Mobile App Retention Metrics

We have written extensively on this blog about acquiring organic traffic, mobile app visibility and converting app store views to installs and users.

One of the key drivers of the value of these users is how long a user keeps using your app – mobile app retention.

Retention metrics you should be using

Many of the metrics used to track app retention are associated or impacted by changes to engagement – meaning, a measurable improvement in user engagement will often improve retention as well.

If retention measures – at a high level – the longevity of product use, engagement measures the depth of product use.

Retention metrics such as cohort-based Day 1, 3, 7, 14, 30 and 90 can help us determine if efforts to keep users coming back are having a positive effect.

Local and push notifications are the tools of choice for many mobile app marketers, but don’t discount great design and perceived value.

Engagement metrics, therefore, provide the insights for improving retention.

What are users engaging with inside of the app, what are they missing and how can we provide more of what they want, or even encourage use of where we think the value is?

Retention metrics examples

Setting key performance indicators (KPIs) for a mobile app should be tied to business goals.  There is not a mobile app template for all mobile apps as Clash of Clans and the Target app are very different with different goals and different ways of measuring the revenue generated from the mobile app.  They really only share a platform (mobile OS and a mobile device).

Assuming goals are tied to revenue, and by extension lifetime value (LTV), finding the cohort that demonstrates the highest LTV informs the rest of the mobile app’s KPIs including retention.

Because retention metrics don’t provide a detailed view of in-app activity (that usually falls under engagement), more (longer) is almost always associated with a higher LTV.

Overall for both Android and iOS apps across all categories – D30 retention tends to be 25-40% of D0 users, with a mere 1-4% at the end of a year (D365).

Mobile app retention statistics

The most common method for tracking retention is Day N retention. What percent of your users opened the app on Day N?

If you have 100 installs on January 1st, 40 open the app Jan 2nd and 20 open the app Jan 8th – for the Jan 1 cohort, your D1 is 40% and D7 is 20%.

Mobile app marketers can look at aggregate data, but reviewing by cohorts tied to new versions or other retention optimization efforts will provide the insights you need to measure the effectiveness of such efforts.

Attrition metrics

A key SaaS metric is churn.  Churn measures what percent of users stop using (and paying for) an app (or service) for a given month.

Consider a scenario where a business has 100 active users paying monthly, are signing up 10 new customers per month, and have a 10% churn rate.

Even with 10% monthly new customer growth, the business is flat as they are losing 10% of existing customers to churn.

The subscription model in mobile apps has not yet become a prominent monetization model with only a few standouts including Pandora, Spotify, HBO, Kylie Jenner’s app.

Most apps are free to play, with some small percent of total users purchasing in-app items. An even smaller percent of users account for the majority of revenue in most apps with some studies showing 60% of revenues coming from only 10% of users.

What that means for attrition metrics for mobile apps is it is easy to dismiss users that churn as those unlikely to make an in-app purchase.

This may be true – but really depends on the type of mobile app.  Shopping apps specifically should have a much longer goals for retaining users than a mobile game for example.

The trend by cohort analysis on D90 retention for non-gaming apps should be a primary attrition metric.

Target and Lowes both shared that customers with their mobile app installed shopped more frequently and purchased more per visit than non-app users.  Keeping these users on your app for 90 or 365 days can have a big impact on store revenues – where many games are making money on Day 1 and much less per user on D90.

Tracking retention

The most popular free and paid app analytics services all have default retention tracking. Apple’s analytics module tracks retention for iOS apps, but not Android. Flurry provides a good, free solution and then Localytics and Mixpanel have complete app analytics offerings.

Improved retention usually means improved LTV, which makes each new acquired user more valuable and allows for increased investment in both paid and organic user acquisition efforts.

The Mobile Impact on Holiday Shopping

As we leave for our holiday vacations, let’s take a look at how mobile and mobile apps have not only nudged their way into retail, but how mobile shopping is now leading much of the shopping experience.

We are not talking about mobile or apps taking over only ecommerce (which they are), but the huge impact mobile is having in “brick and mortar” – meaning the impact of mobile on all retail sales.

Until recently, there was a clear divide between ecommerce sales and “in store” sales.  No one was opening their Mac in Best Buy purchasing online from Best Buy while in the store (maybe someone has done that – but not a significant use case).

With smartphones, purchases can be made digitally and in the store.  As we’ll see, this happened with greater frequency in 2015 than ever before.

