Reliable and in-depth data is critical for startup success. If you don’t know what’s working (and what’s not), you’re going to have to leave you success down to luck – which we don’t recommend.
We’ve been working with OLIO – a free app which connects neighbours with each other and with local shops and cafes so surplus food can be shared, not thrown away – since the beginning. A great example of a lean startup, OLIO have taken a pragmatic approach to analytics, building up a toolkit that allows them to collect the data they need, and use it to affect actual change.
Here’s a look at OLIO’s analytics toolkit…
Google Play and iTunes
Basic app data like number of visits to your app’s download page, number of downloads, country of download, number of updates, device type etc is provided by the stores OLIO is available on – iTunes and Google Play.
This information is collected by Google Play and iTunes as default and while it is useful to know how many downloads you have and if your app description is converting browsers into users, this data is essentially made up of vanity metrics. A handful of active users are much more useful to you than 1,000s of downloads and 0 active users. Even if you’re making money by charging for your app, if users aren’t engaging with it, word will get out.
To get a bit more information about engagement, OLIO needed to implement Google Analytics…
If you’ve ever used Google Analytics for a website, it works much the same way for an app. Google Analytics gives referral data (where is traffic coming from? How are social marketing channels/campaigns performing?), audience data and a real-time view of what’s happening in your app.
OLIO use Events in Google Analytics to track what users do within the app and whether new features are being used or not.
Google Analytics is great for collecting data on user engagement, but it can be pretty complex and doesn’t really allow you to actually do anything with that data, and so that’s where MixPanel comes in…
While MixPanel collects similar data to Google Analytics, it has a clear focus on people. Google Analytics collects anonymous data, letting you know how many people do which things. With MixPanel, you can track individual users, and groups of users, seeing exactly who’s doing what.
OLIO use MixPanel for evidence of engagement in their key areas. How many items are added within a specified time period, and who is adding them e.g. retailers or consumers? What proportion of items that are added are requested, and then picked up? How many items does an average person add in a month? What’s the overlap between people who add items and people who collect items?
On top of seeing what users are doing, MixPanel allows you to create event-driven funnels and then create and send relevant ‘engagement messages’.
One example from OLIO is when a user signs up, if they haven’t posted an item within a few days after signing up, they will automatically be sent an engagement email nudging them to try adding an item.
The one issue stopping OLIO from getting everything they need from MixPanel and Google Analytics is the challenge associated with location data…
MixPanel tracks location using IP addresses to look up where users are. If the user is on 3G or 4G, that could be routed through the data centre that the network provider (eg O2, Three, GiffGaff etc) is with. Therefore, instead of getting accurate location data for each user, you end up with clusters of users marked in Milton Keynes, Dublin and other data centre locations, rather than the location of the actual user. Not much use when OLIO is trying to measure how many items have been added and requested specifically in Brighton for example!
To get more specific location data, OLIO collects longitudinal and latitudinal data from users’ devices and then uses Google Places to convert these values into location data that provides the correct city and area of a user; this is then fed back into MixPanel via the MixPanel API.
A note on qualitative data
Although OLIO now has an analytics toolkit that lets them collect a whole range of quantitative data on their users, the value of qualitative data shouldn’t be overlooked. OLIO have a close relationship with their users, especially Ambassadors (volunteers) who work closely with the OLIO team to help spread the word about the app, and pass on feedback from users. And each of the OLIO team members uses the app several times a week themselves, so that way they get to experience their product first hand, as well as meet users directly and get their feedback.
It’s absolutely fundamental to have a good analytics toolkit in place in order to collect data on your users from the moment you launch says Tessa Cook, co-founder of OLIO. “Often, in the race to get launched, analytics is abandoned. However the minute you’re launched, you need access to data, and implementing it retroactively is almost always a painful and inefficient process, so it really is best to design analytics into your product scope right from the beginning.”
If you need any help with analytics for your product, get in touch with Simpleweb today.