Wesabe: Your Money. Your Community.

More and more web applications are storing sensitive data for their users, a trend of which Wesabe is certainly a part. Security breaches like the RockYou hack show what can happen when a popular web application stores sensitive data unencrypted and then has a lapse: millions of people can be affected at once. As some of the coverage of the attack pointed out, it was a good reminder not to store sensitive data unencrypted.

Wesabe has worked hard to come up with tools to protect our members’ data, both because the nature of our application requires that we ask for extremely sensitive information, and because we believe that all web applications should take security seriously. Today we’re open sourcing a piece of software, Grendel, that we think can help many sites (not just financial applications) protect users’ data from a RockYou-style mass disclosure in a simple way. Grendel is a new project that combines ideas we’ve used on Wesabe for years with other pieces we believe should be common infrastructure for web applications.

Nearly all web sites keep all of a user’s data unencrypted. In many cases this is a necessity, since the web site intentionally publishes that data; an encrypted blog wouldn’t have many readers. In other cases, though, the only time the data is used is when the user is logged in, such as in a word processing web application.

The idea of Grendel is to provide an internal (behind-the-firewall) REST-based web service to keep a user’s data encrypted and ensure its integrity when the user isn’t using it. Grendel uses OpenPGP to store data, with the user’s password encrypting an OpenPGP keyset. That model makes it easy for a web site to store data safely and only decrypt it when the user is logged into the site. Since only the user has their password, once they log out, their data is safe, even if the web site’s database is compromised or stolen. Of course this isn’t an infallible protection — there is no such thing — and in particular it doesn’t protect against web site developers acting in bad faith. It does, though, protect against an attacker getting access to all the secrets stored by users in one step.

Of course, data on web sites is usually shared with at least some other people in some way. Sometimes a user might want to share their information with the web site support staff, so the staff can help solve a problem or fix a bug. Or, the user might want to share their sensitive data with selected other users on the site, such as coworkers or family members. Grendel allows this, letting you encrypt data with multiple keys so that more than one user’s password can gain access.

It’s very easy to screw up when building a cryptography system — check out Nate Lawson’s excellent Google Tech Talk on common crypto flaws, or Matasano’s Socratic dialog on similar topics, for a map of the pitfalls available to you, and us. We’ve been fortunate at Wesabe to have a number of people who think very carefully about security, and they’ve put a lot of effort into designing and building Grendel. That said, we have two goals in open sourcing Grendel: first, to make a tool available to others that could help make “cloud” applications in general much safer for everyone, and second, to open up what we’ve built so others can review and help us improve it. We would love comments on any aspect of Grendel, security or otherwise.

Grendel is available on GitHub now, thanks to the efforts of Coda Hale (also the author of bcrypt-ruby) and Sam Quigley, who designed and built it. If you’re building a web site that stores sensitive data for users, go check it out, and if you’re of a mind to, help us make it better. We’d love to see “cloud” applications have an easy time treating users’ data securely, and we hope Grendel will be a useful tool for that purpose.

I wanted to post a quick thank-you for the fantastic first-week reaction to our new site for banks and credit unions, GetSpringboard.com. I somehow expected that the Thanksgiving holiday in the U.S. would make the first week slow and easy. I was wrong!

George Pasley perfectly nailed why we’re excited about GetSpringboard.com:

Wesabe is doing something that is unheard of in the financial industry. They are offering an online, consumer facing banking product that a FI can have up and running in HOURS. Think about that for a moment. Your FI’s CIO could go the Springboard website, select the plan she wants, break out the corporate credit card, and have an online PFM product running for their customers in a day or two at most.

I don’t know of another banking tech vendor that offers implementation like this. No salesperson to haggle over features and pricing. No project manager to coordinate all the departments needed to make it happen. Just five minutes to fill out the registration form.

Yup. That’s exactly what we’re doing. And the reaction has been amazing. Our guess that everyone likes convenience and low cost — even bankers — turns out to be spot on. Thanks for the great write-up, George.

As a reminder, we have a special offer for the rest of 2009: sign up for any plan and you get the price of the next lowest-priced plan for the life of your subscription. You can save up to $10,000 a year for as long as you use Wesabe Springboard, but only if you sign up before December 31st, 2009. Don’t miss it.