We’ll look at ecommerce vs in-store, and by medium of PC, Smartphone or at a retailer.

Several companies have shared their holiday shopping data including IBM via Watson, and Adobe via their Adobe Marketing Cloud. Then we have forecasts and other informed data points from consultancies like Deloitte.

Black Friday

IBM and Adobe shared data for Black Friday which is “traditionally” a holiday shopping day on the friday after Thanksgiving.

In 2015, Black Friday started on Thanksgiving, with digital sales increasing 26% over 2014, with 60% of traffic and 40% of sales coming via mobile devices.

On Black Friday, mobile accounted for 57.2% of web traffic and 36% of online sales, an increase of 30% over 2014. The average order on a Tablet actually exceeded PC average orders sizes for the first time.

Source: IBM Holiday Benchmark Study
Source: IBM Holiday Benchmark Study

Data shows that this spike is not due to traditional retailers selling in-store, with mobile-first or traditional online sellers (like Amazon) selling online and via mobile.

Of the top 5 ecommerce revenue generators on Black Friday, 4 were traditional retailers.

Source: infographic from Signal.co
Source: infographic from Signal.co

Adobe’s data mirrored IBM’s, but they also shared that of mobile revenues, 70% were from iOS devices.

Note that in the US, the market share by mobile OS is around 50% Android, 40% iOS and 10% Blackberry, Microsoft and others.

Cyber Monday

The same thinking that has brought us Grandparents Day and Boss’ Day has brought us Black Friday, Small Business Saturday and Cyber Monday.

For Cyber Monday 2015, retail online sales increased almost 18% from 2014.

Mobile accounted for 50% of traffic and 30% of sales on Cyber Monday – an increase of more than 25% year over year.

Mobile Purchasing in Retail Store

Customer service consultancy and software platform provider Eptica shared the results of their multichannel customer experience study in retail that showed 23% of respondents completed a purchase from the store they were in, on their mobile device.

This could be related to not having to transport the purchased goods home, to see a wide range of reviews or ratings, a long checkout line or price shopping.

In the same study, a full 20% responded purchasing from a competitor while in a retail store.

Best Buy has long been the poster-child of the challenge of “Showrooming” – where consumers use a retail location to touch and feel merchandise, only to go home and price shop.

Apparently, consumers are not even leaving the retailer before placing a purchase on their smartphones.

Evaluating an App Marketing Promotion Company

With more than 1.5 million apps on each of the Apple and Google app stores, building a plan for visibility for your app is essential to a return on your investment.

Since the majority of mobile app downloads come from app store search, a mobile app marketing plan should start with optimizing for organic discovery in the app stores.

App marketing strategies that include app listing metadata, optimized creatives, ratings and review acquisition and even in-app retention and engagement are collectively generally referred to app store optimization or ASO.

ASO is often, mistakenly thought of as “finding keywords” for an app’s app store listing. This limited approach is still employed by some app marketing companies, but is not the comprehensive approach needed to compete and thrive in the app stores.

What follows are the building blocks of a mobile app marketing plan that an app marketing promotion company should be required to deliver.

Build Metadata for App Store Listing for both Apple’s App Store and Google Play

The metadata in an app listing includes the app name and description, a keywords field for Apple and a short description for Google Play. These fields impact how your app is indexed by Apple and Google, which determines which search results your app appears.

The goal should be broad coverage of extremely relevant search terms.

There are three primary tool for defining and measuring what is relevant and existing coverage of these terms:

1)  Historical and trending app store data

App store data provides insights to how have users searched the app stores for related apps, similar features and benefits, and what is currently trending.

We have written about this extensively, how users search the app stores is very different from how users search the web.

2)  Focus groups

Focus groups can provide both quantitative and qualitative feedback on your app’s features and benefits, which resonate best with your target audience and how they would search for these features in their words.

3)  App ranking analysis for relevant phrases

How your app ranks for specific search terms and phrases can tell you a lot about how the market is reacting to your app for that term.

Your app could be the very first result for “Free Zombie Game”, but if it is a flashlight app, conversions from users seeing your flashlight app result when searching for zombie games will be extremely low.

This signals Apple and Google that your app is not relevant to the search, and they will drop you in the rankings for that search.

Relevance matters to users, and is easily measured in app stores by conversion rates.

