Shipping like there’s no tomorrow

The concept of shipping as per Seth Godin is a beautiful thing. In essence it relates to the idea of getting something out (i.e. “creative output”; a “product”) without the need to achieve perfection before getting that something out. Ship, get feedback, improve; then ship again, get more feedback, and improve once more. Do this again, and again, and again.

In my role as an analyst, I create a lot of data products that many people use. Things like sales reports; spreadsheet tools and models; and data analyses. Not being one to stick with the status quo, I’m always looking for ways to improve the stuff I create (better interfaces, new metrics,  distribution methods, and more).

In such experimentation (which seeking improvement invariably entails) there’s always a risk that something will go wrong: calculation errors (especially common in complex spreadsheets of which aim is “simple for the user; complex for the developer if need be”); interface usability issues (I tend to work on a large screen that not everybody has or uses); software-compatibility (I use Excel 2013 as I need to know what’s “out there”; which sometimes proves problematic when I use a 2013-only function that 2010-users can’t use); well, the list runs long.

And I’ve had my fair share of failed “improvements” (the latest affecting a couple of very senior management, with my experimentation with Excel Pivot Slicers — very cool stuff by the way — breaking a couple of vital sales reports on their Macs*).

But without having taken these risks, without having shipped, there would be no way that I could be doing as much as I currently do, no way that I could have done as much as I have done. If I had listened to my fears and resisted stepping into the unknown; if I had thought to myself, “let’s not rock the boat, let’s stick with what we know works”, I would be a miserable little automaton working in a far less colourful sales operations environment. And many sales people would still be working with the tools they had in 2013.

Shipping like I do now, though, didn’t come overnight. At the start, I had to push the envelope a little by a little, testing the waters to see just how far I could go. Every little success inspired me further, and built up the all-too-important goodwill that proved useful whenever I messed up;  while every misstep gave me insight into how “open to failure” the environment in which I worked in was (I’ve got very accepting colleagues I must say).

It wasn’t always easy though. Still isn’t. Doubt is a constant companion. I remember once spending three weeks thinking and experimenting with a myriad of ways to capture and report sales targets, a job that I’d been tasked to do.

I was looking for a way that was as intuitive and as “smart” as possible. I must have found at least a hundred ways of how it shouldn’t be done. As the deadline approached, I started feeling like I was being too ambitious, and couldn’t help shaking off the feeling that it had all been wasted time. Maybe it doesn’t have to do all that I wanted it to do.

But I stopped myself. There was still time. “Let’s try,” I told myself, “and let’s ship.” And I did just that.

I shipped an initial “prototype” version, ran it past my boss. It was a rough-and-dirty version, but it showed the concept well. I further developed it, and shipped it past sales management. I showed them screenshots. I did a demo.

What they didn’t realise was that the demo focused on the 10% of what worked; there were plenty of bugs, plenty of things that I hadn’t managed to work out. The actual product was still days away (maybe weeks) from being completed, but I sold them the story, and they bought it. They were excited.

But I certainly wasn’t. I was worried. Worried as hell.

But there was no turning back now. I’d demonstrated what the cool new tool could do (more like what it should do, really). I had no choice. I needed to complete this.

Due over the next few days, I worked like a cow suffering from Bovine Spongiform Encephalopathy (i.e. “mad cow disease”, geddit?) coding and re-coding; testing and re-testing. Eventually some of the most major issues that I had encountered I managed to solve. I shipped again, on the deadline, with 80% of what I had envisioned included.

And boy were they were happy (most of them, anyway). Which made me very happy.

Oh, and the other 20%? Repackaged as “future optimisations”. Which, yes, I eventually did ship too.

* I’m happy to announce that we eventually found a workaround, which appeased the Mac users. The Pivot Slicers remain and prove to be an extremely valuable addition to our reporting arsenal.

Let me know what you think