We’re really happy with how many people have reacted — and acted — so enthusiastically to this new launch. Thank you. We’re working hard to make the site even better for you, your members, and all users of Wesabe services.

Today, Wesabe launched a new product line for banks and credit unions: GetSpringboard.com. Earlier this year, we started offering Wesabe tools to banks and credit unions, but this new product site allows financial institutions to get pricing, sign up, deploy, and configure their Wesabe Springboard site, all through the web. Instead of the traditional banking software sales process — where it can take months and tens of thousands of dollars just to get new software launched — GetSpringboard.com puts leading personal finance management (PFM) software into financial institution customers’ hands in minutes.

Since Wesabe had its start as a consumer-based site, and since we have led the development of the online PFM space from that vantage, we know full well how tiring and time-consuming the enterprise banking software sales model can be. We wanted to give our bank and credit union customers all of the advantages we offer our consumer-site users, from easy setup and sign-in to full customization. The GetSpringboard.com site lets us do that. We’ve learned from our friends at 37signals.com and other online application companies, and we think their model is better. We wanted that for our customers, too.

We’ve had a great response to Wesabe Springboard, and a great set of feedback to the GetSpringboard.com development. As one credit union executive put it, “We all grew up on the web, and we want to see our options and place our orders on the web for everything!” Bringing that model to banking software confronts some peoples’ expectations, but we believe, as that executive says, that everyone prefers ease and convenience. Even bankers!

To introduce this new site, we’re offering introductory pricing for this year only. Sign up for any plan at GetSpringboard.com before December 31, 2009, and we’ll give it to you for the price of the next lower-priced plan for as long as you keep that plan.

For instance, if you sign up for the Gold plan (normally $1799/month), we’ll give it to you for the price of the Standard plan (normally $999/month) indefinitely. You’ll save $800 a month for the life of your Springboard site.

Note: if you decide to upgrade to a different plan later, you’ll lose your price break, so be sure to sign up for the level you think you’ll want to keep.

We think this is a great deal and a fantastic incentive to sign up today. We hope you’ll check out GetSpringboard.com and let us know what you think. Hopefully your institution will soon offer Wesabe tools, so you have the best interface available straight from your bank or credit union.

Three years ago today, November 17th, 2006, we launched the Wesabe site to the public for the first time. We were the first of what has become a wide and varied field of online personal finance applications. Our site has changed a ton since our launch, too, as we’ve learned from our users and the market.

I was asked the other day whether I thought that online personal finance applications are now a “defined” market — that is, have the borders of the space been set and is all that is left figuring out who can get the most users. “Uh, no” was all I could think to say. Are consumers’ financial lives somehow solved? Is it somehow the case that people worry substantially less about money now than they did? Has our much-publicized space solved money in the way Google solved search or Amazon solved books?

No, we haven’t. Neither Wesabe nor anyone else has made the personal finance app that we set out to build: one that makes people’s financial lives so easy that they no longer list money as their number one stress in life. There’s still so much more to do.

Wesabe has spent the last six months solidifying and perfecting the core part of our app, and we’re happy with the results. We think we have a great app and a great basis for building more. But don’t think our fourth or fifth birthday will look like our third. We know all Wesabeans, and consumers everywhere, need far more, and we can’t wait to show you what the next year will bring.

Thanks to everyone who has supported Wesabe for the past three years. The messages you post in Groups and on Twitter and send to our support line every day, telling us how Wesabe has helped you and made your lives better, are what bring us to work every day. They are more reward than any other we could ask for. Thank you.

Wow, it has been quite a week.

Over the past five days….

• We launched Wesabe for iPhone, our iPhone app, which allows you to edit and tag your financial transactions, see graphs, use the phone’s GPS to automatically locate where you just spent money for fast entry, and more. The reaction has been fantastic, as you can see from the early reviews on the App Store:

  

  Incredibly cool! It’s been so awesome to hear people’s reactions and feedback: “This is so much better than anything else out there.” Thanks! Wait until you see the next version.

