I’ve purchased a few things lately that I’d like to recommend.

11” MacBook Air

I recently switched from a 15” MacBook Pro to an 11” MacBook Air and I couldn’t be happier. The 11” Air is the best computer I’ve ever owned. Everything else is a distant second.

It’s my only computer. I do all my work on it. I code on it, I design on it, I browse on it, I run 37signals on it.

I originally purchased an external monitor because I thought I’d need the extra space, but I’ve found I like the smaller screen of the 11”. I don’t use the external screen at all. The smaller screen keeps me focused and it’s the right size to run full-screen apps.

If you’ve been considering an 11” Air, and you’ve been on the fence because you’re worried the screen is too small, take a chance and pick one up. You won’t regret it.

Defy Bags: Square

I’ve found a million bags I don’t like. This time I found one I love. It’s the Square from Defy Bags and it happens to be made by hand right here in Chicago.

It fits an 11” Air and an iPad perfectly. It’s small, light, and extremely durable (it’s made from recycled military tarps). The buckle is a AustriAlpin Cobra which is the best buckle I’ve ever used. So easy to confidently click open and close, the buckle is built like a tank, but with lightweight materials. I think these buckles wholesale for about $35 each, so you’ll rarely see bag makers use them. Defy Bags ponies up and uses them because they’re just that good.

So if you’re in the market for a new messenger-style bag, check out Defy Bags. They aren’t the cheapest, but they’re a great value.

Bialetti Brikka stovetop espresso maker

I’m a tea drinker who’s starting to learn to love espresso. And for just around $50 on Amazon, you can pick up a Bialetti Brikka and make some decent espresso right on your stovetop. No fancy or expensive machinery required. No electricity either.

Certainly you can get a better espresso, but if you want to make your own at home, and you don’t want to sink a ton of money into it, I haven’t found anything better than the Brikka. It’s simple, it’s durable, it’s absolutely worth $50. I’ve been enjoying it with Lavazza Super Crema Espresso Beans (ground at home with a Bodum Bistro burr grinder).

Susan Lehman from the New York Times asked me to write an opinion piece for this sunday’s paper.

Because…

Susan liked what I had to say in my “Why work doesn’t happen at work” TEDx talk.

Because…

I was asked to speak at TEDx Midwest by Brad Keywell. Brad was one of the Groupon co-founders, and I met Brad because I served on the Groupon board from 2009 to 2010.

Because…

Andrew Mason, Groupon’s CEO, asked me to be on the board.

Because…

Scott Heiferman, co-founder of Meetup.com, and a mutual friend of ours, introduced Andrew and me and we had lunch in early 2009.

Because…

I had gotten to know Scott over the years after 37signals designed the original Meetup.com site back in 2001-2002.

Because…

Scott emailed me back in 2001 asking if he could meet while he was in Chicago visiting family. He liked our early web design work at 37signals. No one had ever asked to meet me out of the blue before – and barely anyone knew who 37signals was – so I was flattered and said yes.

Because…

I had started 37signals with two co-founders in 1999. One of those founders was Ernest Kim and the other was Carlos Segura.

Because…

A few years earlier, I had interviewed for a job at Organic Online in Chicago. Ernest Kim was the creative director. I didn’t take the job, but Ernest and I hit it off over design and Nike, so we kept in touch.

   and

A few years before that, I was hired on a contract basis by Carlos Segura to help them redesign and rewrite their internal FileMaker Pro database they were using to keep track of their clients.

Because…

I made a popular FileMaker Pro-based app in the 90s called Audiofile which helped people keep track of their music collection. Carlos liked the app and the design and found out I was behind it.

Because…

I couldn’t find a simple tool to keep track of my growing music collection.

Because…

…I can’t remember enough specifics before this, but the chain obviously continues – each link connected to another by a seemingly unrelated event. And I’m sure I’m passing right over a handful of subtle links that made the major links happen.

When you look back on events, it’s pretty incredible how things come together. Nothing happens independently. Everything is tied to something before it. Sometimes the links are more obvious than others, but it’s healthy to take a few moments to reflect on how many things – and people – had to come together in order for another thing to happen.

You just never know.

Since I joined 37signals, I have been working to improve our monitoring infrastructure. We use Nagios for the majority of our monitoring. Nagios is like an old Volvo – it might not be the prettiest or the fastest, but it’s easy to work on and it won’t leave you stranded.

To give you some context, in January 2009 we had 350 Nagios services. By September of 2010 that had grown to 797, and currently we are up to 7,566. In the process of growing that number, we have also drastically reduced the number of alerts that have escalated to page someone in the middle of the night. There have certainly been some bumps along the road to better monitoring, and in this post I hope to provide some insight into how we use Nagios and some helpful hints for folks out there who want to expand and improve their monitoring systems.

We’ve been working on putting together some feature introductions and I thought I’d share what goes into making one.

I like to start off without a script and just start recording some screen capture with voiceover. This is edited together into a rough version of what the story will eventually be.

New Feature test – Rough Cut 1 from 37signals on Vimeo.

Feedback: The design on the blackboard can be more clear. It also takes too long to get into why someone would use this.

Here is the third rough cut. I’ve re-shot the blackboard with a cleaner design and swapped the live-action example with the explanation of the feature.

New Feature test – Rough Cut 3 from 37signals on Vimeo.

Feedback: This still feels too rushed. I think we are trying to solve two very different problems in such a short time. This can be a simple introduction to the feature without having to show exactly how it works. We can also cut a lot out of the live action scene. We don’t need to show people how to send an email.

In the fourth rough cut I’ve re-recorded the voiceover into a shorter explanation of the feature, chopped off the end of the live-action bit and added a title card.

