When Uninstalls Attack! Winning the Battle Against High Uninstall Rates

02/11/16

No, it’s nothing like a zombie apocalypse or Godzilla. But app uninstall rates are a big concern on a lot of marketer minds, and they can do serious harm to your app business. The title of this post is intended to inspire thoughts of taking action to address the dangers facing your app. In those old B-Horror-Movies, the masses panic and run. But the heroes and heroines make a plan and take action.

When you first detect high uninstall rates, it’s important to quickly diagnose your problem and take proactive steps to address problems. To capitalize on app marketing opportunities to improve your retention metrics. It’s actually fairly straight-forward to diagnose the possible root causes of a high uninstall rate – if you have the right data.

A Data-Driven Approach

Currently, the Google Play dashboard offers an overall figure for app uninstall rates across all channels, vendors, campaigns, etc. This number let’s you see if the overall uninstall rate across your marketing programs is above the level for which you have planned. For some insight into uninstall rates, check out this earlier post. General measures like an overall uninstall rate are best at helping you determine if you have product problems. A bad, or slow, or heavy or buggy product will drive uninstalls regardless of acquisition channel.

But real uninstall analysis requires a lot more granularity. Which is why Apsalar launched Uninstall Attribution as a core metric in its platform earlier this year.

Uninstalls Traffic Report

Uninstalls Cohort Report

We’ve incorporated app uninstall attribution into our solution to help mobile app marketers understand the relative quality of users driven by different mobile app marketing programs and install sources. Using uninstall attribution analytics, marketers can make smarter decisions in a variety of use cases including:

  • Campaign Testing and Decisions: By measuring the uninstall rates driven by different campaigns, they can determine the best advertising approaches for driving long-term users.
  • Regional Allocations: By understanding the user retention rates in different markets, brands can make wiser decisions on where to spend expansion dollars.
  • Cohort Comparisons: By comparing the relative app uninstall rates of different user segments, they can make smarter decisions about future acquisition and remarketing campaigns.
  • Vendor Allocation: By comparing uninstall rates of different vendors, they can collaborate with these partners to drive install quality improvements and focus marketing investments on best performing programs.

Marketers work extremely hard and spend big money to drive mobile installs. Attribution for mobile app uninstalls was among the most requested new features from Apsalar clients and prospects alike, and we are very pleased to have developed a solution that offers accurate, high quality insights into this important arena.

But once you have uninstall measurement, it’s important to understand how to use it, and how to interpret the data in order to quickly identify the root causes of issues. Let’s examine how we can use these more granular measures from Apsalar Uninstall Attribution to assess the root cause of high uninstall rates. We’ll do this based upon example scenarios. We’ll first identify an example finding you might spot in your data, and then discuss its probable meaning.

Observation: High Uninstall Rates for Organic Installs
Diagnosis: Product or Messaging Issues

When you see high app uninstall rates for organic iOS and Android installs, that often means you have a product issue. With an organic install, there are no marketing-specific levers in play beyond what you promise the mobile app will do and the actual app experience. Messaging can drive high uninstalls if you overpromise or if your Google Play or App Store page is unclear about actual app experience. But the more likely issue is that people try your product and simply don’t like it or get frustrated by it. The most common reasons why people abandon/uninstall mobile apps relate to product performance.

When an app won’t launch, or launch quickly, or crashes frequently, or functions slowly, or offers bad UI/UX, data show that people quickly lose interest. In 2014, App Dynamics did a survey of Android and iOS app users in the US and UK. The data revealed that two thirds of consumers say their expectations of app performance were increasing, and that almost 9 in 10 had uninstalled an app because of poor performance. A Compuware study showed that crashes and slow app performance were the most frequently cited experience issues.

app performance issues uninstalls

The same survey of mobile app users in the US, UK, France, Germany and India showed that one or two bad experiences can be major contributing factors to a user’s decision to uninstall an app.

app unistall data

Another big product issue is UI/UE. Here are a few of the product characteristics that tend to correlate with high app abandons and uninstall rates.

  1. Unusual UI/UE, especially when clickable content is not clearly delineated.
  2. Inconsistency, when a certain action is described in different words, or when navigation changes on different screens
  3. Absence of default values, meaning no pull down menus, especially during registration
  4. User “dump”, or when an app simply drops the user into the experience without some sort of introduction or tutorial.
  5. Screen size issues, when an app is optimized to only one screen size instead of adapting to the specifics of each device.
  6. Small click targets, and fat fingers don’t mix.
  7. Too many features leading to user confusion, especially during early trial and use launches.
  8. Complex Registration, such as including too many questions that seem irrelevant to the core functionality of a mobile app. Or requiring sign-up via social profile, or not including sign-up via social profile option.

