Qualitative user researchers generate and analyze massive volumes of data as part of their daily work. Information spawned from popular exploratory methods like interviews and observations takes the form of many mediums, including video, audio, and photographs. Each hour of research conducted can easily require three or more hours for analysis.
Why? The attention to detail required to ensure one does not miss insights spans a swath of time-consuming activities. It begins with pre-analysis tasks such as transcribing audio or video recordings, continues with carefully reading the content of these transcripts, then concludes with the monumental task of highlighting key findings and synthesizing them into actionable recommendations. When done right, the payoff is tremendous.
These duties are at the heart of researchers’ responsibilities, and many of us adore performing them. At the same time, mounting workloads, excessive screen time, and a desire for continual improvement have driven many to seek means of optimizing these activities. Fortunately, there are solutions to support research practitioners in automating repetitive tasks so they can spend more time understanding and advocating for their customers (Dovetail, the parent company of this blog, is one of them).
Closely tied to a research activity commonly referred to as coding, tagging taxonomies are structures that formalize how researchers categorize and interpret data. Specific tags to implement vary at every company. However, they generally encompass common themes that emerge in your customer research sessions and data. For example:
Competitor company or product
Feature request or recommendation
When properly implemented and refined, a tagging taxonomy drastically speeds up the research analysis process, leads to the uncovering of impactful themes, and can ultimately result in better evidence-based customer-centric decisions that improve your product’s user experience.
This all sounds fantastic, right? Such a massive reward does not come without effort. Creating, implementing, and refining an effective taxonomy is no small undertaking. Without a seasoned research leader in your team or a dedicated ResearchOperations practitioner, establishing a tagging taxonomy can become a daunting task riddled with deterrents: how do we make time? Where do we start? How do we know we’re doing it correctly?
These are great questions. As previously alluded to, one size does not fit all when it comes to building tagging taxonomies. The rules, parameters, and protocols differ depending on the nature of your research, size of your team, and deadline cadence. When it comes to best practices, practitioners from around the world are still experimenting, refining, and learning as they go based on field experience. In this article, I’ll share lessons learned from my experience establishing a tagging taxonomy as a team of one at a startup.
Between 2016 and 2018, I led UX Research at Reflexion Health, a telerehab company focused on helping individuals recovering from hip and knee replacement surgery conduct physical therapy digitally from the comfort of their own homes. Working in an emerging field, I focused sizable efforts on exploratory research to better understand our demographic of senior-aged users, learn more about the physical therapy space in general, and evaluate the merits of our digital-physical therapy solution. Overseeing the development of multiple product lines and a totally novel user experience versus traditional physical therapy brought the need to balance the speed and quality of my work. This involved executing processes to streamline productivity—one of which involved establishing a tagging taxonomy. Let’s delve into several common pitfalls researchers encounter when setting up a taxonomy and how to avoid them.
Solo researchers often find themselves balancing responsibilities for both User Research and ResearchOps. With perpetual impending deadlines and barrages of requests from stakeholders, advocating for, establishing, and maintaining a tagging taxonomy can feel like a lower priority endeavor versus daily tasks like recruiting users and creating research plans.
Yet, despite the initial time and resource investment required, creating a tagging taxonomy actually saves countless hours in the long run by helping you accelerate your analysis and achieve your research goals with high-quality, evidence-based answers. Over time, a tagging taxonomy can even help you uncover new insights from previous research by identifying common themes between past projects and newly conducted research.
Another potential stumbling block when creating a tagging taxonomy is creating a new group of tags for every project. While there will be certain project-specific and unique situational tags for each user research project, it can be a mistake not to establish global tags that relate to common occurrences researchers encounter in nearly every session—things like pain points and motivations. It can be very valuable to establish a set of global (also called universal) tags, then compliment them with project-specific tags after analyzing the data from the sessions. For an outstanding in-depth discussion of global versus project-specific tags, check out our article on Creating global and project tags for your research repository.
As researchers, we tend to love digging into the specifics of data—slicing and dicing it in countless ways to identify interesting patterns about our customers. While intellectual curiosity can often reveal helpful learnings from your research, going overboard can lead to the creation of too many hyper-specific one-off tags that don’t have enough merit to inform an important decision that impacts your customers.
Conversely—as champions of design thinking—we might have the urge to define very general tags, so we don’t overwhelm our stakeholders with an excessive number of tags. This oversimplification can result in missing the richness, nuance, and epiphanies hidden in our research data. As detailed in What we learned from creating a tagging taxonomy, limiting your taxonomy to 25 tags across five groups strikes a great balance between efficiency and depth.
Your tagging taxonomy will evolve with the needs of your customers and organization. As professionals who conduct research nearly all day long, it can be highly tempting to desire a “set it and forget it” approach to maintaining a tagging taxonomy. However, the principles of user research remain: we must build for our users, evaluate our experiences, and iterate. It can be invaluable to frequently think about who will consume your research tags and configure your taxonomy for their needs. Also, remember that, as with implementing any new process, there will be a learning curve. Ask your teammates for candid feedback, regularly inquire about its effectiveness, and consider creating help documentation.
Building a tagging taxonomy can be intimidating. As an incredibly useful but relatively new tool, it’s natural to feel uncertainty while implementing one in your organization. Sometimes the best approach is to start small—begin by identifying global tags for categories you frequently encounter in your research and match it to evidence from research sessions using a repository like Dovetail. From there, you can begin to refine the protocol for using these tags, evaluate the effectiveness of the taxonomy, and continually evolve your processes.
We’ve covered some common pitfalls you might encounter along your tagging taxonomy journey and how to mitigate them. We’d love to hear about your questions, experiences, and lessons learned from implementing tagging taxonomies throughout your career. Head over to the Dovetail Slack to connect with fellow researchers and continue the conversation today!