Well, after a couple of weeks of fussing over the finer details, PocketSmith is finally in Private Beta, and this week we’re starting discussions with a number of organisations around the possibility of taking on the application as a pilot.
It’s the point in time we’ve been waiting for pretty much since we started, really. It’s been difficult getting the last couple of weeks out of the way as the activity of developing software for our own eyes only starts to get a little tedious!
The Private Beta was delayed for about a week because we realised a week prior to our planned launch that there were many things that simply weren’t good enough.
So, what have we learned thus far?
–
Fine details take time
Bug stomping and UI fixes really sucked up the final run up to the Private Beta launch. Hey thanks, Internet Explorer! All web developers of the world should take note of the time taken to write markup specifically for this ass of a browser and send a bill to Microsoft as penance for its remarkably poor adherence to web standards. The timesheet should add up to a couple of millennia.
If the comments above don’t make much sense – what I’m trying to say is that PocketSmith’s layout, though compliant enough to work across Firefox, Safari, Opera, Camino, the iPhone and even Google’s new Chrome – will still need further adjustment and exceptions in order to render properly in Internet Explorer. This is akin to shooting a movie that plays on just about all TVs, and having to re-enact key scenes and re-build props just for Microsoft’s special TV. You know, the one that came with your Microsoft House that you never bothered to flush down the Microsoft Toilet in favour of a better free TV, like Firefox or Chrome.
Writing copy takes time
One of the things we’ve been putting off is proper guidance for our users. It may sound strange, but when you’ve got your head so deep in the application it seems almost odd to have to take a step back to write down what seems obvious to us.
Here’s a little something I’m experimenting with. I’ve always been enamoured with Haiku, and of its ability to convey meaning in a simple and concise manner. Nicely analogous to what we’re trying to do with financial information, and perhaps a measure of good application design is in the ability to describe what’s happening in 3 sentences, and for extra points, 5-7-5 syllables! Or maybe I’m just a lazy copywriter.
A little ambitious perhaps. But if at all possible, I’d still like to see Haiku sprinkled in amongst PocketSmith’s pages.
Timing is everything, perfection comes later
Words from a wise man that came just in the nick of time (a gift this particular individual seems to be endowed with). We fretted about the lack of one particular feature that we felt would bring the Final Beam of Coherence to our application. We’d built quite a bit of emphasis around this that eventually, we felt that we simply couldn’t release the application without it.
But were we losing sight of one of our core principles? We need to iterate often. If we are too focused on building to perfection, we could lose sight of what our market wants; worse, we may even miss out on a number of key opportunities. It doesn’t hurt to release a little earlier and to continue to release often.
PocketSmith is already much matured from the Alpha, and has certainly seen innovations beyond the scope of my original business plan. Fact of the matter is, the product will never be complete. So we’ve decided to release the Private Beta without the feature (which will emerge in due course, and I assure you, it will be great). Thanks, Des.
–
The strategy for the Private Beta release is a little bit different to the Alpha. While we really sweated it out in the final hours leading up to the Alpha launch, we decided that this should be a slightly better-managed occasion. Here’s what we’re doing.
Load-testing the server in user chunks
One of the issues plaguing usability in the Alpha was the speed of the application. We deployed to Heroku, who are still in beta themselves. The technology is amazing and they’re still free, but understandably not running at full steam until they’re ready to.
So for the Beta we’re on Slicehost, and we’ve got a pretty good idea as to how many users a single slice can handle. So what we’re doing is shelling out private invites in stages and monitoring the load on the server. I suspect we’ll have quite a few users out there before the server starts to creak as there aren’t going to be too many concurrent users.
Setting up a mailing list
The Alpha culminated in a survey that we encouraged our users to fill out. This was good as we had a number of key assumptions we needed to validate. It was a pretty lonely affair for the users however, as communication was pretty much between themselves and us. We’d like more interaction between the users and some active discussion, and so we’ve set up a mailing list.
Our users are busy people, and so forums aren’t the best.. well… forum for hosting discussions. Mailing lists deliver messages directly into inboxes, and responses from a user go back to everyone on the list. Simple. Hopefully, this will be the start of a little community – if only for the Beta.
James has found that Google Groups does this remarkably well, and we suspect it does as much as most of the paid-for versions out there do.
Iterating often
We’ll continue to develop and deploy changes to the Beta environment. The Alpha was a static affair, save for a few bugfixes. We’re still in the process of determining a rollthrough schedule and some of the operational details that go with it, but we’d like to reward our Beta participants with continual updates to check out perhaps once a fortnight.
Talking turkey
Yes, we’re going to see if people will actually pay us money to use PocketSmith. Sticking with the principle of continuous validation even through development, we’ll figure out as early as possible what potential customers will pay for. By our reckoning, the product is *just* at the stage where we can do this – we certainly couldn’t have done it any earlier, although there’ll be continued improvement as we carry out discussions.
For early-stage commercial discussions, it’s all about striking the balance between having a good working prototype that won’t let you down, versus an over-developed final product that has either needlessly cost you months of opportunities, or one that needs to be re-engineered to suit the client or market.
Throughout the process, we’ll be polling our colleagues and friends in related industries to get a rounded opinion.
–
So this is where we’re at! When it’s a little less commercially sensitive, I’ll write a bit more about our recent change in strategy too. I’m particularly excited about this one, and have been somewhat looking forward to our first fundamental change in the perception of the direction of our business.
Paul Graham of the Y-Combinator states that the initial idea is just a starting point – not a blueprint:
The fact is, most startups end up nothing like the initial idea. It would be closer to the truth to say the main value of your initial idea is that, in the process of discovering it’s broken, you’ll come up with your real idea.
We’ve been making little iterations along the way, but it’s actually reassuring to know that we’re on our toes enough to perceive – and even more importantly, agree on – the changes in our plan that need to happen in order to make the business as much of a success as it can be. The journey continues, and with it, our exploration.