The important thing here, whether you are a product leader or an app marketing person, is to find out exactly why people are uninstalling. What are the product or UI issues, specifically? One powerful way to get at those answers is to conduct a survey of recent uninstallers. Using Apsalar Remarketing Audiences, you can define such an audience and use a research supplier to deliver surveys to those specific anonymized individuals. While you might think that the response rates for uninstallers would be near nil, I think you’ll be surprised. People want to help if they think you are listening.

Observation: Installs From a Vendor or Vendors are Well Above Average
Diagnosis: Mobile App Marketing Publisher or Traffic Quality Issues

Small differences in uninstall rates across media sources are normal. You’ll find that these rates are constantly changing in small ways over time, as may the ranking of vendors by uninstall rates. But a vendor showing very high uninstall rates relative to others warrants your attention.

Most networks and sources are anxious to deliver the highest quality Android and iOS users for your investments. After all, that’s how they will get renewals/more of your budget. You’ll find that the vast majority of vendors welcome the opportunity to understand the data and take steps to make things better.

Uninstall rates are a great place to collaborate with a vendor. Sometimes they are driven by a single publisher or class of publishers. Other times the mix of ad formats may require adjustment. Or, in some cases, high uninstalls can be a sign of low quality inventory or even fraud.

The first way to collaborate with a vendor in order to improve results is by ensuring that they have the uninstall data necessary to incorporate into their optimization modeling. Collecting and sharing the data are essential to helping a vendor do the best job possible for you. Securely share uninstall information via the Apsalar platform so that networks can proactively adjust the mix of units and publishers to yield the best possible users.

Second, letting a network know that you are seeing relatively higher rates of uninstalls from them helps alert them to the issue so that they can adjusts the dials on your campaign for better results. As with all forms of post-back and other performance data, uninstall rates are an important input to improving your results for marketing programs and deepening your partnership with the best vendors.

Observation: Campaign or Creative Showing Well Above Average Uninstall Rate
Diagnosis: Likely a Messaging, Offer or Aesthetic Issue

As we all know, different ads and campaigns pull differently, just like Google Play and App Store pages do. When a particular campaign or creative is driving high uninstall rates, it’s time to adjust your mix to favor better performers, or make revisions to those ads in order to boost your quality install yield.

Remember, though, that you should focus on optimizing to your KPIs, not (just) uninstall rates. One campaign could, for example, yield fewer but very high quality users. Though generally a high uninstall rate will also yield low ROI.

Observation: High Uninstall Rate for Incentivized Downloads
Diagnosis: Time to Calculate Cost per Quality Install

Incentivized app downloads, or the practice of rewarding users for downloading/installing the app, have been around for a long time. If you are unfamiliar, incentivized app downloads use a digital reward – like free game gold or free WI-FI service for an hour – as an incentive to get you to download an app.

Most marketers expect that incentivized app downloads will yield higher uninstall rates. After all, the user’s motivation is different. But what matters is how much higher. Let’s take an example. If incentivized components of your plans are yielding a CPI that is half the price of installs via other methods, and the uninstalls are only 20% higher, then incentivized may represent a good value. It’s all about calculating a cost per quality or persistent user.

Observation: High Uninstall Rates in Some Markets
Diagnosis: Messaging or Language Issues, Device Capacity Issues

Sometimes the data show big disparities in uninstall rates by country or region. The causes we most often see for this include communication issues like no or poor localization of content, poor app performance because of connection quality and speed, and the fact that in some countries – especially in developing markets like India and China, devices have small memories so users must constantly uninstall mobile apps in order to make room for other apps.

Addressing localization issues is relatively straightforward, though it requires time and resources. Network and connection issues are very difficult to mitigate without a major redesign of the app to be lighter on less taxing on processors. The mobile apps device space issue is simply a fact of life in some markets, and you need to be cognizant of it as you examine your uninstall rates.

It also may point to an opportunity for remarketing to uninstallers to get them to reinstall when the need for the app resurfaces. Using Apsalar Remarketing Audiences, for example, you can define an audience of recent uninstallers and use your vendor(s) or choice to specifically market to them.

Conclusions

High uninstall rates can be worrying. But taking a measured approach to quickly understanding and addressing your uninstall issues can yield big benefits quickly. Consideration of uninstall rates should be part of your efforts to optimize performance to your core KPIs. The key is to have the information needed to pinpoint the problem areas on your business, and then a concrete plan of action.

Share Button