Launching a New App? What Metrics Should You Track to Be in Control?

At this stage metrics like churn rate, recovery rate, lifetime value mean pretty much nothing to us (but if you are interested in them, here is an article that does them great justice). We need to count on metrics that help us convert those first users into customers. More than anything we need hints on how to make them love our application.

Track down those pesky errors

Like it or not, users will encounter errors when dealing with any app, especially if it is new one. Nobody likes errors, especially app users. At this point, reaction time and communication are crucial. If you are notified in real time about possible errors, you will react accordingly and try to fix it asap. It’s even better to communicate with your users via e-mail and let them know you area aware of the error and it will be fixed asap.

Being proactive is something we can afford at this stage and users love it.

Here is how I usually deal with it: setup an e-mail account with the sole purpose of receiving e-mail errors. I then forward all those e-mails to my own account. On my smartphone I use an e-mail push app that will notify me with a splash screen and sound signal each time an error occurs. So, no matter if I am in front of my e-mail or not, I am aware of the errors going on in my app in real time.

In time, the number of errors will grow and this will overwhelm you. You’ll move on to better error tracking methods, but for now, it’s the easiest way to track the errors in your app and react quickly to them.

Completed actions, just as important as errors

As the first users are getting welcomed by the application they’ll need to go though a number of steps and actions to see the benefits of using it. Without getting through them, they will never become potential customers, so tracking how they are dealing with each step or action is vital.

The easiest way to deal with this is to setup a Google Analytics funnel for the process. We’re taking a bit more complex approach, but it pays off: as user actions are happening in more than one visit session, we setup logs in our app that save each user action.

While the first version is easier to implement, the second allows us to check the actions for each user so, if he will contact us via support or sends a tweet or visits our website from a different device, we’ll already have a nice history of his actions.

Based on the logs we can then generate charts like the following, that give us a broad look on actions taken by the users: img inn

Action intent, true feedback

We are big fans of the Minimum Viable Product model, though Rand from moz.com got us very curios about the Exceptional Viable Product model. This means we launch products not when we believe that particular products is ready, but when we think that, even if not ready, people will find it useful. Our users will be in charge of how the product will be shaped from the launch moment and on.

One of the best way to do this is by using action intent hints. I place links and buttons inside the app that lead users to functionalities that are not yet developed. Users are often welcomed with messages like this:

alert_feedback

I then count how many people clicked on the button to measure their intent of using a certain undeveloped functionality and check the feedback they leave for it.

With such data, it is much easier to schedule future developments that that people want, based on actual facts..

E-mail response rate

Depending on the app you build, it will send more or less e-mail notifications to users. What I do is always go through a 3rd party system that sends the e-mails for me. This way I’m making sure they don’t reach the spam folder and I don’t lose any ime with it.

I tried different e-mail services out there like SendGrid and Postmark but ended up using Mandrill. They offer the MailChimp quality, it is free for up to 12,000 e-mails per month and they report open rate and click rate for the e-mails, per user.

If you send e-mails that are not opened or acted upon, better stop doing it and try other timing or different content/messages. Ultimately, e-mails will play a huge role in getting users back to your app.

Visits and pageviews

The only time when visits and pageviews are important is when there are only a few people getting to your app. You can’t have conversion rates and thus profit if you don’t have visits in the first place.

I’ve done this mistake in the past of focusing only on conversion rates and not on increasing the number of visits, especially organic traffic. In the end there is only that much you can do to optimize your conversion rate out of 500 people visiting your website per week.

Paid traffic is an option, but not for all, as budgets are probably limited. As time goes by, the number of visits will become a vanity metric and organic growth will take its place as the KPI to look for.

Your metrics?

Early startup metrics are amazing because you literally have the time to act on each of them by picking up the phone or starting an e-mail and get in touch with the user. In fact, the success rate of getting the first customers always depended on our reaction to each of the above metrics.

Are you just starting a web app? What metrics are you focusing on at this stage? Is your web app well established? What’s your advice for the ones starting now?