In this TED Talk by Margaret Gould Stewart she talks about making design changes to giant websites such as the like button for Facebook with over 1 billion users. The presentation is about the importance of data-driven design, but also about how data can never replace the human touch.
And that’s the secret sauce. Combining data with human design and interpretation.
Today everyone talks about predictive analytics, A/B testing and data science, but how many companies and app developers really use data beyond analytics telling them downloads, active users and maybe churn? Not that many based on our experience.
So how do you increase your chances of success using data?
Test the Concept
Have a great idea for an app (or website)? Before you spend time, resources and money designing or developing your idea, test it with real users. Start by understanding the problem you are solving and verify that it is a real problem through interviews and surveys with real people. Once you’ve confirmed the problem, start testing the solution using prototypes.
We’ve previously written about the importance of user testing every step of the way from design to launch. Use the data to guide you, but not to control the processes. If your idea or design is truly unique then users may not get it straight away.
Continue user testing throughout development. Implement analytics early on to see the behavior among your internal users. Continue with concept testing whenever you’re uncertain.
Before Commercial Launch
Alpha testers, beta testers and pilot users. Don’t be afraid to launch before you’re ready, even Google does it. Once again, make sure that you collect as much data as possible during trials.
Launching an app is just the start of the journey. Monitor your analytics, dig through the data, continue to ask for customer feedback and test new features and changes before launching.
Will this advice guarantee success? Not always, but it will increase your chances.*
And the 281%? I asked three people to give me a number between 1 to 1,000 and took the average. What I can tell you however is that your chances of success will be much higher if you follow the advice in this blog. Probably 10-fold based on my experience of over hundreds of apps in the last couple of years alone.
* Note: Coca Cola famously launched New Coke based on data telling them it was a better Coke and it still failed. If you’re uncertain about what your data tells you then ask a market researcher or statistician to give you input. E.g. what data is statistically a safe bet and not to ensure that you use your data in the right way.
How many people read this blog? What did they do after reading it? Did the topic and headline perform better than other blogs? Data-driven blogging? You bet!
President DMI International