Bob Lawson, director and co-founder of Kumulos, looks at the reasons why a lot of mobile apps aren't very good and provides advice on making sure an app is successful
On the face of it a very controversial statement for a mobile first company, isn’t it? So, let’s first explain what we mean. It’s not a blanket statement that all apps don’t work. Of course, many apps DO work and work very well. But, it’s also the case that many apps just disappoint.
In a flurry of enthusiasm an app is created, but users don’t come, or those that do, don’t stick around for very long. The publisher gets disillusioned and assumes that “Apps just don’t work for our business”. It’s the disillusioned we are talking to here and how they can learn from others to do the right things, often the simple things well, to greatly increase the chances of app success.
Before we get started, it’s important to set a bit of context here around why we think we are qualified to comment. We currently work across more than 24 countries, spanning seven continents. This encompasses more than 17 industry verticals including Retail, Financial Services, Fan Engagement, Customer Loyalty, Fitness and Wellbeing, Logistics, Retail and a host of others. Apps that serve a whole range of purposes. Internally facing apps for employee engagement and workforce efficiency. Apps that are a digital channel delivering incremental revenue for businesses or apps that ARE the business. We work closely with apps designed to deliver fan or investor engagement or loyalty apps designed to drive footfall to physical locations like high street retail outlets. So, lots of apps for lots of reasons, but all of them have common threads running through them – secrets for success if you like.
Kumulos isn’t an app developer or an app publisher. We sit in this unique position between the two, providing technology that’s used by some of the biggest and brightest brands in the world. So, we’ve seen a lot of apps in our time. Some good, some quite frankly great. Some that weren’t great to start with, but got there in the end. This privileged position lets us see mobile apps, built for a host of different reasons and a view of the data and metrics that sit within them. When you look at the analytics data, it throws up some really interesting trends. Something we just had to share with you. So, we have pulled together this quick read that looks at the common mistakes we see being made by app publishers and their developers.
The result is this blog – five reasons why apps don’t work. We hope it gives you some ideas on how you can think differently about how you approach mobile in your business, so you deliver better results.
Really understanding why you need an app is a good place to start. We often hear,
“We need an app“… why do you need an app? “Well because everyone else has an app, we don’t want to be left behind.”
Or worse App Washing. That’s where businesses have apps, so they can say to their bosses that we are with it and Digital.
No matter how good your app is put together, if the underlying reason for having an app isn’t sound then chances of success are slim.
It’s also not just whether you need an app at all – it’s about thinking carefully around what type of app you need.
Wrong type of app is as bad as trying to use the wrong tool for the job. You may get the job done, like hammering a nail into a board with a spanner, but the results will be rough and ready. Not optimal.
So, think carefully about the technology behind your app.
It’s important to approach this first from what exactly you NEED the app do to, what simple processes and flows you need your app to do before you head down the road of building your app. That way you won’t find yourself up a blind alley nine months down the line, expecting your app to have a function that the platform you’ve chosen can’t support.
For example, if it’s important to be able to track the location of your users to send them information based on their proximity. Or if push notifications are a critical function to send important information to users, or as a tool to ensure regular use of the app. Not all these functions are available across all mobile app development platforms. So, choose your platform wisely.
It’s important to take good independent advice to make sure you take the right approach. If you work with internal development teams, they could be guiding you down the path that best suits them to make their life easier. So, talk to experts with no axe to grind, so you work with the best platform, the best tool for the job, from the start.
Building the wrong audience for your app is a sure-fire way of hard-won users not hanging around after the download. Also, alongside this, we’d sometimes question whether the app has the right organisational structure built around it, but I’ll come back to that in a moment.
Over promising and under delivering is what we are talking about here. Making bold claims around Mixed Reality or Alexa Voice that deliver little value to the app for example. Or the compelling App/Play store description that will spark interest and get the download, but when they first open the app it simply doesn’t do what they expect it to do.
App publishers work hard and spend hard to attract new downloads. There are lots, and lots and lots of people that will help apps acquire downloads. The fraud element of this aside, if it’s the wrong person then they will never be a valuable user. We have seen some apps celebrate how awesome their app is with thousands of downloads a day. But then suffer an average three-day churn of 97 per cent of daily downloads. Does that make the app bad, maybe, but also is the app just not setting the right expectations for users, or chasing downloads from any source, just to make daily downloads “vanity” stat impressive.
So, you have to ask are they attracting the right audience? Were the users that downloaded your app the people that will get the most value out of the app. There’s a lot of focus on cost of acquisition, lots of attribution analysis on where the best yielding media sources are to drive downloads, but we see lots of very smart people that then don’t (or don’t want to) (or can’t) extrapolate that into customer value.
