Adding and Measuring the Impact of Mobile Apps

MobilePhoneMeasureSmallFor week 7 of analytics, we focused on how to measure the impact of an app — be it for entertainment, practical or promotion purposes.

Given the development work behind each app, I think it’s particularly important to measure their analytics so that you can course correct, as required, to maximize the investment. This means using app analytics to learn: who downloads the app,  how they discover it  and how frequently they use it (if at all).

Although app analytics measures many of the same metrics as web analytics, like number of visits and sources, I thought it was interesting to note some of the subtle differences between the two.  For example, I noticed that app analytics metrics include:

  • Screen resolutions (as well as screen size and orientation)
  • Varied types of mobile devices used
  • A stronger focus on groups of users — including user’s friends who also use the app and ‘cohorts’ (i.e., segments of customers that downloaded the app during a common time period).
  • “Usage” instances and frequency, since downloads/installs don’t correlate with engagement/use.

Like web analytics, you need to define metrics and KPIs. For example, you need to define an ‘active’ user’s level of engagement (frequency, duration).

We also looked at various tools for measuring app analytics. However, Sofia pointed out that the metrics they measure are not evenly distributed among the categories of apps available. Specifically, current tools are more focused on marketing metrics, like search engine keywords used to discover them, versus content metrics, such as the app’s longevity/life cycle and number of app/widget installs a user has.

Of the tools we looked at, I particularly liked Flurry Analytics because it provides benchmark measures and several distinct metrics, such as user personas. However, it didn’t look like it provided search engine keywords, which you needed to measure through other tools, such as App Annie, as its Key word ranks feature.

For the exercise, I selected a made-up ‘package ingredient analysis app’ promoted through an Instagram campaign. I identified KPIs and measuring tools as:

  • „Number of active users (Flurry- Event- Use)
  • „Average increase in users (over 2 week time periods) – (Flurry – Funnel Analysis)
  • Number of people using the app at 2x per week or more (Flurry – Retention – Return Rate & Event – Use)

I identified metrics and their tools as:

  • „Search engine words used to find app (App Annie – Key word ranks)
  • „Number of installs (Flurry – Funnel –Downloads)
  • „Number of users who have installed the app vs. Instagram users (app installs and Instagram audience)
  • „Personas using app (Flurry – Audience Interest – Personas)
  • „Use of this App compared to other lifestyle apps (Flurry – Audience Interest – My Apps Interest)

We also explored and discovered several other worthwhile app analytics tools, including:

  • Amplitude – Focuses on user engagement, including: segmentation, cohorts and churn.
  • AppSee – Indepth analysis of users and everything they do with app (for a steep price)
  • Ninja Metrics – Focuses on customers and is well suited to game apps
  • Localytics – Unique tool that integrates analytics with marketing to help produce push campaigns

This class opened a new world of analytics — and there may be an app for that too.