done. As news about Irene started to build, I took out my emergency kit and go bag and assessed what was missing; early Thursday I placed an overnight order to fill the gaps. Amazon shipped all but one of the items on time, but unfortunately UPS delivered them late at night on Friday, after I’d already evacuated. (I’m impressed they got the package to me at all, though, considering what was going on.) Mike O’Dea, on the other hand, ordered a portable generator and a bunch of MREs via Amazon Prime, and got them in time on Friday. If you have enough warning and can wait out a late delivery, a well-timed Amazon order can route around local supply shortages, make your life a lot easier, and make you a lot safer. (Other people reported that Fresh Direct, a local grocery delivery service, had helped them get stocked up with water and non-perishable food.)
• One thing I bought was this mobile device battery charger. As it turned out I didn’t have a need to test it, but if the power had gone out for us as it did for so many people, it could have helped a lot. I also bought this hand-crank and solar-powered radio/flashlight/cell phone charger. Since I’m a big believer that having good information is the best way to stay safe (see above), I wanted to keep SMS and radio information flowing indefinitely. A hand-crank radio is probably your best bet for that in an extended emergency.
• If you’re building an emergency kit, I highly recommend equipped.org for general information and product reviews. The author is a pilot and some of his recommendations are more suited towards wilderness survival, but he also covers preparation for urban emergencies well.
• Being cooped up all weekend with the windows rattling is pretty stressful. Entertainment is essential for stress relief. Since New York is a city of high-functioning alcoholics, nearly everyone local I follow was tweeting pictures of their preferred cocktails (Hurricanes and Dark & Stormys included, of course), and Sarah and I had a bourbon tasting. We later redesigned her blog – entertainment that would have ended quickly if power, internet, or our tolerance for TypePad and TypeKit’s dashboards had given out. The best source of entertainment by far was Stellar, which served as a highly-efficient collector of the best Irene jokes out there. Check out my Stellar feed from this weekend for a sampling (I have a few invites if you want to join).

Between the information provided by the Mayor’s office, speculation from Wunderground, and a bunch of news on Twitter, I made a decision to leave about a day before many other people had made a call either way. As a result I was at a grocery store by 7am on Friday, and got to choose whatever supplies I needed while the place was empty; many others faced empty shelves and long lines. I also had time to wrap many of our valuables in plastic and place them at the highest elevation in the apartment, and hopped on a half-full subway train Friday evening and left Manhattan. If you’re a data-driven geek, use the list above and make preparations to leave fast, and you’ll be better off than otherwise. Big crowds trying to all do the same thing at once are a huge problem in disasters; avoid those if you can.

In this case I was pretty well-prepared and got off without having any of my preparations tested at all (and I’ll happily forego A/B testing on this, please); in another case the time I gained might have made a difference for much more than just my comfort on the subway. I’m hugely thankful for all the great tools and help we had. Hopefully this list will help others make preparations in the future.

In 1999, I think right after the iMac came out in a range of colors, I happened to sit in on an internal meeting at Apple, one in a large theater filled with employees. Steve Jobs came out and the whole theater burst into applause, and the clapping went on for minutes, with people standing and cheering.  The success of the iMac was just becoming evident – the first act of Steve’s big return, leading from there to what Apple is now.

Steve let the applause go on for a little bit, then, with much effort, settled down the crowd. When things got quiet, the first thing he said was: “That’s an awful lot of applause considering that you guys are the ones who do all the work.”

Everyone leapt to their feet and applauded again for several minutes more, this time with Steve egging them on, applauding each other as a team.

That moment has since defined what I think about as leadership. I’d have to think that however wistful Steve is about leaving Apple today, there must be some part of him looking at the incredible company he’s built, and thinking to himself about today’s news, that’s a lot of applause considering that you’re the ones who do the work. He’d be right to say that, but he’s the leader, and I have always admired that more than any other quality for which he’s praised.

Take care, Steve, I wish you all the best.

I watched the clip on kottke.org today from ”God Grew Tired of Us,” a documentary about Sudanese refugees coming to the United States. There have been a few documentaries on this topic in the past few years, perhaps inspired by this great NY Times Magazine article from 2001.

