The Filter (week of June 30) 37signals 30 Jun 2006

4 comments Latest by JF

A few interesting comments posted this week at Signal vs. Noise:

Edward Hall: The perfect group size = 8-12

Daniel 28 Jun 06
The statistics showed that testing with 8 subjects will indentify 90% of the problems a given interface can have. More people, and the rate at which problems are identified will only asymptotically approach 100%.

Dean Brooks 28 Jun 06
1. Group output tends in many situations to fall logarithmically with group size. Each doubling of group size reduces average output by 20 percent on average. 2. Up to a certain point, groups of odd-numbered sizes tend to have a wider range of outcomes than groups of even size…

As I said, my focus has been on historical and military problems. However, I did turn up one interesting business example: companies that make the INC 500 (a list selected for extraordinary five-year growth rates) tend disproportionately to be founded with 3 or 5 employees, rather than 4 or 6. Obviously, as they hire additional people, the group size changes — but the inference would be that they stand a better chance of making big early gains or establishing a really strong business plan with a group of 3 or 5.

Growing in vs. growing out

Scott M 28 Jun 06
“Make sure you make it easy for new people to get in. ” Another way to state your advice is to favor high growth over high margin. There are two ways to make money - with long sales cycles of high margin items to a few customers or short sales cycles of low margin items to many customers. Small software companies will find the best chances of success with the later approach. When you are small, there is simply much more opportunity for revenue increase through growth than incremental product margins and high margin activities bring a risk level (a couple big customers can make or break you) and costs (long sales cycles, higher support costs) that most small, growing companies should avoid.

Josh Williams 28 Jun 06
This is so very true. We occasionally get asked if Blinksale will grow with a business as they add more employees and salespeople. As in say, over 100 employees. And the answer is no. We built Blinksale to service smaller companies. If I had 100 employees I wouldn’t be using Blinksale. This isn’t a knock on our own software. It was simply not intended to do that. And the nice thing is that somewhere, out there, someone is making invoicing software for larger companies. It may be ugly and hard to use, but someone is making it. Just not us. And that’s okay. For every person that outgrows Blinksale, we add plenty for whom Blinksale is just right.

Phil 28 Jun 06
Why can’t something be simple but offer more for bigger companies as well? I agree with your philosophy of keeping Basecamp the way it is, but why not create another product (“Summit”) that does the more advanced things in a simple clean way? Then not only would you continue to sell to the same audience basecamp appeals to, but also sell to people who need more in the first place, and those who grew out? Software is not like a pair of jeans, for many people it requires training and getting used to the way something works (even when its simple) and it’s a pain to simple grow out of it and move on to something completely different. We love your products and are throwing the money at you, why not build it for us? :)

Liz Fraley 28 Jun 06
In many ways, I agree with Dan Bricklin: There’s no reason that software cannot last 200 years http://www.bricklin.com/200yearsoftware.htm

John Critz 29 Jun 06
The problem with the “it’s just a simple little feature, why can’t you add it” perspective is that you’re not thinking about the hundreds of other feature requests people have for the way they do business. Sure it’s probably a valid request and would save you some time, but only you and maybe a handful of others. So it’s usually not worth the development time or cluttering of the app. for a handful of people to benefit from it.

Reality vs. drawings and words

Mark 29 Jun 06
Far too many times, especially with home builders, clients are taken to a “design center” to look at swatches and samples of carpeting, paint, bricks, tile, marble…to have installed in their own homes. The problem is, there is an insant disconnect between what the customer sees and agrees to, and how it comes out in reality. The spaces aren’t the same, the play of light and shadows on the materials is different, the sunlight bouncing off the material is different than the fluorescents of the design center.

Chris 29 Jun 06
Another great facility used to help reduce the cycle for the client to understand something is quick builds. We deploy to the target environment on a nightly basis so that our client can get her hands on the latest and greatest of “what she knows”, which is the application. Instead of saying hey isn’t this great, you’ll see it in a month or so you see it the next day. Again, if there are issues or the client doesn’t like the way something feels, we know right away.

4 comments so far (Jump to latest)

Not a copy cat like you guys 03 Jul 06

Not sure if anyone pointed this before to you bunch of plagiarists but you are guys are by far the worst wanna be googlers

so mcuh so you copy each and every philoshphy of google and sell them as your own?

getting real shit is just a crappier copilation of art of projet management google approach to product management and agile process

you guys are cheap i mst say claiming concepts that are not evn remotely yours

Matt Todd 03 Jul 06

I’m not an expert, nor part of 37signals, but I’m almost certain that there really isn’t an ownership to these concepts and philosophies, merely a practice and communication of the concepts. For 37signals, they practice their philosophies, but then they capitalize on their understanding by teaching them to those who might not know or understand them fully. They aren’t copying to be copying, it’s just a part of being an adherent.

And, on top of these cohesive philosophies, they blend in their own understandings and ideas to make a blend that is all their own. This isn’t copying but derivation: a new piece of art.

So, the real question comes, then: why all the hate?

M.T.

Hm 03 Jul 06

So, the real question comes, then: why all the hate?

they make money, so fuck them. And by association, fuck YOU! AND by association, fuck google.

*flails arms and leaves*

JF 03 Jul 06

From the introduction in our book:

“We�re not claiming to have invented these techniques.
Many of these concepts have been around in one form or
another for a long time. Don�t get huffy if you read some
of our advice and it reminds you of something you read
about already on so and so�s weblog or in some book pub-
lished 20 years ago. It�s definitely possible. These tech-
niques are not at all exclusive to 37signals. We�re just telling
you how we work and what�s been successful for us.”