New Feature test – Rough Cut 4 from 37signals on Vimeo.

Feedback: The transition between the example and explanation is too abrupt. Some of the mouse clicks are unclear, maybe there is a way to highlight those.

Rough cut number 5 has a second title card in the middle of the piece to help the transition and both cards have been re-worked by one of our designers. I’ve added some simple camera moves to highlight the important bits and make the video more interesting. I also added one of our bumpers at the end to wrap it all up. Again, I recorded a new voiceover track for this version though I think I sound a bit grumpy.

New Feature test – Rough Cut 5 from 37signals on Vimeo.

Feedback: This is getting very close. Can we fix the fish-eye effect on the blackboard scene? Also, I don’t much care for banjo.

Now that we’re in a pretty solid place for this video it was time to really polish it up. I re-did the camera moves in After Effects, re-worded one of the titles, recorded a much friendlier voiceover and re-shot the live-action scene with a better camera and some fancy focus pulling by Jason Fried. I also, of course, took out the banjo music.

Basecamp – Emailing Content to a project from 37signals on Vimeo.

It’s easy to dismiss the value of multiple iterations in a project like this. It took 8 cuts to get this simple screencast to a place we liked. A lot less went into the production of this than some of my other videos, but that doesn’t mean it doesn’t have as much to accomplish. A great man once said “Size matters not.” Don’t think about projects in terms of size and scope. As long as the the goal is reached it doesn’t much matter how you get there.

Customers don’t just buy a product — they switch from something else. And customers don’t just leave a product — they switch to something else.

It’s in these switching moments that the deepest customer insights can be found.

On the 1st of October, a select group of 24 people will attend a unique, hands-on, full-day workshop to learn about “The Switch”.

Most businesses don’t know the real reasons why people switch to — or from — their products. We’ll teach you how to find out.

The workshop will be at the 37signals office in Chicago. The cost to attend is $1000. The workshop will be led by 37signals and The Rewired Group.

You’ll participate in live customer interviews.

You’ll learn new techniques for unearthing the deep insights that most companies never bother to dig up.

You’ll understand why people switch from one product to another and how you can increase the odds that the switch goes your way.

And you’ll be able to put everything you learned to immediate use.

There’s only one simple requirement: You’ll be asked to bring something with you. It won’t be a big deal. Details will be provided one week before the workshop.

Spots are limited. Only 24 people will be able to attend and participate.

If you really want to know why customers are switching to — or away — from your product, attend. If you’re only mildly interested, skip it to make room for someone else.

Want to be one of the 24? Register now.

Officially, 37signals customer support is only available during standard U.S. business hours — 8 a.m.-5 p.m. CST (14:00-23:00 GMT). In reality, we’re usually open a little longer than that, because some of us are morning people (Merissa!) and some of us are night owls (Joan!). We’ll sometimes pop in on weekends too, especially when it’s busy. Still, that only amounts to about 60 hours of availability in a 168-hour week.

That’s been a bummer for folks in other parts of the world — nearly a fifth of our customers are in Europe, and at least 5 percent work from Asia and Australia. While it’s often easy to get back to our American customers in less than an hour, Europeans could be waiting more than six hours for a response, and our friends down under could wait a whole business day. When you’re unable to log in to your account, or somebody removed your admin permissions, or you can’t find that file you know you uploaded yesterday, well … that sucks!

It sucked even worse after we launched the new Basecamp and suddenly had hundreds more cases per day with the same tiny team of seven. Cases were piling up to 400, 500, 600 deep. It got so bad that a few times, we straight-up shut down intake, just so we could catch a breath. Not cool.

In June, it was taking us about 110 minutes to reply to emails during business hours, 279 minutes on average overall. Only 41% of customers were getting a reply within an hour when we were on the clock, and a dismal 29% received a response within an hour overall.

Ouch.

We had already decided to bring on some folks across the pond to better serve our customers in different time zones. We added the lovely and talented Monika (Netherlands) and Jim (UK) to the team (more on them soon!), and I’ve been working from Europe this summer as well — so lately, our day has been starting around 3 a.m. CST.

We’ve been experimenting with later shifts, too — Kristin, Ann, and Joan have been taking turns coming in around 11 a.m.-noon and leaving around 8-9 p.m. CST.

What this means is that, at least during the Monday-Friday workweek, we’ve gone from 8-12 hours of availability to about 18. So far, that’s making a world of difference.

Yesterday at 11 a.m., for example, the median time to a first reply during business hours was 24 minutes and 73% of cases were being answered within an hour. (You can always see how quickly we’re getting to things on our “Happiness Report” page.)

In the three and a half weeks before we started European hours and late shifts, our median weekday response time was six hours. Since we spread the hours out (in addition to some awesome deploys and feature improvements, too!), median weekday response time has fallen dramatically, to around 55 minutes. We’ve been reaching “Inbox Zero” at the end of the day — something that hasn’t really happened since we launched Basecamp on March 6. No more 100+ case backlogs to work through and prioritize first thing in the morning, because we’re constantly working through the queue.

Is this perfect? Heck no. We know that commerce is 24-7, and 55 minutes can feel like an eternity when stuff isn’t working. We know people rely on our apps to do their jobs, and when something goes wrong, it can create big problems. We take that super-seriously. We’d like to move toward around-the-clock support, where everyone is getting a reply within an hour. Ideally, no one would wait more than a few minutes to hear back from us.

That’s the direction we’re heading in — this is just a step toward it. But it’s been a major step, and we’ll all breathing a little easier and sleeping a little more soundly!

Thanks to Noah for his help with the numbers here.