I’ve been working at Sabre for 4 years and I’m currently a product engineer on the TripCase team. I have years of experience developing consumer-driven products on cutting-edge technologies. I enjoy sharing my passion for using open source technology by speaking at various conferences.
My current emphasis is building modern solutions based upon open-source technology including HTML, CSS, JavaScript, Rails, and NodeJS.
Read on to learn how developers can use concrete analytics to discover how mobile customers interact with websites.
Measuring Customer Success
How many customers visited your website yesterday? For many that’s the key performance indicator measuring success.
Telling your team that 50,000 people visited your site sounds impressive, but it doesn’t say anything actionable. That number won’t tell you what your customer used on your site, when they dropped away, or how long it took them to figure out what to do.
How can you measure your customers’ success?
What if your business goal isn’t publishing pages of articles on the web? What if instead you have a highly interactive, dynamic webapp? Webapps have buttons, switches, checkboxes, and other UI complications. Screens chain together forming complex workflows.
Move beyond reporting vanity metrics like visitor counts and page hits.
Beyond Page Counts
Single page webapps (SPAs) are the type of modern programs delivered from webservers and consumed on browsers. Measuring customer visits by page count doesn’t mean a thing with SPAs.
Why? Because as the name implies, single page webapps are created from viewing just one page.
It constantly mixes in, trims away, and rebuilds itself in response to user interaction. It’s an advanced architecture to be sure, but you’ve probably used one this week. Gmail, Netflix, and TripCase are all examples of SPAs.
How do they measure their user’s experience? What data is available to help their designers, product owners and programmers decide what to change?
I recommend tracking individual events as your users tap specific elements on your webapp’s interface. Only then can you accurately know people are using the features you’ve built.
Tracking analytics events will help you realize the ideal of build -> measure -> learn.
Analytics Event Tracking
Popular, traditional analytics tools won’t entirely track the use of a single page app. Instead, find an analytics tool allowing programmers to send tracking events linked to specific UI code.
Send an event when users interact with significant UI widgets and controls in your UX. For example clicking a button, selecting a menu option, and tapping an image.
How you name events is up to you. I suggest keeping it simple. Choose major categories (nouns) and minor actions (verbs). For example:
- Flight, Add
- Hotel, Edit
- Trip, Share
Sometimes you’ll send extra data with an event. Adding specific data provides better understanding through parametrized context. Extra data is usually a simple type like a number or string. In the examples above you might send: flight number, room type, and duration.
UX events are collected anonymously. Individual users won’t be matched up with their stream of click events. That’s rarely useful. Product developers can learn how their app is used analyzing reports of events accumulated in bulk.
Study analytics reports listening for your user’s voice in the numbers.
Custom Reporting Dashboards
Mass anonymous event collection is useful. It enables you to scale up the reach of your understanding. Event tracking is always recorded no matter how far away, or in what timezone, your users are.
Sometimes tracking events is more accurate than asking someone in real life how they use your app. People often tell interviewers how they aspire to use an app forgetting what they actually do.
Create custom reporting dashboards based around your product hypotheses. Consider what problems your users have, and how your changes solve them. Think of gaps in the marketplace that you want to fill. Understand real-world pain-points that you want to relieve with software. Ship out your solutions and watch how they perform.
Dashboards help your team cope with big-data overload by reporting even tracking data measuring your hypothesis. High-five each other when your features succeed. Pull together and improve features when they aren’t hitting key indicators.
Celebrate wins by visualizing the numbers.
Many Tools Exist
I’ve been using Google Analytics for the past few years. It’s an industry leader and integrating it isn’t overly complex.
This is a highly competitive market, and many alternatives exist. Some of them are: MixPanel, Flurry, KISSmetrics, ChartBeat, and Piwik.
Each of these analytics tools has superpowers. Demo a few of them for a few days and write some test code. Choose the one that best fits your app’s needs and tech stack.
My Upcoming Talk
Why am I writing about user experience and analytics? I’m speaking at a technology conference called HTML5DevConf held October 19–23.
My talk is called “Measuring the Mobile Experience: The Analytics of Handheld UX.” If you’re attending HTML5 Developer Conference, join my talk to hear concrete tips and tricks on how to program UI event tracking, discover user devices, and customize reporting dashboards.