OnStartups

37signals: Feature Bloat vs. Business Bloat

Posted by Dharmesh Shah on June 22, 2006 38 Comments


As a quick overview, 37signals is a small web software company that creates a series of simple applications for project management and collaboration.

Those that follow 37signals (the company) have likely heard at least a few times from them why products should have less features and be kept simple.  I for one, agree with the importance of avoiding “feature bloat” whereby products become so bloated with features trying to do everything for everybody that they don’t do anything for anybody particularly well.   The reason for feature bloat is quite simple.  Developers feel that adding “just that one more feature” will give their product an advantage and attract a few more customers.  Besides, it’s software, so if the feature isn’t really “in the way” anyway, what harm does it really do?  And, if it’s a natural extension of what the product already does, the feature may be easy to add, and as such profitable.  As long as the feature is adding value to one or more customers, isn’t it a good thing?  The simple answer:  not always, and not usually.

I’ve talked about this phenomena before and though there are times when you need to make sure the right features are in your product, you should not use “everything and the kitchen skink” model in building products.  Adding features, even those that are useful to some number of people, has a cost.  There is the “hard” cost of developing and maintaining the feature, but there is also the opportunity cost because adding this feature takes time away from other things you could be doing to help a larger pool of customers.

In summary, Feature Bloat is a bad thing – but it’s not hard to understand why so many (including me) fall into the trap.  

So, now that we’ve taken a look at feature bloat and why it happens and why it’s a bad thing.  What about business bloat?

Business Bloat:  When businesses try to do more and more, not all of which is related to their core business causing their businesses to become “bloated”.

So, the phenomenon of business bloat is simply feature-bloat but at a higher level of abstraction.  In business bloat, we are looking at the business instead of an individual product.  Let’s jump right in to an example (using 37signals).  37signals has been proselytizing the need to avoid feature bloat.  But, let’s look at the company from a different perspective – at the business level.  
  1. Core business:  Building insanely simple web-based software products for micro-firms.  In this regard, they have done exceptionally well (and most would argue that BaseCamp and BackPack are the two products that put this small company on the map).  Nothing to talk about here.  This is what they do well.

  1. Signal vs. Noise Blog:  Nothing wrong with startups having a blog and communicating with customers.  In fact, this can be very, very helpful for a startup to start building visibility in the market and start having conversations with their customers.  However, I’d argue that there is (ironically) much more noise than signal on their blog, but that’s just my opinion.  The blog gets a lot of traffic and is a great vehicle for the company to promote itself and its products.

  1. Workshops:  The team is passionate about what they do and what they have learned in building the company.  They’re using this to educate new startup entrepreneurs on how they can “get real”.  Nothing necessarily wrong with this either, and it’s a noble cause (and makes money) – but these kinds of things do take time and one could argue that it has become less about the customers.  (Question:  What percentage of “real” customers benefit from one of these workshops?)

  1. “Getting Real” Book:  Similar to the workshops, this is just another vehicle to share all the wisdom and insight that the founders have accrued.  To avoid the pain and hassle of going the traditional route, they “self published” and sold the book online to make money.  It’s a pretty good book (and they made lots of money:  $175,000 at last count, I think).  Once again, they’re passionate about what they’ve learned and want to share it with the world.  They’re also making a fair amount of money at it.  Not a bad thing.  But now, we’re starting to see more and more of a “services” company buried inside 37signals.  Nothing wrong with that, just a different company.

  1. Job Board:  As an extension of the blog, they start a simple web-based job board where employers can post recruitment descriptions ($250/month, last I checked).  This is seemingly a way to make “easy money” by leveraging the immense base of designers and programmers that frequent the 37signals blog.  Once again, I have no doubt that this is a profitable endeavor for them (in fact, I can’t think of a whole lot that the company does, that doesn’t make money).


So, the question is this:  As long as the company is generating much-needed cash and making money, what’s wrong with all of the above?  Maybe nothing.  Maybe in the early stages, startups should be as “opportunistic” as possible.  But, some of these arguments start sounding a lot like the same arguments that lead to feature bloat.  Sure, each individual feature is easy and makes some money (by acquiring a few more customers), but if adding more features leads to feature-bloat, doesn’t adding more and more “non-core” activities to the business lead to business bloat?

Extending the 37signals example, how have items #2 - #5 added value to customers and strengthened the core business?  Could time spent on these activities be better used to further the direct value to customers?  As the torch-bearers for the “less features, less bloat” mantra, are the smart guys at 37signals following their own advice?

What do you think?  Am I simply envious of what they’ve done and how easily they can make cash doing just about anything?  Should they continue to exploit opportunities to monetize their brand and current popularity?  What about a 37Signals PR firm (they do most of their own PR, and know a thing or two about getting press)?  What about a 37signals “Usability Certification” program whereby they “certify” other people’s products for usability (at $750/each)?  There are likely many more things the company could do, all of which would likely make money.  But, what should the company do?

Would love to hear your thoughts.