Before the Redesign

So you've got this product. So I've got this product.

It's totally unique, the only entry in its field, and everybody knows it. It's a rookie entry into an extremely crowded (if not particularly well-differentiated) field. Almost nobody knows it exists.

It's a time tracking application, and it's called Freckle Time Tracking. And it's totally different, but who knows and who cares?

Sizzling steak—yum! Raw meat of unknown pedigree, not so much

The product site sucked. We focused on building, launching, and polished the product. Naturally, I wanted to build substance before trying to add—much less sell—the sizzle. The app itself has plenty of sizzle but the product site? Soggy.

This is what the product site was like, circa a couple weeks ago:

Go ahead, click it for a full-sizer, complete with legible text.

And that was it. There's no screenshot of a site tour or a screencast because, well, there was no tour, no screencast, no list of features, no… anything.

Not just an ugly face

Aside from being generally objectionable on those grounds, it was ineffective.

People didn't understand what the service was. Our conversion rate was pretty typical (just under 3% sign up for accounts), but I think we can chalk that up to New App Novelty Syndrome more so than to anything else.

Moreover, it's ineffective because it doesn't convey what really makes the product difference. The copywriting hints at it, but doesn't go all the way. The best parts of Freckle are left hidden (partially because they weren't all there when we made this first site).

Lastly, and perhaps worstly, because it's embarrassingly bad and we perceived it to be ineffective, we didn't promote the product. And that's a kiss of death.

Why the old site had to die

1. Because it was ugly and useless, it was ineffective at promoting the wonderfulness of our product.
2. Because it was ugly, useless, and ineffective, we didn't want to waste many potential 'first impressions' on it.
3. Because we didn't want to waste first impressions, we didn't pimp it.
4. Because we didn't pimp it, it didn't get traffic.
5. Because it didn't get traffic, people didn't know about it.
6. Because people didn't know about it, they didn't sign up.
7. Because people didn't know about it, and most importantly because they didn't use it, we didn't get buzz.
8. Because we didn't get buzz, we didn't get serendipitous exposure.

And here the situation turns around and bites its own tail. Without serendipitous exposure, we didn't get signups, which would beget buzz, which… you see how this is going.

For want of a shoe, the kingdom was lost.

If you polish shit, all you get is shit that's shiny

Obviously a complete redesign was in order, and I don't mean just the visuals.

There were all these other problems. The content was insufficient. The content was badly written. The tone of the product site didn't match the app. It wasn't effective.

If that site were a house, it'd have an acceptable front façade, but only two outside walls and half a roof. And no furniture. And probably there would be ground squirrels colonizing the first floor and puffins eating away at the foundation.

And what can you do when puffins start eating away at the foundation of your house? You firebomb it and start over. Stinking puffins. It's impossible to get rid of them once they get settled in.

So I didn't start by analyzing the current site and figuring out how to improve it bit by bit.

You don't polish things that don't work at all, which were wrong in their very conception. Hello, shit polisher? Your deceptive local maximum is calling.

No, you replace them with something that's fundamentally going to work, and you polish that.

And to be able to build something that's fundamentally functional, you need to get clear on your goals.

Enough philosophy—Let's talk numbers

Our goal is to be "ramen profitable" on Freckle alone—for four people—by January 31st, 2010. Normal numbers just aren't going to do it.

To be "ramen profitable"—to be able to pay all our bills and be able to buy cheap blocks of instant noodles—we need to have about five times the number of paying accounts we have now, assuming the same distribution.

That means that we have to:

• not waste time and resources reaching people who won't like Freckle anyway (e.g. opposite of the Right People)
• reach our Right People more effectively
• convince more of our Right People to sign up
• convince more of our Right People to sign up for / upgrade to paying accounts
• convince more of those to use the service regularly, and keep their accounts open

Now that we have goals, the question becomes: how do we achieve them?

What happened next? How did I start? Did I make it?

Obviously the newly redesigned site is online now. You should check it out and see what happened. But there's more to the story.

Tune in for the next installment on Friday, where I'll explain how I broke down those goals into design requirements, and how we got from there to the current site. I'll also talk more about the numbers.

Don't miss the next post: subcribe to my RSS feed.

Feed reader averse? No problem, sign up for email delivery of new posts.

And, yes, obligatory modernity: If this stuff is up your alley, you should follow me on Twitter, too.