I spun around in my chair and nudged the designer sitting next to me, “wow, look at what just happened.” I showed her my phone, which displayed an ad that was seemingly frozen to my screen. My music was playing, but I could only see the ad on the screen. “What is that?” she asked. I told her it’s an ad on Spotify Free, and this was clearly a bug that I should report to our engineering teams.
It was then that I learned the importance of understanding the nuances of our product. I had been a Premium listener for years, but after switching teams to Spotify Free, I realized I needed to reacquaint myself with the Free experience. This was the day I really learned about the power of dogfooding our own product.
Dogfooding Augments Research, It Doesn’t Replace It
Dogfooding is a practice of using the product or service that you’re working on. Some have criticized dogfooding as a sub-optimal replacement for User Research. I agree that dogfooding should never be a replacement for User Research; however, I have also found that it can powerfully augment our work when done in conjunction with rigorous research.
Three Reasons You Should Add Dogfooding to Your Research Toolkit
Six months ago, I, along with an engineer colleague, Jonah Katz, co-founded a dogfooding practice within our team. And since then, we’ve learned three important benefits of dogfooding:
1) Building a deep understanding of the product
As User Researchers, it is evident that we need to do research to develop a deep understanding of our users. But, we don’t talk enough about the importance of knowing our product. Dogfooding is not a tool to learn about users. We have many more rigorous and in-depth methodologies that give us those insights. Instead, dogfooding gives us an important opportunity to learn about the ins and outs of our product and feel the experience for ourselves. The closer we are to the product, the more precise our studies and the resulting design and product recommendations can be.
2) Identifying opportunities to innovate
After several dogfooding efforts with my product teams, I’ve been surprised by the quality and quantity of innovative ideas that have resulted from experiencing the product for themselves. In understanding the nuances of the experience, we’ve been better able to identify small changes that can make a big impact to the user experience. I especially encourage my stakeholders to dogfood before hack days, as it is a powerful tool to identify gaps in the experience and generate new ideas for how to improve it.
3) Empowering the team to take ownership of a high-quality experience
As product Managers, Engineers, Designers, Data Scientists and other people on tech teams, it is important that we all take ownership of the product we’re building. But, we really can’t if we don’t know what that experience is truly like. We have to understand the experience, assess its quality and be diligent about fixing anything we don’t feel is good enough. Dogfooding is a great way to ensure that the product you’re building is a high-quality experience.
How to Add Dogfooding to Your Ongoing Research Studies
Now, I’ll walk you through two possible ways for using dogfooding to augment your research projects, depending on the stage of the product.
If you’re testing a brand-new product concept…
Dogfooding can help you understand a new feature experience, catch bugs and ensure a product is high quality. For instance, if you’re launching a new feature, you can do exploratory research to help direct product ideation, and then dogfooding to test for high-quality experience. It might look something like this:
- Exploratory in-home research
- Concept testing
- Iterative usability testing
- Dogfooding
- A/B testing
This setup will allow you to understand what your users need from this new feature in addition and help inform the concept, thereby ensuring the final designs are usable.
If you’re iterating on an existing product…
Suppose your product team wants to revisit a product years after it launched. Through dogfooding, you can understand the holistic experience, and generate ideas for revamping it. In this case, a research process might include:
- Data analytics
- Diary study with follow up user interviews
- Dogfooding
- Ideation workshop
The data can show you interesting trends in feature usage patterns. The diary study can give you a deeper understanding of how, when and why people use your product throughout their days. Finally, dogfooding will enable the team to reflect on the edge cases and confusing —or even frustrating — experiences. In taking all three perspectives into account, you can ensure you’ll have a rich ideation session.
Regardless of the stage of your product, dogfooding can be a useful addition to your research regimen. It will allow you and your team to understand the nuances of your product, which deepens the meaning of your User Research and Data Science insights. Dogfooding can help you generate empathy for what a user describes in an interview or a hypothesis for any hard to explain trends in the data.
Tips & Tricks for Successful Dogfooding
Over the past six months, I’ve learned a few things that’s helped us increase the impact of our dogfooding:
Make it easy to access the experience in daily life
In our case, employees weren’t naturally interacting with Spotify Free regularly. Since employees get free access to Spotify Premium, we had to encourage people to use Spotify Free regularly. To solve this, an engineer on the team built a system that takes all of the playlists, music and podcasts from an employee’s Spotify Premium account and copies it into a Spotify Free account. This reduces the barrier for people, who have already invested in building collections and playlists, to find the music and podcasts that they love. As a result, they listen to Spotify Free more often!
Build rituals around dogfooding
Building conversations around dogfooding made it easy for everyone to easily share their stories and learn from others’ experiences. We’ve found that making Slack channels specifically for dogfooding helps people regularly post screenshots and share their reactions with the group in an efficient way. And it encourages others to do the same and express how similar or different their experience was. We’ve also had debrief sessions with teams after dogfooding, where they can share stories, reflections, and ideas with the group. Everyone walks away from these sessions with a shared understanding and an inspired feeling, ready to go forth and improve the experience.
Document the feedback
Regardless of how you build the rituals around dogfooding, it’s helpful to have systematic ways for people to report, and for you to collect, reactions to the experience. This way, you can capture their feedback just as they’re listening to Spotify Free. This systematic feedback loop will enable you to learn about their experience in the moment and continuously throughout the dogfooding effort.
I hope you find dogfooding to be a helpful addition to your User Research toolkit. In addition to typical User Research methods that yield a deep understanding of your users, dogfooding can help you understand the ins and outs of your product. Tell us how this process works for you as you apply this in your research in the comments section below!
Colette Kolenda
Colette Kolenda is a mixed-methods User Researcher at Spotify. She enjoys combining qualitative and quantitative methods to deeply study research questions and fuel product decisions.
