For Mobile, Is Google Analytics Enough?

Rob Anderson / 20th June 2019 / Comment / Analytics

hand holding a mobile phone

When considering analytics tools for your business, website, and digital marketing efforts, it’s important to understand the benefits and limitations of (and alternatives to) the industry standard, Google Analytics. Let’s dig in!

Google Analytics is Free, So Why Should I Pay for an Analytics Tool?

One common critique is that Google Analytics focuses mainly on site traffic, so when you need to analyze your marketing efforts, track events, and analyze your sales funnel, Google Analytics may not be simple enough to understand and use. And depending on the application, it may have difficulty in delivering the needed actionable data for marketers.
While it does support events and goals, users can find it complicated to set up and analyze, and eventually give up, even though the program is installed.

Legacy Versions of Google Analytics For Mobile Apps is Getting Shut Down

If you have a free Google Analytics property collecting hits exclusively from the Google Analytics Services SDK (Android or iOS), know that Google is starting the process of deprecating “legacy” Google Analytics for Mobile Apps. The following will take place as a result:

    • In 2019, Google will begin to decommission properties that receive data exclusively from the Google Analytics Services SDK.
    • Data collection and processing for such properties will stop on October 31, 2019.
    • Reporting access through our UI and API access will remain available for these properties’ historical data until January 31, 2020.
    • After Google’s service is fully turned down, these properties will no longer be accessible via our Google Analytics UI or API, and their data will be removed from Google Analytic Servers

Things You Can’t Do in Google Analytics

Closely Track Individual Users

Google Analytics only allows users to use a unique user ID and prohibits sending personal information such as user name or IP address. So one can’t really see and understand how specific users behave on the site, or how one acquired them. (Although depending on the privacy regulations – e.g., GDPR – of your target market, this might be seen as a Feature rather than a Bug)

Focus Tightly on Events

Without a well-defined event naming taxonomy, Google Analytics can make it difficult to understand website events – both in terms of who’s using them and why they are being used or not. Using tools like MixPanel or Heap Analytics makes it much easier to focus on events and can help hone the approach to optimizing the site and marketing efforts.

Get Product Support

The only way you can get full customer support from the Google Analytics team is to pay for the premium version of Google Analytics (GA 360), which will grant you access to more features and deeper personal support options. If one is using the free version of Google Analytics, there are various resources available for support, but not having access to Google as a support option is something to take into consideration.

Alternatives to Google Analytics – A Brief High-Level Comparison

Define screen views, events, and conversion tracking, deploy code updates to application codebase for tracking. Treats Users and Events as separate metrics for analysis. While providing more flexibility to analysis itself, it does make simple exploratory analytics more difficult to do than what you might be used to using with the Google Analytics UI and reporting API.

“I like free tools and don’t mind setting up the integration in my app. I’m comfortable understanding the data model and doing the analysis I need in the Firebase UI. I understand there are limits to what I can collect.”
Free. The Firebase team has been working hard to make the user interface more intuitive, so usability is something they are certainly paying attention to.

Define screen views, events and conversion tracking, deploy code updates to application codebase for tracking. Associates individual Users with their in-app Events and allows for a deeper dive into session activity.

“I want to define a key limited set of conversion events, dive deeper into user activity, and send push notifications and run A/B tests within the app using my analytics tools and funnels.”
Priced per Event hit sent and data retention window. The fewer core Events that you need to track, the less it costs.
Automatic Data Capture – after deployed in an application, tracks every click, tap, change, swipe, and page view. Events can be defined in Event Visualizer tool and grouped into funnels; event tracking definitions are applied retroactively to data captured in Heap.
“I want to track EVERYTHING, then decide afterward how to align the data with events, users, and conversion funnels. If something in my app changes, I’d like to update tracking and have it apply to historical data.”
Priced per Session. Good for apps that need to track a lot of data

Conclusions

  1. The trend line is clear – as digital marketers, we can’t just ignore mobile and focus only on the web.
  2. Mobile presents measurement challenges – the nature of mobile apps and application development may require a shift in thinking on how we approach analytics as a discipline.
  3. A strong strategy in mobile has clear rewards – the potential size of the audience alone has enough reward to justify spending time changing our approach.
  4. There’s no such thing as a free lunch – free tools may not be the best tools anymore when it comes to mobile apps.
  5. Start with a decision tree when comparing approaches – the table below is an example of how one can start to structure the conversation to begin making strategic choices around analytics for mobile:
A:  Decide What I Should Be MeasuringB:  Align on Methodology for TrackingC: Choose Appropriate Analytics for Tracking
How Do I Want to Track:How Should I Track:Taking Into Account:
  • Usage
  • Retention
  • Engagement
  • Conversion
  • Sessions
  • Users
  • Events
  • Does it Align Philosophically with the Choices in Columns A and B?
  • Organizational Cost/Budget Alignment (Pricing Model)
  • Ease of Deployment into Application (Engineering Costs)
  • Ease of Analysis Features
  • Integrations Within the Target Application (A/B Testing, Push Notifications, etc.)
  • Integrations With other Third Party Tools (Google Data Studio, Tableau, 3rd-party APIs, etc.)