So, it’s obviously important to make sure you get THE RIGHT people to download the app. But we see so many people latching onto big stats that make them feel good. Our recommendation is ignore (completely ignore) downloads as a meaningful stat. They may be top of funnel stat that sounds most impressive, but they overshadow the real success measures. Engaged users is what really counts. Focus on DAU/MAU trends over time. That’s what’s important. Of course, we’re making a general point. All apps are different, so different measures are important. But if downloaders don’t then go on to use your app, then surely there’s something wrong somewhere.
Engaged users are what count. For success, it’s essential to understand what your ‘model’ customer looks like. How you expect them to use your app, now often and for how long. For example, are there seasonal variations you need to consider in different countries.
That brings me onto what I said a moment ago about wrong structure.
It’s common place for larger mobile apps to have a clear split in responsibilities when it comes to the acquisition funnel. Top of funnel it’s all about acquiring new customers. Bottom of funnel it’s all about engagement and retention. All very sensible. Often that maps onto different teams. An acquisition team 100 per cent focused on driving downloads and CRM team 100 per cent focused on retention and monetisation. That’s OK if they work closely as a collaborative team, but often we find Key Performance Indicators (KPI) poorly aligned. The Acquisition team pouring poor downloads into the funnel, the poor Acquisition team “failing” as they try to manage 97 per cent three-day churn. Sound familiar?
Everybody has analytics running on their mobile apps. Right. Lots of charts, lots and lots and lots of data. The bigger apps will have dedicated teams of Big-Data Scientists pouring over every aspect of their app to understand what’s happening. What are users doing, and who aren’t doing what you’d expect. For apps without that luxury of such resource, I expect that’s mainly businesses that the app is a digital channel, rather their entire business, it’s tougher to get real insight into what’s happening.
If the app’s a task completion app, how many successfully complete the tasks, what’s the average time to complete and what can you do to ensure that this average is reduced over time.
AND REALLY IMPORTANTLY – churn – everyone has churn. For some apps we’ve seen (and the customer seems to accept) 97 per cent of users abandoning the app after just three days. That’s painful. But where churn REALLY hurts is with your super-users those that have been highly engaged, those model customers that you’d like all your customers to be like. When they churn then that really hurts. If the inflow is greater than the outflow, then some apps see that as a success. But with just a small percentage reduction in churn this can have a dramatically positive impact on the success of the app. We’ll cover much more about Churn in a few paragraphs time, so read on.
Why do so many apps not have greater visibility of these vital signs of success? What we see is a combination of factors. The numbers are just too difficult to get in an understandable, easy to digest format. Google Analytics is a great tool, but in a lot of cases it’s just the wrong tool for the job. It’s a web tool trying to serve mobile apps. Firebase Analytics at least is focused on mobile, but it’s very immature and not intuitive and easy to set up. So everyone ticks the box when you ask, “Do you have Analytics on your app?” but lots and lots can’t answer some basic fundamental questions on how their app’s performing and most essentially gain insight into what you need to improve to stem the bleed points in the app keep users engaged with the app and coming back for more.
Which leads me onto the next point.
Starting a journey without a clear understanding of where you want to end up, leads to an inevitable result. Without a clear end goal, you can’t take the corrective action you need to adjust your course to where you want to go. We see this with mobile apps. To be fair, apps are usually launched with an objective in mind – increase worker efficiency, reduce task processing costs, increase brand engagement, or generate incremental revenue. All worthy and sensible objectives and things that mobile apps can definitely help businesses deliver. But how many start out with a very specific objective in mind. To generate 25 per cent reduction in processing costs across 95 per cent of transactions in the next six months. SMART objectives Specific, Measurable, Attainable, Relevant and Time-bound.
That way you have a much clearer view of the levers you can pull to drive success and meeting your objective.
We see a very strong correlation between apps that are failing and those that don’t break the main objective into clear measurable meaningful digestible chunks – Key Performance Indicators. A lot pick the measures that are easy (and to be honest superficial), like Star Ratings in the App/Play Store and Daily Downloads rather than those that are true lead indicators for success, like Onboarding Ratio (Active Users/Total Downloads), Stickiness Ratio (DAU/MAU), Customer Acquisition Costs (CAC), Customer Life-time Value (LTV) or arguably the most important measure Churn Ratio.
Churn is a fact of life. Every app will have churn. And where churn hurts most is with your Engaged “Model” users. Churn with those engaged with your app are the customers you most want to retain.
