In her previous life, Nancy Hua was a successful algorithmic trader, making a career out of analyzing the wealth of data that drives the financial markets. When she made a transition into technology and software development, Hua almost immediately gravitated to analytics. She applied that same obsession she had with understanding numbers in the financial world to software user experience.
“There’s a story that can be told with every ebb and flow in the financial market, and the numbers are the foundation of that story,” Hua says. “The same is true when it comes to user experience in software. In mobile, in particular, developers have an incredible opportunity to understand and improve how people are engaging with their applications — and the story starts with data.”
Seeing this opportunity, Hua founded Apptimize in 2013. Apptimize allows digital product development teams to iterate faster and better by enabling them to develop apps in real-time through updates, A/B testing, and analytics. Apptimize’s list of customers is impressive, with top apps including The Wall Street Journal, Vevo, HotelTonight and Strava.
We recently had a chance to talk with Hua about Apptimize, why she started it, and how analytics are playing a crucial role in the maturing mobile industry.
Why did you start Apptimize?
When I started in the industry, companies knew they needed a mobile presence — clearly it was emerging as one of the primary ways people were interfacing with the digital world. But it’s also the area where the user experience standards are the highest and there are the most barriers to making that experience great.
At the time, a lot of the measurement tools were lacking. I wanted a tool that made it much easier to understand the data users were generating — then iterate and deploy new releases, and see how those changes impacted the user experience. That’s where Apptimize comes in.
There are a lot of app analytics tools. For those who don’t know Apptimize, what’s the simplest way to explain how you fit into the ecosystem?
Historically, a lot of focus in the mobile industry has been put on user acquisition. Getting people to discover your app is challenging, but getting them to stay is even harder. 80% of apps are deleted after one use. More mobile leaders are realizing the importance of engaging and retaining users in their app as a key growth metric by providing a great user experience. Apptimize is really focused on retention, with A/B testing and feature release management at the core of our offering.
“Without good data that tells a story and showcases a need, teams will feel a lack of meaning in why they are building features, and why users are behaving in certain ways.”
Apptimize found that app teams that launch eight or more tests per year see their app grow at twice the rate of those that don’t launch any tests. In terms of consistent data-driven mobile growth, it’s absolutely critical to test tactics that can increase app engagement.
So, how is Apptimize different than some of the other tools available?
A key differentiator of Apptimize is the audience that we serve. A lot of app analytics tools are positioned for marketing teams — but Apptimize is designed for product teams. We’re for product managers, engineers, designers and QA teams that are responsible for improving an application’s user experience and retention over its lifecycle.
Our mission for Apptimize is that it must be really easy to use and adopt — product development teams can implement it and very quickly see a return on their investment. Making it easy to get started also encourages a culture of experimentation — people get excited to try different things once they know how simple it is to test and see results.
Ease of use was the inspiration for features like our WYSIWYG visual editor. With Apptimize, the first iteration doesn’t require anything to be tagged. You can just use the visual editor to change and test visual elements without any coding. We’re also the only ones to have done the work of figuring out parity between iOS and Android on features such as autoimport, so that teams can easily look at user data across platforms where it makes sense.
For a company that’s just getting started in mobile, how do you explain the importance of analytics?
Analytics should be the foundation of all of your decision making. Without good data to tell the story of user experience, it won’t be clear what’s actually causing anything. It’s guesswork to add new features or make changes to the UI without the ability to measure the impact of those changes. Consumers stick with a particular brand or app because they trust that the app will deliver the relevant experience — good analytics helps you meet those expectations and even exceed them.
Aside from answering questions about your product and informing product changes, analytics helps with team cohesiveness. Without good data that tells a story and showcases a need, teams will feel a lack of meaning in why they are building features, and why users are behaving in certain ways. The main defensible advantage that your app has against competitors who may want to build a similar app, is the data you have from your users. With the right analytics tools, you can optimize your app and the experience will continue to mature.
What types of apps are best suited for A/B testing?
All apps need A/B testing. I suppose any app team that doesn’t really want to build product expertise or user expertise as a scalable competitive advantage might not want to test — but I don’t really know anyone like that. I guess if you’re a genius who is completely in tune with users’ evolving preferences and will always be able to iterate based on intuition, you might not need A/B testing.
But seriously, the top apps in the world, regardless of category or industry, use Apptimize to optimize the experience more quickly and with more certainty so they can improve and tailor the experience for each of their valuable users.
What’s an interesting experiment that you’ve come across?
I am constantly inspired by the tests that our customers and my team run — it’s amazing how much there is to experiment with. One particular initiative that comes to mind is from our customer busuu, which builds language learning apps. Antoine Sakho, their head of product, presented about their subscription optimization strategy at our Mobilize 2017 event in October.
The company historically allowed free users to switch at will between learning any language, but they would encounter locked content in each of them. The team tested restricting access to all languages to premium subscribers only. The hypothesis was that hardcore free power users who learn several languages at a time would upgrade to continue learning.
What they saw was significant — an 83% conversion rate increase for the power user participants — one of the strongest results they’ve seen through hundreds of experiments. Content locking is a really interesting tactic to experiment with. We have a lot of case studies on the topic.
You live in data. What’s a surprising finding that your team has come across during the course of your work in the past year?
We often see that different customers run very similar tests but the results may differ. For example, someone will run an experiment about their onboarding call-to-action (CTA) and come out with a clear winner. Another customer might try the same variants in a test and get results that tell a completely different story. This goes to show that each app and app user are different, and what works with so much certainty for one app is not a general best practice. This further shows the importance of testing to determine the best UX for your specific product and your customers.
What is your point of view on building an in-house A/B testing solution versus buying it?
To build something well in house, you need to make sure you have the right resources — specifically skillset and development time. Unless you’re a tech giant like Facebook or Airbnb, that’s likely not the case.
Many companies try building in house and then come to Apptimize. We’ve heard that in-house solutions tend to be much more restrictive and difficult to use, so the tool actually doesn’t get a lot of use. It’s typically built in a very specific way for a specific kind of user, so other people will get errors or stuck when they try to use it. Mobile leaders increasingly want to test more frequently, so this lack of accessibility poses a barrier. With Apptimize, our users are testing hundreds of times per year and that’s because the product is so accessible — everyone at the company can contribute to the experimentation program.
Also, companies tend to underestimate the cost of maintaining these custom in-house tools. It’s not the kind of thing you can build once and be done — there are a lot of advancements in mobile. As an app evolves, the types of things you want to test change from what you initially conceived, so the analytics tool also needs to be updated. When the app advances, but the analytics tool doesn’t, the testing culture loses out.
Frankly, experimentation software is hard to build. Apptimize is our only product, so we are continuously improving our tool and solving new technical problems. By purchasing an experimentation solution, companies save time and money, and are able to focus their internal resources on what they do best — building a better product.