Note that there are several factors that go into not only the indexing but ranking of an app, but conversion and ranking provide clues as to how the market is reacting to how you are positioning your app.

Tested and Optimized Creatives

The icon, screenshots, video and even order of the screenshots can have a surprisingly huge impact on app installs. Testing these creatives with a target audience – ideally a focus group – should be part of any app marketing promotion company offering.

This is a chance to not only test a wide range of directions without affecting actual downloads, but also provides a channel for receiving qualitative feedback.

One design, color scheme, or text on a screenshot may perform much better or worse than others – the data tells you this.

Qualitative data provides access to why, giving the publisher (and designer) a chance to better understand their target audience and adjust designs before publishing.

Even when testing minor variances in icon design, or the text used or order of screenshots on a published app store listing, publishers can see 100%+ improvements in conversions.

Google Play provides multivariate testing to Google Play app publishers (Apple does not), which allows for experimenting with smaller changes on a subset of app store traffic to try and attain statistically significant, incremental improvements to conversions.

Produce Measureable Results

Companies have different goals for their mobile apps. No matter if your goal is a measurable ROI, more emails, better engagement with your brand, customer service or anything else, goals should be defined and measurable.

App promotion companies that deliver measurable results generally assist with defining and testing differentiating features of an app, help acquiring ratings and reviews, advise or build a strong presence outside of the app store and more.

You can read more on our approach to app store marketing here, or request a demo on how Gummicube uses app store data to grow our clients’ mobile portfolios.

Deep Linking and iOS 9 – How Deep Linking Affects Discoverability

Apple’s latest mobile operating system, iOS 9 introduces exciting new features not only for iPhone and iPad users, but for developers and marketers as well.

Siri and Spotlight now index and include mobile apps in search results. Apps are indexed not only by the metadata in the app store listing, but also metadata used for specific in-app content.

A user searching Safari, Siri or Spotlight for a taco recipe could see taco recipes from apps already installed on their device, and tap through to the specific location in the app.  In-app content is now accessible, instead of results being limited to websites or the top level of an app.

ios-9-search-results

We have seen some of this in-app content being indexed and discovered in Apple’s recent presentation of the AppleTV. The user searches for a movie, and results are apps with a direct navigation to the movie in the app.

apple_tv_lg

For “Ghostbusters” on the Netflix app to appear in AppleTV search results, Netflix added metadata to that specific piece of content within their app.

From a user perspective – locating relevant apps and navigating to the specific content in the apps is drastically improved user experience.

From a publisher perspective – indexation of apps and in-app content is not limited to the app listing metadata – which has primarily been weighted to app name and the 100-character keyword field.

The long-tail of search is finally here for mobile apps.

Optimizing your app and content for iOS 9 search can impact discovery, engagement and retention – which represent some of the biggest challenges for app publishers and marketers.

From Apple:

Search in iOS 9 gives people great ways to access information inside of your app, even when it isn’t installed. When you make your content searchable, users can access activities and content deep within your app through Spotlight and Safari search results, Handoff, Siri Suggestions, and Reminders. Making your content searchable helps you enhance the user experience of your app and improve its discoverability.

Introducing the Apple’s iOS Search API

A mobile app is indexed by Apple for the App Store using the app name and keywords field.

Description, ratings and reviews, category and related keywords also impact indexation (the search results an app is included in), but the largest weight for how an app is indexed is the app name and keywords.

For many apps, there was no way to capture all the content in the app that may be relevant to a search in just the app name and keywords field.

A search in Safari, Spotlight or Siri for “When does the new Star Wars movie play near me?” would take a user to a website, even if they had Fandango, IMDB or any other movie tickets app on their phone.

Same for any app with lots of content like cooking, recipes, travel, real estate, news – the list goes on.

How do Fandango and others enable Apple to index their in-app content? The Apple iOS Search API.

Apple has created two indexes:

  • A private index of on-device activities and user created content
  • A public index of publicly available data

For now, this is an important distinction to understand when contrasting with how Google (Android) approaches indexing in-app content.

This topic can get technical quickly, so let’s focus on the most important opportunities for marketers:

  • Core Spotlight – for indexing in-app content, and
  • Web Markup/Universal links – for syncing and indexing content that is also available on your website

Core Spotlight

Core Spotlight helps you make items searchable in the private on-device index; you don’t use Core Spotlight APIs to make items publicly searchable.

