Monthly Archives: August 2009

Open City Data

Yesterday I was part of a press conference by Mayor Gavin Newsom promoting, which is one of San Francisco’s first steps at opening up. Tim O’Reilly also spoke and made the point to me afterward that as he dives deep into every part of the intersection of technology and government he’s most excited about the prospect for change at the city level. Here are some pictures from the event. I think we’ll see more along these lines, and more WordPress, for San Francisco in the future.

Starting a Bank

I often get asked something along the line of, “If you weren’t leading Automattic, what would you work on?” There’s not a single answer to this question; the answer changes day to day. But I think if you asked me today, I’d say I would like to start a bank.

There are very few people who really love their bank. We’ve all dealt with overage fees that stack up, brain-dead fine print, and a general malaise. There’s also a unique opportunity in that mainstream contempt for financial institutions has never been higher, while at the same time there is an incredible amount of government backing that essentially makes it a no-risk environment. People are hungry just for anything different, something contrarian. A David to the Goliath banking industry.

The name of my bank would be something supremely boring, like SafeBank. The idea behind it is that bad behaviour in the banking world has been largely inevitable because their compensation structures incented people to do overly risky things. SafeBank would maintain a reserve level 2-3x higher than Fed requirements and any other bank. SafeBank would have no bonuses. Critics would say this would make it impossible to attract top-shelf talent. Every time the bank gets attacked we’d turn it into an advertising opportunity to emphasize why we’re different. “We can’t attract top-shelf talent? We take your money and put it in a vault. We don’t need the million-dollar bonus geniuses on Wall Street to do that. SafeBank. Bank, safe.”

In fact, the first few years of SafeBank would be largely focused on acquisition through every trick in the book. At the very beginning pull a Gmail/ and make it invite-only, which will create a buzz and also allow you to give amazing white-glove service to the initial customers, who will in turn tell their friends and make a ton more buzz. (You can also target certain profitable segments and ultra-safe depositors at first, like Gmail users in San Francisco (using Firefox with an ad-blocker) who make six figures a year.) There would be only one style of checks and debit cards and they’d need a distinctive design so if you saw one you’d say, “What’s that?” which would then start the whole conversation again about how SafeBank is different.

For the first two years you could also do things like not allow accounts larger than the FDIC-insured limit. No one has ever heard of a bank turning away money! But you’d say that although everything SafeBank does is risk-free, it’s still a startup and if people have more than the insured limit (250k for single, 500k for couples) in an account, they should put the extra somewhere else. Again, this will impact a very low percentage of customers, but everyone will think it’s remarkable. This can be phased out after a few years; in fact, it would be a PR opportunity. “We’ve been in business now long enough that we feel comfortable with larger accounts.” Boom, free coverage.

I’m a tech guy so of course a lot of focus would be on the website. Imagine an old-time vintage design aesthetic combined with a Google-like simplicity and attention to speed. All logins would be two-factor, with the default being it’d SMS you a one-time code to log in when you gave your email address. A big part of the website would be the blog, of course. It would have a strong Ben Franklin-like common sense voice, and in addition to giving a few cool saving or home tips each week, it would cover at least one financial industry story a day.

  • “Bank of America spent $40,000,000 dollars on airplanes last year. We spent $40,000 to develop an iPhone application so you can check your balance from anywhere.”  (Hmm, the iPhone app should cost like $2.99.)
  • “Here’s how to block advertising when you browse the web with Firefox; it makes the web faster and less annoying.”
  • “Goldman Sachs just paid out 16 billion dollars in bonuses to their employees. If we had an extra 16 billion dollars lying around, we’d put it in the bank for a rainy day. (If Goldman had never paid out bonuses they never would have needed government intervention.)”
  • “So-and-so Bank’s website requires you to use Internet Explorer. We beg that you don’t, because there are way cooler and faster browsers. Here are 3 open source browsers you can switch to today.”
  • “68 Million Reasons Your Bank Sucks. That’s the amount BoA collected last quarter in needless ATM fees.”

(That’s all made up.) The headlines would almost write themselves, and every time a financial institution is in the news it’d be an opportunity to contrast why SafeBank is different and what the underlying philosophy is behind why it’s different.

All of the marketing would be on the web and viral, because it’d be an online-only bank like ING Direct. No storefronts where people have to wait in line or risk a bad interaction with a teller, or that get robbed and need insurance; basically a lot of the historical risk of running a bank could be eliminated. When you sign up it would have a “tell your friends about SafeBank” address book feature that would connect you to them if they signed up for an account, give you both money (Bank of America has something like this), and also make it easy to send them money, PayPal-style, if they have an account.

How would the bank make money? I think it wouldn’t touch anything risky on the financial side — it would be a data company. The first 3 years the focus would be entirely on customer acquisition, marketing, PR, and building a world-class tech team building a rock solid infrastructure. SafeBank would make way, way less money than banks currently do, but it would be more than enough to build an amazing product in a sustainable way, like Craigslist did with newspaper classifieds. After a certain milestone, say 100 billion in deposits, I would buy or clone Mint. SafeBank would have more (and more accurate) data about its customers than almost any other company in the world other than credit card companies, so the online interface would have Mint-like lead generation offers based on that information. For example, you spend $140 a month on electricity, but if you switch to this new solar provider you’d save $200 a year. Think of it like Gmail contextual advertising but based on where you spend your money rather than the words in an email. There also might be aggregate data opportunities for economic research or targeting, but I’m not sure if I like the privacy implications there.