• We changed our tagging system, which — since tags run everywhere in Wesabe — is a major change, both in amount of work and the impact on the site. This one caused some confusion, partially because people were accustomed to the old system, and partially due to some bugs in the first launch of the new system. We’re working out those bugs, and then I’ll post a whole guide to tagging, which will explain what we did and why we did it. Until then, please be assured that all of the capabilities you had (including split and one-time tags) are still there — we simplified the interface and gave you more control, even if it may not seem like it yet.
• We redesigned every single page on our site. Most of the reactions have been completely fantastic:
  

  

  

  

  

  Those have all been wonderful for us all to read, and we’re happy to have hit such a great level in the design. Some of the reactions, though, have been less happy:

  

  Oof. We know that redesigns can be disorienting. Other people were unhappy about the lack of search and export, and others about the readability of text in the new design. Not to worry — to those who has expressed dismay at the changes or missing features, we have heard you, and we’ll fix those concerns. We may take a different approach than the old design did, but we’ll do it at least as well, if not much better. QIF import is already fixed, and the rest will follow shortly.

  A few people have asked why we removed features during the redesign. Sometimes it was because they just hadn’t been updated to the new design yet, or had bugs that were too severe, we felt, to ship. In other cases, though, we wanted to get rid of or revise features we didn’t think were working well enough, and pare down to the essential parts, the ones we know are neeed and work well. Was this the right decision? I absolutely still think it was — for instance, we removed Goals and Tips — two whole tabs out of four — and haven’t gotten a single complaint about either of them being gone. We do believe reducing the site to its essential features makes it easier to use for everyone, and we’ll add back the features people are actually missing. We’ll be happy, though, to keep out or substantially revise the features people didn’t like or haven’t missed.

  Overall, the reaction to the redesign has been completely incredible. So many people have said, “Now I can introduce this to my [less techinically savvy person]!” That’s what we were going for — getting Wesabe to more people without giving up any of the control or power we’ve always provided.

• Last but certainly not least, we announced our first Springboard customer, Delta Community Credit Union, the largest credit union in Georgia. We’re wonderfully happy to be working with them — they take helping their members extremely seriously, and have been excited about working with us, and we’re happy to be bring Wesabe services to a new group of people by working with them. We like working with people who share our values about helping consumers, and the credit union community, and Delta Community especially, line up perfectly with that.
  

So, yeah. That’s a lot for five days. Thanks so much for all the great feedback and response, and for those of you looking forward to a fix or addition, no worries, we’ve heard you and will respond quickly. Follow @WesabeUpdates to get up-to-the-minute detail on new fixes and additions, and we’ll announce the big ones here. Thanks again!

In November of 2006, we launched the Wesabe site after about two years of talking about it, figuring out what it was going to become, and then building it, one piece at a time. Wesabe was the first of what are now known as “Social Personal Finance” sites, and we didn’t have anything to guide us except the history of Quicken — from which we were explicitly trying to get away. Wesabe was new and the idea was new, and that meant we had to make it up as we went along.

We spent a lot of the year leading up to Wesabe’s launch sitting in our cramped, informal office (in the back of a clothing store owned by my co-founder, Jason, and his wife) sketching, arguing, trying things out, and getting the pieces to fit together. With the graphic and UI design of the site, much of that process took place between me and Jeff, the site’s original designer, sitting in front of the gigantic monitor in his office. We essentially “pair programmed” the design, which helped me enormously as the product engineer, even if it sometimes frustrated Jeff’s better aesthetic instincts.

Now that the Wesabe site has been live for almost two and a half years, we’ve taken a chance to go back to our initial design and redo it, incorporating what we’ve learned since launch, and letting Magera, our designer, have the lead. It has been a chance for us to clean up some of our mistakes, make the site easier to use all around, and introduce a new standard of design for our site. Where with the first design I was extremely controlling about every little piece, this time I told Magera, “Make something you would like” — and it turns out we all like what she likes.

Wesabe’s whole site has been completely redesigned — every page is different. Rather than having a developer-led UI, this is designer-led development, and we’re all extremely happy with the result. We hope you will be, too — and please do let us know what you think.

(Note: As part of this launch, we’ve disabled or minimized some of our site’s features to get the simplest version out the door that we could. Some of the site’s graphs, export and search features, and side features like Wesabe Mail and Wesabe Cutback will be offline for a short while. They will be coming back very soon. Our Tips and Goals tabs are on the blocks for more major revisions — those will both return but in different forms. At the same time, Paul has completely revised the new User Manual, so we now have substantial and valuable online help for almost all of the site’s features. We’ve also added in some long-time requests from our community, such as spending target histories so you can go over your budget for months past, and a tag cloud that takes into account how much you spend on each tag.)