The Fast and the Uninstalled

Why Responsiveness is Probably Your App’s Best Feature

5 min readApr 27, 2021

Experiment or Die

Despite what Agile and Scrum methodologies teach us, the software development world is addicted to predictability. The need for predictability is in our human nature. Not only do we want to know what the future holds but we also want to appear competent when the time comes to account to our stakeholders for the time and money spent: is what we’ve done consistent with what we’ve promised.

Unfortunately, when it comes to meeting our users’ needs, reality is often unpredictable, and the more your work environment values predictability, the farther away it gets from reality. Some studies suggest that up to 80% of features in the average software product are rarely or never used. It’s the classic output vs outcome situation.

A way to avoid developing features nobody wants is to run cheap experiments to test them before developing them further. As Stefan Thomke explains in his compelling article Building a Culture of Experimentation, experimentation is the future, and companies that aren’t doing it are destined to be outcompeted by the ones that do. While you’re guessing, your competitors are learning.

And if you think this is something only young startups with greenfield projects and small teams are capable of, think again. Experimentation has long been embraced by giants such as Booking.com, Amazon, Facebook, Google, and Microsoft, with as much as 80% of new features resulting from experiments.

So how do you compete in this new data-driven world? Experimentation requires special processes and infrastructure in order to be effective. And with a low success rate of only about 10% you need to experiment on a large scale to see some positive results. This can be rather demoralising, especially since you often have to kill new features based on negative experiment results.

But there’s one feature that you can always be sure your users want because it’s already been proven by over 50 years of research and data: responsiveness.

Why Does Responsiveness Matter?

The mobile share of the web is growing rapidly with 68.1% of all website visits in 2020 coming from mobile devices (it passed the 50% mark somewhere around 2016). But while mobile is “winning” in terms of most traffic, desktop users remain more engaged with 53.3% of total time on site in the U.S. and 46.4% of total time on site globally, and most conversions still happen on desktop. And while bounce rates are decreasing overall, that of desktop is shrinking faster than mobile.

We know that load times equals bounce rates equals conversion equals revenue. We even have data that shows how much your progression rates will improve by shaving milliseconds off each step of your funnel. A study by Deloitte, Milliseconds Make Millions, suggests that a load time improvement of only 0.1 second can boost conversion by 10%, and lead to consumers spending almost 10% more.

Unfortunately, the inverse is also true: slower load times are often the reason customers abandon the app, switch to another brand, and share their bad experience with others, the study reports. Depending on a company’s revenue this can amount to millions as the price they pay for their app being slow.

While the average connection speed has soared in recent years, mobile load times haven’t, and some studies even suggest that mobile load times have actually increased over the last 10 years. Meanwhile the human brain hasn’t changed and our perception of speed has remained the same. This means not only that making your mobile app snappier can give you a significant advantage over your competitors, but that your app is racing against a static benchmark. Great news, right?

So if this is such a no-brainer, why aren’t big companies doing it? They are! If you haven’t heard of Germany’s 1-Second Club, you’re missing out. The list includes such companies as booking.com and otto.de whose mobile sites load, as the name suggests, in under 1 second. If these giants can get hundreds of their developers move toward the same goal in order to hit this benchmark, what’s stopping you?

How Fast Is Fast Enough?

The benchmarks mentioned by Jakob Nielsen in his article Response Times: The 3 Important Limits from 1993 still stand today and have been confirmed by other studies. It seems 0.1 second (or 100 milliseconds) is an important one, which is very close to the speed of a blink of an eye if you need a real-world reference. Some companies have run experiments that slowed down their apps to test how much they can get away with, but that, of course, doesn’t change our human hardware.

Things get a bit more complicated if we look at animations. When things change faster than our brain can register, Nielsen recommends using animation to help the user keep up. However, because users are exceptionally good at tracking motion, animations that aren’t smooth are perceived negatively. The Deloitte study puts the benchmark at 60 frames per second or 16ms per frame, including the time it takes for the device to paint the new frame to the screen, leaving an app about 10ms to produce a frame.

This might seem like a lot of work, especially when it doesn’t deliver value to the user like new features do. But there’s a strong argument for treating speed as its own feature with its own KPI attached to it. Remember, virtually any new feature is useless until proven useful, but responsiveness has revenue numbers, 50 years of software research, and 200,000 years of human evolution backing it.

Conclusion

Responsiveness can become the best feature of your app that, all other things being equal, adds the most user value, translates to more revenue, and can be your best shot at overtaking the competition.