Nearly every project we undertake can benefit from user research. This can be as simple as analyzing patterns in Google Analytics, testing a new design with a paper prototype, or releasing a feature to a small percentage of users and seeing what happens. Because we may ask you to participate in a study, or you may come across a test we’re running on the Kiva website, we thought it’d be helpful to share some of our research methods with you.
We use a couple of different tools to conduct surveys -- email and online forms. These surveys enable us to reach a relatively large group of people in a short period of time. They are also often the cheapest and easiest option we have to engage our diverse lender base spread out over 200 countries. For example, we surveyed thousands of lenders who are members of lending teams at an early stage in our recent team page redesign  to learn how they use teams and what is most important to them.
Taking the time to talk with lenders and understand how they use Kiva is invaluable. This helps our development team understand what motivates people to use Kiva and get their opinion on current and future features. Over the past two years, we’ve conducted a broad set of interviews where designers, product managers, engineers, and our marketing team sit down with someone to probe on topics or ask clarifying questions to better understand the needs of our user base.
A usability study (or testing) is a technique to evaluate how successful a product or feature is in meeting a goal. Watching people use the stuff you design and build can be a humbling but valuable method to identify what works and what doesn’t. At Kiva, we run a number of different types of usability studies. We’ll print designs early in the process and create paper prototypes to show participants, asking them to complete a task. We also do more formal studies with the help of our friends at Watchlab . They have incredible testing facilities here in San Francisco where our team can observe a potential new lender and ask them questions to understand where our designs can be more effective.
A/B (Split) Tests
Once we’ve developed a feature, we also may want to test it with a percentage of visitors to our site to understand whether it's meeting its goals and gauge lender reaction before releasing it to 100% of Kiva’s visitors. This is a very reliable way for us to determine how successful a project will be. At any given time, we might be running four or five tests on different portions of the Kiva website. In a recent example, we displayed a few Kiva Zip borrowers  in the list of fundraising loans to 10 to 25% of people. This is part of a project to determine the best way to expose existing Kiva lenders to our pilot program Kiva Zip . With the results of an A/B test, we may decide to launch a new feature, make changes and test again, or pursue another idea entirely.
These are just a few of the ways we're conducting user research at Kiva. We hope this gives you a bit of insight into what you're seeing on the Kiva website, and why you shouldn't be surprised if you receive an email asking you to take a survey.
While the method of research may change from project to project, know that our ultimate goal is to gain the knowledge necessary to design the best solution for our lending community. We're interested in your thoughts and opinions, and look forward to hearing from you someday soon.
Send us your questions, comments, thoughts at firstname.lastname@example.org !