We used to be a managed service provider.

Before we started building TrueSight, a lot of what we did involved sitting in chairs, trying to fix things, and answering the phone. We had boxes we relied on daily. Simple things — like a power strip that you could SSH into and reboot a machine through — saved us thousands of dollars and untold hours on planes.

And we had software. Complicated, hard-to-use, never-implemented software. I personally authorized a huge software implementation — half a million dollars — that we never got running (I won’t tell you the name of the company, to spare myself the lawsuits.) It was painful, and wasteful, and led to a ridiculous amount of finger pointing. In the end, a smart employee built something in Lotus Notes in his spare time that let us remotely manage an entire rack of multivendor equipment.

One of the things we’re brutally aware of, then, is complexity. When we started making equipment for data centers, we vowed to keep it simple. It’s one of the reasons we make things in an appliance format. We think that user performance monitoring should be transparent, secure, and as easy to maintain as a load-balancer or router.

It’s also one of the reasons you can see, at a glance, an entire user’s session and drill into it. And it’s one of the things that’s fueling the creation of a wide range of simple, intuitive visualizations (we call them Lenses, and we haven’t really shown them to anyone but our customers yet, but everyone who sees them can’t stop staring and clicking.) More on these in a future post.

So with web operations scars all over me, it makes me really happy when I see quotes from our customers that suggest we’re on the right track. Someone forwarded me a note on Friday that made my weekend:

“Every time a client calls up, the first thing someone says is, ‘Are they on Coradiant?’ So it’s become sort of the standard for troubleshooting. It went from sort of an exploratory product to one that has become almost mission-critical in regards to its usage … I mean, I’ve seen a lot of products that tout all their abilities, but this is one that actually truly delivers and delivers out-of-the-box, which is really the best part of it.”

Simplicity is great. But simple doesn’t mean limited features. In fact, we were talking to a very big software company last week and they looked at me an hour into the discussions and said, “this thing is like Excel: I can start using it in minutes, but there’s just so much depth and detail here I can get it to do pretty much anything.” (Although you can’t use it to play pac-man, as this completely unnecessary Excel macro does.)

I’ve often thought that product management isn’t about giving people what they ask for. It’s about not giving them what they don’t really need.

This year, we’re working on communicating that simplicity to people so they can grasp what we do and understand how useful it is. This video (about the iPod, but apparently created by folks in Redmond) is a great example of how to position for simplicity.

Part of making a simple product is understanding the problem you’re trying to solve really, really well. We spend a lot of time agonizing over small details and using smart defaults so our end users don’t have to. A lot of times, we don’t give people choices; instead, we make a choice for them. Choices, it turns out, are the enemy of great software. Every choice doubles the number of test cases you’ve got. Great technology is smart enough to make its own choices, and let the advanced users adjust those choices if they feel like running with scissors.

A friend showed me fellow Montrealer Andy Nulman’s blog on surprise — it’s great reading if you’re a marketer — and in his first post he claimed that:

“I have always shouted that consumers don’t know what they want. Well, I lied.
They know what they want.
They WANT to be led.
They WANT us to lead them.
They WANT to follow.
And they WANT to be surprised.”

That might sound a bit arrogant. And I’m confident that a lot of technology companies think it doesn’t apply to them: They’re selling non-consumer products, so their buyers make rational, economic buying decisions. Right?

Wrong.

Most people are innundated with rational pitches. I’m with Andy: Provided that the surprise includes novel, efficient, easy-to-grasp solutions to known problems, web operators are OK with us innovating on their behalf. Lots of our customers rely on us to plow the snow on their behalf; in other words, they don’t have time to figure out where web performance is going, so they expect us to do so.

Having sat in their seats for a few years, we feel their pain.