In that article was a moment that has stuck with me for ten years, from a visit newly-arrived refugees take to a large grocery store:

My friend Jon made one of the documentaries about the refugees, Lost Boys of Sudan. Highly recommended.

A number of people wrote me about my ”Why Wesabe Lost to Mint” post to ask about this line: “…nor do I agree with those who obsess over failures for years after (as I have done in the past).” A few said they were currently agonizing over a past failure and asked how I had gotten out of that state. Here’s what I wrote to one of them:

For whatever reason, I haven’t ever gotten to that state of obsessing over Wesabe’s failures. I mostly just moved on and started thinking about what I wanted to take away from the experience (positive and negative) and what I want to do next. Maybe now I have confidence that shipping something new will work.

Thanks, by the way, to everyone who responded to that last post. I was overwhelmed by the response and I’m glad if it helped others.

A number of people have asked and speculated about why the company I co-founded, Wesabe, shut down earlier this summer. Some of the claims or guesses about it are just factually wrong; others seem misinformed to me; others seem to have some truth. I thought I’d add my own opinion.

In November 2006, Wesabe launched as a site to help people manage their personal finances. We certainly weren’t the first to try to tackle this problem through a web app, but we were the first of a new wave of companies that came out in the months that followed, characterized but what some would call a Web 2.0 approach to the problem. Like Flickr and del.icio.us, we relied on community and features such as tags; unlike some of the previous attempts, we tried to automatically aggregate and store all of our users’ financial accounts on the web (instead of relying on manual data entry, say, or desktop storage of the data); and most especially, we tried to learn from the accumulated data our users uploaded, and make recommendations for better financial decisions based on that data. If every copy of Quicken started out as a blank spreadsheet, Wesabe tried to accumulate knowledge from users and data that would fill in some of that spreadsheet for you, and point you in the direction of better choices.

Even before we launched, we heard about other people working on similar ideas, and a slew of companies soon launched in our wake. None of them really seemed to get very far, though, and we were considered the leader in online personal finance until September 2007, when Mint launched at, and won, the first TechCrunch 40 conference. From that point forward we were considered in second place at best, and they overshadowed our site and everyone else’s, too. Two years later, Mint was acquired by Intuit, makers of Quicken (and after Mint’s launch, the makers of Quicken Online) for $170 million. A bit less than a year later, Wesabe shut down.

That’s the history; now, some interpretation, with the completely obvious caveat that I am anything but an unbiased observer. I was the first person to start work on Wesabe and formally co-founded the company (as Chief Product Officer) with my high school friend, Jason Knight (who was CEO). I made nearly all of the product decisions myself, and was notorious with Jason and our board as being very hard-headed about those decisions. While I relied on many other people in making product choices, I also hired and managed all of those people, so that group was at the least a reflection of who I thought had the right values and ideas. When Jason had to leave the company due to a family illness, I took over as CEO and led the company without a formal peer for the final two years. All that adds up to me having absolutely no one to blame for Wesabe’s failure but myself, and as a result I can’t now nor could ever be dispassionate in thinking about what happened.

With that in mind, here are what I believe are a number of myths about why Mint won and Wesabe lost:

1. Mint launched first - I hear this surprisingly often; they didn’t. Wesabe launched about 10 months before Mint. More the shame that we didn’t capitalize on that early lead. There’s a lot to be said for not rushing to market, and learning from the mistakes the first entrants make. Shipping a “minimum viable product” immediately and learning from the market directly makes good sense to me, but engaging with and supporting users is anything but free. Observation can be cheaper. Mint (and some others) did well by seeing where we screwed up, and waiting to launch until they had a better approach. 
2. Wesabe never made any money - untrue. We started generating revenue in late 2008, a year and a half before we shut down; and ran completely out of invested funds almost nine months before closing the company. (Put another way: we survived solely on revenue from mid-November 2009 until the end of July of this year, when we turned off the site.) Obviously we didn’t make enough to keep us going indefinitely, but we were not far off from supporting a small company indefinitely on revenue. 
3. Mint was a better name and had a better design - both of these things are true, but I don’t believe they were primary causes for our company to fail and for Mint to be acquired. Mint’s CEO likes to talk about how ridiculous our name was relative to theirs, but I think the examples of Amazon, Yahoo, eBay, Google, and plenty of others make it plain that even ludicrous names (as all of those were thought to be when the companies launched) can go on to be great brands. Mint’s design, while definitely very appealing and definitely a factor in getting people to trust the company, doesn’t seem to me to be enough to explain the different outcomes, again based on what I’ve seen from other companies (del.icio.us versus Magnolia, eBay versus Amazon Auctions, and now iPhone slipping against Android). Design matters a huge amount, without question, and Mint’s design was exceptional, but if other, stronger forms of lock-in are in place first, design alone can’t win a market, nor can it keep a market. 
4. Wesabe wasn’t viral and Mint was - half-true: neither product was. Neither of us bore any resemblance to a typical Silicon Valley success story, with traffic surging up and to the right (YouTube, Twitter, etc). Mint aggressively acquired users by paying for search engine marketing (reportedly spending over $1 for each user), while Wesabe spent almost nothing on marketing; yet in the end we grew at about 1/5th the rate they did. Take a look at Mint’s Compete.com graph today and you’ll see that their traffic dropped substantially for the six months after their acquisition, and has had sawtooth traffic after that. That pattern matches with a decreased and then increased marketing budget, but not with viral growth. Our patterns similarly followed non-scalable curves (influenced primarily by press wins, economic conditions, and sometimes drafting on Mint’s coverage). 

Those are what I see as the common beliefs about the outcome of our competition with Mint that I don’t believe were right. Why was there such a difference, then?

I see two primary reasons that, if they had changed before Mint’s launch or had never occurred in the first place, could well have allowed Wesabe to maintain its early lead and remain the leader against Mint’s entry.

First, we chose not to work with Yodlee, but failed to find or make a replacement for them (until too late). Yodlee is a company that provides automatic financial data aggregation as a web service. They screen-scrape bank web sites (that is, read the payee and amount and date by parsing them out of the bank’s web site, writing a custom parser for each bank they support). When we talked to Yodlee in 2006, the company was crumbling, having failed to get acquired and losing executives. They were also very aggressive in negotiation, telling us they would give us six months’ service nearly free and then tell us the final price we’d be charged going forward. Since they had effectively no competitors, we didn’t believe we should tie our company to a single-source provider, especially one in very bad business shape. Mint used Yodlee (at least until they were acquired - I’m not sure what they’re doing now) to automatically get user’s data from bank sites and import them into Mint, and as a result had a much easier user experience getting bank data imported. Wesabe built our own data acquisition system, first using downloadable client programs (partially because that was easier and partially to preserve users’ privacy) and later using a Yodlee-like web interface, but the Yodlee-like version didn’t launch until six months after Mint went live, and even then didn’t really work as a near-complete replacement for some time after.

A good friend argues that our mistake was not using Yodlee in the first place, and maybe – probably – it was. I believe, though, that we could have made that choice as long as we immediately assumed that someone else would eventually sign up with Yodlee, and that we had to be at least as good if not better than what Yodlee provided, however we got there. (PageOnce, for instance, has not used Yodlee, but has grown very significantly in the same time, using a combination of other aggregation methods that were more effective than ours.) Mint’s dependence on Yodlee apparently suppressed their acquisition interest among companies that knew Yodlee well (such as Microsoft, Yahoo, and Google); since we had developed our own technology for aggregation, we didn’t have that particular problem, and in fact had some acquisition interest simply for the aggregator we’d built. We just didn’t build it nearly fast enough. That one mistake (not using or replacing Yodlee before Mint had a chance to launch on Yodlee) was probably enough to kill Wesabe alone.