SafeBank couldn’t raise VC or anything like that because having any sort of exit expectations would completely kill the safety story, but I think it could be bootstrapped and after a few years would be hugely profitable. Its existence would also put huge pressure on existing banks because depositors would be leaving in droves, putting pressure on their reserve requirements. Existing banks couldn’t compete in a traditional way because they have such a sordid history of customer apathy and bad PR. SafeBank wouldn’t be trying to capture their profits, it would largely be destroying them and making much smaller amounts of money in non-traditional bank ways. It would be somewhat like a credit union, but for the masses.

Anyway, this is just how my mind wandered this morning while brushing my teeth. Tomorrow I’ll think the last industry I’d every want to be in is banking. 😉

Would you trust your money to SafeBank?

OS X Optimizations

Monolingual is a Open Source utility for Mac OS X that removes all the not-needed languages from your computer, freeing up hundreds of megabytes. My Mac mini is going “laggy” with the mouse jumping around instead of being smooth when I move the cursor around — any more tips for optimizations?

6 Steps to Kill Your Community

There have been a number of new platforms popping up recently that claim to increase your user engagement, get you more comments, increase your traffic, and more, through means that I consider short-sighted and harmful. Since people seem not to mind, I thought I’d write a guide for how to increase the number of comments you get by 400-1,000% and ruin whatever shred of community you had on your site.

  1. Don’t Moderate. Allow anybody to post anything regardless of whether it contributes to the conversation or not. Stupidity, libel, hate, curse words are all okay because in the comments you have plausible deniability. Make sure people know that whatever they post will live forever, and anything goes. The few smart people you did have in your comments will enjoy responding to these folks. Advertisers love being next to a good fight, too.
  2. Allow Spam Through. I don’t mean the obvious viagra mortgage stuff, but the human-written and surface-thoughtful comment that “Florida Real Estate” or “Poll Widgets” decided to leave on your entry. Or the guy who comments on every post and has a 3-link signature. Or the lame startup that mentions itself at every possible opportunity, however tangentially related it is. Once spammers catch a sniff of this stuff getting through, they’ll descend on your site like locusts and instantly double or treble your “community.”
  3. Force Signup. You’re not a blog, you’re a social network cum media empire and even to leave the smallest comment you should make me fill out a profile, preferably with demographic information you can use in advertiser pitches later. (*cough* CNET) Please ignore useful services like Gravatar and try to get me to upload yet another profile picture because you think that makes your site more “sticky.” (In a sense of the word, it does.)
  4. Don’t Participate in Comments. Make it clear that your post itself is for annointed authors who don’t mix with the hoi polloi in the comments ghetto. Don’t link or highlight anything good from the comment section; those people silly enough to contribute content to your site for free should feel ignored. If an author does happen to drop in and make a comment, make sure it doesn’t stand out from the rest so it’s lost amid the sea of…
  5. Random Crap from Around the Web. Make sure any comments you have are buried by every random piece of “conversation” from around the web, especially retweets, Delicious links, Digg and Slashdot comments, pretty much anything will work here. Bonus points for unmoderated pingbacks, so every scraper spam blog copying the content of the post gets a free link in the comments.
  6. Design Like NASCAR. The more buttons, widgets, stickers, and visual clutter the better. I want to see every possible login system including but not limited to Facebook, Twitter, OpenID, Google Friend Connect, and that Myspace thing. Because of their respective crappy terms of service, use the giant buttons they insist on. Also include a megabyte of “share this” icons for every obscure service in every language. People love options! Complexity is for closers.

As a bonus, here are a few extra that don’t make any sense to me, but seem to be popular:

  1. Abandon Search Engines. Make sure your comment system is externally hosted on a separate domain and injects comments in via JavaScript, so everything is completely invisible to search engines. For bonus points, even if you have an existing database, say WordPress, with all of your comments, make sure the external service doesn’t synchronize any of the new stuff back into it, so you’re at the mercy of the external provider forever and ever.
  2. Be Famous! You’ll get thousands of comments on almost everything you post and make sure only to let through the most sycophantic and saccharine, don’t tolerate real conversation or debate. To spice it up every now and then opine on a known controversial subject like abortion and let your audience loose on each other like gladiators while you watch from the stands.
  3. Put the Comment Form at Top. This ensures everyone making a comment hasn’t read any of the discussion so they’ll leave a comment anyway even if the exact same thing has been said already or a question has already been answered.
  4. No Subscriptions. Don’t allow me to get email notifications of new stuff, make me visit your site and reload the page constantly to see if there’s anything new.
  5. Make People Click Click Click. Ideally do 1-comment-per-page CNET-style and your pageviews will go through the roof, but if you can’t stomach that just make comments-per-page setting low or have some sort of complicated nesting scheme.
  6. Treat Everyone the Same. If I’ve left hundreds of great comments over many years on your site, please make me wait in the moderation queue like some random stranger off Digg. Don’t let anyone know I’m a regular, or talk to me, or invite me to test out beta stuff, or pretty much anything that acknowledges my existence or shows any degree of trust.
  7. Don’t Ask Anything of Your Audience. No polls, surveys, or open-ended blog entries. What do those plebes know anyway?

Shameless: IntenseDebate does all of the software stuff right, or will shortly. Core WordPress threading, moderation, and whitelists are already built in but also check out Subscribe to Comments and of course Akismet.

What are your pet peeves and rules?