It can be tough to get accurate churn measures for apps. You don’t know if the app has been completely deleted from the phone, or whether its just been temporarily off-loaded (particularly if it’s a seasonal app, with a long cadence). But one very good measure of engagement is looking at the churn in your reachable audience. Those that have opted into receive notifications.
And importantly tracking that measure over time to see if the trend is increasing or reducing is important.
In that same vein, looking at the ratio of users that sign up and then also opt into receive notifications. It can of course depend a lot on the type of app in question, how important notifications or in app messaging is, but it’s still a good measure and early indicator of audience engagement looking at the proportion of new downloads that sign up to receive notifications.
Knowing whether you are growing your addressable audience over time shows you whether you are either better at selling the benefits to the user or you are more successful at acquiring the right audience, keen to hear from you regularly. And notice that the measure is Active Users Acquired and not total downloads. That bit’s important.
Knowing if your addressable audience is growing or shrinking could be one of the most important performance indicators. Which takes us on nicely to the last point.
The cardinal sin here is telling people what you want to tell them and not being aware enough of what they want to hear from you. Quite simply this is the best way to destroy your high value, hard won engaged users is to hack them off with spam messaging.
It’s not necessarily about volume here. Although frequency can be a problem, obviously. What it’s really about is relevance and personalisation. Making it strongly permission based by giving users control over what you send them. And legislation comes into play here. Getting it wrong isn’t just about reputation risk and hacking of your users, it could cost you hard cash in fines.
GDPR legislation in Europe and the new California Privacy Laws (California Consumer Privacy Act , CCPA 2020), make it even more important that users are opting in to receive information from you. That should be for both Push Notifications AND In App messaging. But personalisation doesn’t stop there. Knowing what type of content they want to receive and giving it to them in the format they want is just as important. And personalisation isn’t just the superficial greeting “Hi Jon”. It’s about language, time of day, rich media like video or pictures, proximity based and relevance to what they are doing in your app and what they are interested in.
So, understanding how users want to be communicated with is essential.
Of course, a successful app must start with a sound idea, something that enough customers will value to make it viable. An understanding of how large your addressable market is, is also important. If there are billions of people who could use what your app does, then you’ll be onto a winner (maybe) even if you have a 95 per cen churn rate every three days. OK, so you may not be fully exploiting the opportunity with such high bleed, but you’ll probably still make it a success. But in the ‘real world’ its all about doing the following things well to increase the chances of success.
Strong & experienced development partner – experience in your sector may be useful, but better is a company that works across a broad range of industry sectors – this will give you the benefit of cross pollination. Taking one idea from a different sector and understanding how to adapt that for your industry. At Kumulos, we don’t build apps, but we can introduce you to our highly talented partners. Just get in touch. Also, back to #1 the right tools for the job. Picking a developer that has strong skills with the technology behind your app (native, hybrid Xamarin, React Native, Unity) or PWA’s could be more important than a strong understanding of your particular industry. So, choose your partner carefully.
Value based KPIs – don’t just pick the ones easy to measure, but those based on the objective of the app, eyeballs (ad revenue), # reduction in manual processes, percentage reduction in call center calls. Study the obvious bleed points (but start simple – two or three main ones 20-30 events) so you don’t get blinded by data and have an achievable actional plan to move the numbers in the direction you need to bring success.
No software is 100 per cent perfect ever – let alone when you launch. We see successful apps adopt an MVP approach, minimum viable product. Then in each development iteration, study how your app’s being used, map this against your key objective and your core KPI’s (that directly drive towards that Key Objective) and make incremental improvements to your app based on what the data is telling you. (analyse, learn, change). It’s important to use use real customer data to tell you what you need to do – not manager’s hunches.
Be smart about how you communicate and engage with your users – Segment on interest and relevance, micro-strategies automated is better than blanket communication. Understand the usage patterns of your super users then use intelligent engagement strategies to get more of your users behaving that way.
So why do we care and why do we have so much to say on the subject. Well because it’s our thing. We understand Intelligent Engagement. We know it’s not just limited to things like Push Notifications, it’s about taking a holistic approach, understanding the user experience as well as attracting and retaining the right types of users. That’s what the world’s biggest apps do, with an army of analysts, data scientists and the rest.
We are on a mission to stop mediocrity and bring a powerful technology stack and 360 view of user behaviour and user experience into the reach every app, and not just those blessed with seven figure staff budgets. By doing the simple things well, understanding the three-five levers that will steer your app to success and then using technology like Kumulos to guide you and we can all deliver incredible app results.
So, I hope this blog has been able to make you think a little and suggest a few things you can try to help mobile apps work for you. Good luck and contact us with your thoughts!