RefreshTheNet

from Car­son­i­fied Blog

Edi­tors Note: In his first arti­cle for Think Vit­a­min, Des Traynor – UX Lead at Dublin based Con­trast, out­lines some extremely use­ful post launch strate­gies for ensur­ing the suc­cess of your web app. You can hear Des share more devel­op­ment sto­ries at Future of Web Apps Dublin 2010.

Launch Time

Launch is a stress­ful time for web start-ups. It’s the cul­mi­na­tion of many month’s hard work, argu­ments, agree­ments, fea­ture debates, and design deci­sions. Some­one runs a deploy script and poof! You’re not in Kansas any­more, and you’ll cer­tainly need a brain, a heart and a fair share of courage.

Since most start-up advice cov­ers the period before you’re live, I thought I’d write about two of the most com­mon issues I see in web appli­ca­tions post launch.

Bud­get for Fur­ther Development

You’re plan­ning on being suc­cess­ful, right? One thing you should know is that the vast major­ity of devel­op­ment will hap­pen after you go live. Only when the rub­ber hits the road will you know what’s essen­tial, what’s a pri­or­ity, what fea­tures are “nice to have” and what ones are pre­vent­ing sign ups, and real rev­enue.

You need to bud­get for devel­op­ment that will hap­pen when you’re live, and you need that money in the bank when you launch. You can’t count on prod­uct rev­enue. You don’t have that much time.

Even if things are going really well and three hun­dred cus­tomers sign up for your app and even­tu­ally con­vert to pay­ing cus­tomers. These days you have to at least offer a thirty-day free trial to let cus­tomers find their feet. So on day 31, you’ll have three hun­dred cus­tomers ready to pay you their $19.

When you strip out fees, host­ing costs, etc, you’ll be left with maybe $5,000 to plan fur­ther devel­op­ment. That’s a long time to be run­ning your appli­ca­tion with­out any devel­op­ment, and $5,000 prob­a­bly won’t buy you what your three hun­dred cus­tomers have been scream­ing for all month.

Design still Hap­pens after Launch

At MIX10 this year, Bill Bux­ton spoke about the devel­op­ment of the Seat­tle pub­lic library. One point he made about the design process has mas­sive par­al­lels with appli­ca­tion development…

‘

“The design of the build­ing wasn’t fin­ished until the build­ing had opened and there were peo­ple using it, because there are some deci­sions that you can’t make prop­erly until you’ve seen the build­ing occu­pied, because you’re just not gonna get it right.”

Mar­ket research, wire­fram­ing, pro­to­typ­ing are nec­es­sary to get you a good start, but once you’re live your users will show you how it all should work. Hope­fully you got most of it right, but there will be changes, and you need to bud­get for them.

Where Rev­enue Comes From

The sad real­ity is that unless you’ve already amassed a sig­nif­i­cant fol­low­ing, through your rep­u­ta­tion, your blog, your pre­vi­ous com­pany, or by win­ning a com­pe­ti­tion, you most likely won’t have three hun­dred pay­ing cus­tomers on day one. Try three. But that’s okay, you’re just get­ting started.

A motto I drill into client’s heads is that web appli­ca­tions don’t sell them­selves. Even the beloved Net­work Effect still requires a net­work to begin with.

It’s easy to obsess over the one bug that stops one ancil­lary fea­ture work­ing in one ver­sion of one browser, or dwell upon one user who swears they’ll pay for Ning inte­gra­tion. This isn’t why you’re strug­gling to hit three dig­its. It’s just your scape­goat to hide a big­ger problem.

Num­ber of users, times value of plans equals rev­enue. Right? So all you need are more users and you’ll have more money. It’s sim­ple, and that’s the prob­lem with it. Say­ing “we need to increase users or “we need to increase our rev­enue per user” reduces you to the sleazy, pin-stripe con­sul­tant who chirps in with tru­isms like “guys, we need to move for­ward”, and “let’s aim for our deadlines”.