Second, Mint focused on making the user do almost no work at all, by automatically editing and categorizing their data, reducing the number of fields in their signup form, and giving them immediate gratification as soon as they possibly could; we completely sucked at all of that. Instead, I prioritized trying to build tools that would eventually help people change their financial behavior for the better, which I believed required people to more closely work with and understand their data. My goals may have been (okay, were) noble, but in the end we didn’t help the people I wanted to since the product failed. I was focused on trying to make the usability of editing data as easy and functional as it could be; Mint was focused on making it so you never had to do that at all. Their approach completely kicked our approach’s ass. (To be defensive for just a moment, their data accuracy – how well they automatically edited – was really low, and anyone who looked deeply into their data at Mint, especially in the beginning, was shocked at how inaccurate it was. The point, though, is hardly anyone seems to have looked.)

Between the worse data aggregation method and the much higher amount of work Wesabe made you do, it was far easier to have a good experience on Mint, and that good experience came far more quickly. Everything I’ve mentioned – not being dependent on a single source provider, preserving users’ privacy, helping users actually make positive change in their financial lives – all of those things are great, rational reasons to pursue what we pursued. But none of them matter if the product is harder to use, since most people simply won’t care enough or get enough benefit from long-term features if a shorter-term alternative is available.

I am, of course, enormously sad that Wesabe lost and the company closed. I don’t agree with those who say you should learn from your successes and mostly ignore your failures; nor do I agree with those who obsess over failures for years after (as I have done in the past). I’m hoping that by writing this all out I can offload it from my head and hopefully help inform other people who try to start companies in the future.

You’ll hear a lot about why company A won and company B lost in any market, and in my experience, a lot of the theories thrown about – even or especially by the participants – are utter crap. A domain name doesn’t win you a market; launching second or fifth or tenth doesn’t lose you a market. You can’t blame your competitors or your board or the lack of or excess of investment.  Focus on what really matters: making users happy with your product as quickly as you can, and helping them as much as you can after that.  If you do those better than anyone else out there you’ll win.

I think in this case, Mint totally won at the first (making users happy quickly), and we both totally failed at the second (actually helping people).  No one, in my view, solved the financial problems of consumers.  No one even got close.  Yes, both products helped some people – ours mostly through a supportive community and theirs mostly through giving people a rough picture of where their money has gone. But when we analyzed the benefits we saw for our users, and when Mint boasted about the benefits they saw for their users, the debt reduction and savings increase numbers directly matched the national averages.  Because our products existed during a deep financial crisis, consumers everywhere cut back, saved more, and tried to reduce their debt. Neither product had any significant impact beyond what the overall economy led people to do anyways.

So, yeah. Changing people’s behavior is really hard. No one in this market succeeded at doing so – there is no Google nor Amazon of personal finance. Can you succeed where we failed? Please do – the problems are absolutely huge and the help consumers have is absolutely abysmal. Learn from the above and go help people (after making them immediately happy, first).

-M

I wrote this to a coworker just now, and it feels like something everyone should agree with me about. Please be persuaded:

I don’t use Google Calendar nor any shared calendar - but I do receive your invitations and despite missing [a recent meeting] I am usually good at responding to meeting requests. I saw your request just now and am happy to attend that meeting.

I don’t like the idea of shared calendars at all. If the calendar says I’m free and you put a meeting in it, does that really mean I’m free? Usually that assumption leads to conflicts and missed meetings. A blank space on my calendar might mean that I am using that time to write or to talk to people individually. Blank != available for me, and I think that is good. Also, I hate meetings, so I want it to be a little harder to get meetings set up. “Easy to create a meeting” is a bug, not a feature, in my opinion. :)

In my opinion there is no substitute for calling or emailing and saying, “Hey, can you join us for this?”

But - I will be sure I send you replies to your meeting requests promptly. I am not trying to prevent you from getting your job done; I am trying to make sure I spend the least amount of time in meetings possible, and that when I am scheduled for a meeting I have really confirmed that I am available.

I should say, I feel this way at work, not at home. My wife and I use a shared Backpack calendar and it’s been a huge help. We use it for social planning, appointments, reminders, and important dates. I bet that shared calendars could work well with any group that is small and shares a very high level of trust and respect. I just usually don’t find those are the prerequisites when companies start using shared calendars. Maybe the lack of those prerequisites is the the real problem at work – in this and other areas.

-M