Getting To Know Our Users


Warning: Illegal string offset 'status_txt' in /home3/buzznova/public_html/entertailion.com/blog/wp-content/plugins/share-and-follow/share-and-follow.php on line 1243

Warning: Illegal string offset 'status_txt' in /home3/buzznova/public_html/entertailion.com/blog/wp-content/plugins/share-and-follow/share-and-follow.php on line 1243

Warning: Illegal string offset 'status_txt' in /home3/buzznova/public_html/entertailion.com/blog/wp-content/plugins/share-and-follow/share-and-follow.php on line 1243

Warning: Illegal string offset 'status_txt' in /home3/buzznova/public_html/entertailion.com/blog/wp-content/plugins/share-and-follow/share-and-follow.php on line 1243

Warning: Illegal string offset 'status_txt' in /home3/buzznova/public_html/entertailion.com/blog/wp-content/plugins/share-and-follow/share-and-follow.php on line 1243

Warning: Illegal string offset 'status_txt' in /home3/buzznova/public_html/entertailion.com/blog/wp-content/plugins/share-and-follow/share-and-follow.php on line 1243

Warning: Illegal string offset 'status_txt' in /home3/buzznova/public_html/entertailion.com/blog/wp-content/plugins/share-and-follow/share-and-follow.php on line 1243

Warning: Illegal string offset 'status_txt' in /home3/buzznova/public_html/entertailion.com/blog/wp-content/plugins/share-and-follow/share-and-follow.php on line 1243

Knowing how well our projects perform is very important to a small startup like ours. To keep our costs low, we need to know as early as possible which ideas don’t pan out. We want to follow a strategy of determining from real users what features they want. We are just beginning to understand how we could achieve these goals using various tools.

For marketing our first app, we created a website which was hooked up to Google Analytics. Although we aren’t getting a lot of traffic, the tools Google provides look very promising for tracking our web site’s visitors. The analytics dashboard displays data using graphs that can be drilled into and filtered based on various criteria. The data show that we are getting returning traffic which is a good sign. Also, most of the traffic has come from LinkedIn, Facebook and Twitter, so the little bit of social marketing we have done has worked. For the company blog and the human quirks project we used a WordPress plugin to hookup Google analytics too. We’ll stick with this until we understand our needs better when we hopefully have much more web traffic.

To promote the human quirks website, we used some free vouchers for Google AdWords. This service is somewhat intimidating with lots of terminology and settings. We used the keyword tool to find some keywords to bid on and created a campaign with a fixed daily budget. We created a simple ad and went mostly with the default settings, in particular let Google handle automatic bidding. We haven’t had a single click on our ad yet, but since the human quirks contest runs over a month, we’ll wait a while before we might need to tweak the ad to drive more traffic to our website.

The RSS feed for the company blog is hooked into FeedBurner to track subscriptions. We have very few subscribers. At the moment it seems that most of the feed traffic are coming from bots for various search engines and blog directories. A nice feature that FeedBurner provides is to allow email subscriptions; we don’t have anybody on the email list yet.

For tracking usage data about our first app we use the Android Market publishing website. The site tells you how many apps have been downloaded and how many apps are currently installed. It also tells you if any errors occurred, but we haven’t had any yet. The data seem to be only updated on a daily basis. We had an initial spike in traffic which surprised us since we didn’t expect a lot of users for our little game. We also got a comment from a user which indicated that our backend had gone down. It made us scramble to make our backend scale better. Overall we are not very impressed by the site; it seems to just have the bare minimum data about our app and doesn’t give you any deeper insight into the kinds of users or devices that are running our app.

The backend of our game is implemented in Google’s App Engine. We picked the Google App Engine since it comes with free quotas which we thought would be more than adequate for our little Android game. The App Engine provides an online developer dashboard with data on quotas, logs and administrative settings. Unfortunately, it does not provide a notification mechanism when any free quota limits are reached. Also, the logs do not give us enough insight into what the user is doing. We found several cases where the logs would not be updated correctly. We could add more detailed logging to our app, but the dashboard doesn’t provide good tools for visualizing that kind of data.

Since the game is free, we decided to include AdMob advertising to generate revenue. AdMob provides a website to track the income generated by the app. We’ve only made pennies so far. The website provides data about how many impressions and clicks occurred. It also provides basic geographic data about users. However, this is nowhere near as detailed as with Google Analytics. We are not impressed with the data we are getting; we are looking for much more detailed information on our users and on what they are doing within the game.

We found some blogs online recommending Flurry analytics. The developers on those sites used it in their games and got better data than with AdMob. Flurry has two components: a free analytics service and an app advertising and promotion service. What makes their analytics service particularly interesting is that you can track custom events in your app. For example, we can create our own events for when the user plays a game, when they guess the answer correctly or even when they reach a new high score. This kind of granularity isn’t available from AdMob. We’ve done some developer testing and the results look very promising. Their online UI is also very nice and allows you to explore the data from various angles. You can also download the data as CSV files.

So we have decided to replace AdMob in our next app update with Flurry analytics and their advertising. We’ll report back on what we have learnt about our users.

Image courtesy of Flickr user nurpax.

Tags: , , , , ,

Leave a comment