Usage Trends Across US Digital Landscape

Mobile Continues to Erode Desktop Time Spent

Time on Mobile is up 31% YoY; Desktop dropped 2% in the same period:

Three-quarters of all Internet Time is Spent on a Mobile Device

In 2018, Mobile increased share of time spent to 74%–up from 68% in 2017
Desktop lost share, dropping to 26% from 32%

Cross Device and Mobile continue to erode Desktop

Product Health – Measure What Matters

What Should I Be Measuring? How Am I Doing Compared to Industry Standard?

    • Usage
    • Retention
    • Engagement
    • Conversion

Usage – Are People Using My Products?

Methodology – Track Daily Active Users (DAU) and Monthly Active Users (MAU), then Benchmark these Metrics over Time

Count a user as Daily Active if they record an action in the product over the course of a given day. Requiring a user to perform a specific action (not just opening the app) to be counted in DAU/MAU is recommended – if a user is not performing activities that drive value to the business, they should not be counted as active.

Calculating average DAU over the course of a month divided by MAU gives a percentage of how active a product’s user base is. For example, let average DAU be 120 and MAU be 2000. We then divide 120/2000 to arrive at an active user base percentage score of 6%.

Industry benchmark studies show that 90% percentile (elite) products have an incredibly high active user base score.

Average products (50% percentile) have a 5% or higher active user base percentage score.

source: 2017 Mixpanel Product Benchmarks Report

The active user base percentage can see high variation depending on the industry – One will want to benchmark these numbers within a specific vertical (E-commerce, SaaS, Media, etc.) to determine how active a user base is vs. the competition.

Tracking average DAU growth month-over-month is recommended. (Are more people using the app this month vs. last month? Is usage growth flat? If so, why?)

Retention – Are People Coming Back?

Methodology – Track Users that Return and Perform Another Action

Users that return week-over-week and continue to engage in activity are valuable and demonstrate continued long-term value for the application.
As with Usage, weekly retention is highly variant depending on the industry.

source: 2017 Mixpanel Product Benchmarks Report

At 8 weeks, most retention rates decrease to the point where attracting new user growth (DAU) becomes important.

Engagement – How Are People Using My Products?

Methodology – Track The Sum Total of All Actions Performed

Track the average baseline daily actions performed in the application, then compare for a given day how the average actions deviate (+/-) from the established baseline. This will give a picture of how consistent application usage is throughout the week, or if certain days experience lower activity (Fridays, Weekends). Different application verticals have a different expected baseline for activity deviation and dropoff.

If there are certain days where activity is high within the application, these days are good times to target users with campaigns to drive the activity even higher. If there are certain days where activity is particularly low, these might be good times to see if you can get users to engage. You will also want to see if there is any correlation between application activity with conversions (more on this in the next section).

source: 2017 Mixpanel Product Benchmarks Report

Conversion – Are People Doing What We Want Them To Do?

Methodology – Track The Key Goals (KPIs) Performed

Typical Use Cases:

    • For Product Managers – Track and analyze the usage of different features, funnels, and individual user behavior.
    • For Marketers – Analyze Marketing and user acquisition channels by the main KPIs (for example, which content/ campaign/ landing page leads to the best sign ups), and optimize funnels.
    • For Support/Customer success – Track individual user journeys to understand their behavior better.
    • On the Company level – Track the main KPIs (account sign-ups, paid membership engagement) and their trends.

source: 2017 Mixpanel Product Benchmarks Report

At the Company level, all app activity should roll up into distinct Conversion funnels and events. The frequency of conversion can vary based on vertical. The easier it is to convert, and the shorter the funnel, the higher % for conversion rate.
(e.g., Media apps where the Conversion event is defined as “Play Media” are over 35% more likely to convert compared to an E-commerce app where a Conversion is a Purchase event)

Methodology – Track Users that Repeat Conversions

Users that return to the app and convert again are valuable! We want to track how we acquired those users and provide them with messaging that encourages them to continue to re-engage with the app and convert on a frequent basis.

source: 2017 Mixpanel Product Benchmarks Report

Methodology – Track The Time It Takes Users to Convert

The fewer barriers we can place on a user before they convert, the more likely it is that they will end up converting. Even things like screen load speed, network connectivity, and ease of application use can have an effect on time to conversion. A well-thought-out analytics strategy can identify where users are encountering difficulty in conversion. Once an area of difficulty is found, different approaches in messaging can be utilized for the user, application changes can be made, or A/B tests can be conducted to improve conversion rate.

By Rob Anderson