Apple uses the example for a news app:

  • Index the most recent, trending, and relevant articles and add attributes such as location or topic
  • Index articles the user marks as favorite or for reading later
  • Remove articles from the index after a period of time to keep the index fresh

Web Markup/Universal Links

Web markup allows the Applebot web crawler to index your content in the public index, which makes it available in Spotlight and Safari search results.

Adding universal links routes users to your app from search if they have it installed, and to your website if they don’t.

Apple has created a website for validating iOS 9 Search readiness called the App Search API Validation Tool.

ios-9-search-api-validation

You can then convert mobile website visitors to app users with an app banner prompting visitors to install your app.

Here is an example of an app banner using Branch.io’s technology:

smartbanner-branch

How Apple Ranks Content

If the rise of the app stores has taught us anything, it is that “Googling” or web search is not the only or even preferred way to search.

Apple, Facebook and Google all are invested in creating search channels that provide relevant results regardless of where the content comes from.

Here is Apple’s specific statement regarding how they rank content from various sources in search results:

iOS 9 users expect app search to give them the most relevant results, regardless of where the content comes from.

To deliver the best user experience, iOS determines the quality and ranking of a searchable item by measuring:

  • The frequency with which users view your content in the app

  • The amount of engagement users have with your content (determined by the engagement ratio, which is based on the number of times users tap an item related to your app and the number of app-related items that are displayed in search results)

  • The popularity of a URL in your website and the amount of structured data available

Indexing in-app content, syncing in-app and web content, deep linking, universal links, ranking specific in-app content by engagement all provide significant opportunities for those who invest in optimizing their mobile apps for this new mobile search paradigm.

Holiday App Marketing Tips

The holidays are upon us, with “Black Friday” and “Cyber Monday” pointing to a very mobile holiday season.

Data from IBM and Adobe showed web traffic, sales and even in-store purchases significantly impacted by mobile and mobile apps.

The end of the year and January are always very big for app publishers as new smartphones are received as presents, and down time with family is time spent on “the second screen”.

So how to make sure your app is well positioned for the holidays and the January spike?

Don’t Neglect Push Notifications

One of the huge advantages of acquiring users for your mobile app is the ability to reach them with notifications.

Notifications can be automated via some in-app trigger (local) or sent manually remotely (push) and are the tool for engaging with your app audience.

One reason people prefer notifications to email, SMS/text or other marketing is because the information is short,  time-sensitive and actionable.

A user can look at a short notification, and immediately decide if it is relevant and if they want to take action – or to ignore it.

No saving for later like email, or even having to open an email app to view – notifications go directly to the user’s home or lock screen, and may even buzz or sound an alert depending on the user settings.

Unless your app is Facebook, your app is not top of mind to your users no matter how much they love your app or brand.

Give them a reason to re-engage with your app, and share that reason with a notification.

Expect Mobile Sales

Even from customers are in your store.

That’s right, data from Thanksgiving weekend in the US showed 23% of purchases from consumers in a retail store were via mobile. The shopper was in the store and was either price-checking the item, or finding ratings or reviews or more information, and purchased from the retailer’s website via their mobile device.

IBM reported that mobile accounted for 57% of online traffic and 36% of online sales – up 30% from 2014.

Holiday Icons & Screenshots

Updating your app regularly is generally a good idea for several reasons, and having an icon that is aligned to the season (or at least the holiday season) signals to prospective users that your app is up to date.

A holiday-themed icon is also eye-catching for existing users who notice your updated icon in their device.

app-marketing-holidays-episodesholiday-app-marketing-tips-me

 

 

 

 

 

Likely your app is either a brand or an extension of your brand, so don’t stray too far from your existing icon, but adding some seasonal treatment can show users your app and brand are engaged with the season just like they are.

holiday-app-marketing-kim

Convert Mobile Web Visitors

The latest operating system for iPhones and iPad, iOS 9 introduces new features for users and publishers that makes searching the web, the app store, content within apps and their device much easier with Siri and Spotlight.

To capture mobile web traffic and convert these visitors to mobile app installs and users – add an app banner.

smartbanner-branch

This alerts your mobile web visitors that you have a mobile app and links users directly to the app store for installation.

Even if you miss Apple’s app store shut down (Dec 22 – Jan 4th), get your app updates submitted as  January is generally a big month for mobile and mobile apps!

Best wishes and happy holidays!

App Store Optimization (ASO) Blog | Mobile App Marketing