It’s Never Too Late in a Product to Conduct User Research

Problem

"When I was working at Glassdoor I was a product manager that owned all of our salary experiences. Salaries were the founding product for Glassdoor but they had not evolved much over the course of the company's growth. Instead, the company focused on introducing new content areas and expanding other verticals. I was given the challenge to redesign the salary experience to fit the needs of our job seekers, to make sure that we were staying competitive in a market where other companies were now thinking about salaries and salary data."

Actions taken

"At the beginning it was difficult because we already had a successful product, something that people used and already loved. So we needed to make a v2 that didn't feel like we were reskinning the product. We needed to make sure that what we built would solve existing problems, new problems that were currently unsolved, and customer needs that were not yet met. Keeping this in mind, after we completed our initial designs, we felt as if they were too iterative. We didn't have a strong enough vision of what we wanted our salaries product to look like and how it was going to enable us in the future."

"It was at this point that I took a step back and thought about the product as if it were brand new. We went back to the drawing board and thought about our user segments in terms of job seekers. People who would newly enter, or re-enter, the job market. People who were looking for a career change. People who wanted to stay in their current track but were looking forward at what their career trajectory could be. We took all of these user segments and thought about how they related to salary and compensation. This forced us to back to our roots. What were the needs of these various users and how were they each different?"

"Next, we brainstormed a list of all the possible solutions to solve the problems that we could think of. We created a set of 10 to 15 different concepts- an example of a concept might be how your pay will differ depending on level of education- and for each of those concepts we built very simple modules that showed respective information. We took these modules and did use case studies and user research, as if we were building a brand new product. We placed the concepts in front of users and asked if the information was helpful based on their current situation. Through this research we were able to get a clear roadmap of the users' top needs and identified which segments we should target for our company."

"In conclusion, the data gave us a roadmap of what v2 of this product should look like. It also gave us concrete evidence of why it was worth investing in. This data would enable us to build a product for the needs of the current user as well as be a foundational base of information for the future when it came time to build v3. Thus, utilizing the data gave us a much more strategic vision for what salaries meant to the company and to the user. By returning to the research, it helped us both qualify and quantify what we wanted to build and the impact that it would have."

Lessons learned

• "I believe the biggest thing is that it is never too late in a product to conduct user research. Sometimes user needs change. Sometimes technology changes. Maybe the competitive landscape shifts. Or maybe the demographics of your users fluctuates. It's always a good idea to go back to the foundation, even if you feel like you are iterative, and know exactly what users want."
• "Sometimes you can learn new things just by starting from scratch. And if not, at least you have reaffirmed that you are still solving the right problems for your users. This is better than simply relying on the current iteration of the product and assuming that you are still solving the correct needs."