This sim­plis­tic way of look­ing at your appli­ca­tion leaves you with no next step. There is no lever called “more users”. There is no obvi­ous next step asso­ci­ated with “increase value”.

You need to focus your efforts. That’s why I draw this sec­ond dia­gram for clients (full size dia­gram).

This high­lights the real issues. It shows exactly what goes into a typ­i­cal cus­tomer acqui­si­tion. If you have some basic ana­lyt­ics tool installed, you can quickly quan­tify all of these num­bers and see where your prob­lem lies.

If you’re los­ing cus­tomers on day one of your free trial, you need to eval­u­ate what the first login is like, and talk to your customers.

If your ana­lyt­ics high­light a lot of errors at the sign-up stage, or your fun­nel view shows that very few get past it, then you need an eas­ier sign-up form, or more moti­vated users.

In my expe­ri­ence, it’s rarely one of the above prob­lems. It’s usu­ally far ear­lier in the process; no one has heard of you or your appli­ca­tion. In ana­lyt­ics speak, the num­ber of new unique vis­i­tors to your site each day is too low to see a solid increase in your user­base, and it itself isn’t increasing.

If/when we agree that’s your prob­lem, then we look at that in detail. This usu­ally involves a third drawing.

How can I put this in a way so as not to offend or unnerve? Basi­cally, start-ups all suck at marketing/branding/pitching and most of all, sell­ing. I don’t mean sell­ing up-and-to-the-right bar-charts to VC’s, I mean actu­ally sell­ing the appli­ca­tion you built to the peo­ple you built it for.

This is one prob­lem that you can’t code your way out of. There comes a point where cod­ing is pro­cras­ti­na­tion. Unless of course your busi­ness depends solely on your respect from soft­ware devel­op­ers, in which case pro­gram­ming is your best mar­ket­ing tool.

Shout­ing isn’t Selling

As I type this post, there are four peo­ple in my twit­ter stream “mar­ket­ing” their appli­ca­tion by post­ing tweets say­ing “guys, I’d love some feed­back” or “Check out xyz.com for all your xyz needs”. Using a link mon­i­tor­ing ser­vice like Bit.ly shows you exactly how inef­fec­tive this approach is. It’s the online equiv­a­lent of storm­ing into a night­club and scream­ing “Okay ladies, who wants a drink?”, and it’s about as effective.

By way of com­par­i­son, I often receive per­sonal e-mail from both friends and strangers ask­ing if I’ll try out their new project/application/piece-of-fun. These e-mails explain why they’re writ­ing to me, how they know me, why they value my opin­ion about this app. It’s clear the author has taken their time to speak to me.

These guys get a 100% click-through rate. Some of them will get a “looks cool, but I’m not your tar­get user”, some get tweeted, some get blogged, some will get a list of show-stopping issues, but they all get some­thing, because they showed me some respect.

Broad­cast mech­a­nisms such as tweet­ing, blog­ging, spon­sor­ing, adver­tis­ing, even speak­ing at a con­fer­ence are all great for mak­ing noise. But the value dimin­ishes very quickly after­wards. Hav­ing gen­uine con­ver­sa­tions with gen­uine peo­ple, is a far bet­ter approach to sell­ing your appli­ca­tion. It’s a far scarier prospect too.

It’s much eas­ier to hide between your tweets say­ing “I’m doing all I can, users just aren’t com­ing”. The dif­fer­ence between broad­cast­ing and com­mu­ni­cat­ing is that com­mu­ni­cat­ing involves answer­ing tricky ques­tions like “why is it so dear?” or “can it import Excel files?”

Lessons Learnt

You need to work hard for every user in the early days of a web appli­ca­tion, and as you get them you need to have bud­get and plans for what you’ll do to make them hap­pier as the rela­tion­ship grows. Get­ting one hun­dred pay­ing users is a huge mile­stone, get­ting to the myth­i­cal 2,000 takes patience and hard work. Act accordingly.

The rest is here:
Your Web App has Launched – Now What?