Unlock the secrets to effective UX research for SaaS products with our comprehensive guide
Everyone knows that great design is critical for exceptional SaaS user experience, and most agree that great design begins with truly understanding the user.
But how do you do that?
When it’s time to uncover who is using (or will use) your products, what they want, and what they don’t, research can provide all the answers you need to get started. In fact, if you look critically enough at the SaaS UX design process, you’ll find that research is nearly inseparable from great design.
Unconvinced? Consider the following ways that UX research informs the very foundation of SaaS design decisions.
Research-informed UX site redesigns have led to 500% revenue gains in some cases.
Quality, frustration-free design can increase conversion rates by up to 400%.
91% of unsatisfied customers don’t explain why they abandon products after bad experiences.
Increasing conversion by 400% or revenue by 500% sounds very attractive. However, you won’t be able to ascertain what “frustration-free” looks like or what a “research-informed” redesign entails until you start asking questions and digging into data.
As critical as design is to creating a great product, research is a major building block to ensuring success. Of course, like any serious endeavor in user experience, UX research presents challenges. It takes time, costs money, and requires team buy-in to be truly effective. However, those obstacles don’t negate the opportunities for growth and improvement that research provides.
At its core, SaaS UX is about the user, and research is the best way to clarify their needs. In this post, we’ll explore the basics of conducting this critical step in UX design and refinement and how to leverage research insights to create better products.
UX Research Methods
1. Performing Qualitative Research and Interviews
In-depth user interviews, sometimes called IDIs, are about as deep as you can get with qualitative UX research. Whatever you call them, long-format interviews with end users are one of the best ways to understand what people think about a SaaS product.
By building rapport with an end user in person, over the phone, or via Zoom, you can extract more information from them than you might through an anonymous pop-up survey or Net Promoter Score rating (a composite made from those “how likely would you be to recommend ____ to a friend or colleague” attitudinal questions).
While you won’t get “data” per se, these interviews are a great compliment to numeric survey data and more sophisticated analytics. For many, in-depth interviews are often the first step in a larger research program. They can also help explain why people rate things the way they do in surveys and allow respondents to explain things at a deeper level than is possible in a small survey text box. As long as you recruit the right participants and provide a fair incentive, in-depth interviews are fairly easy to execute and provide a great return on investment.
2. Implementing SaaS UX Usability Testing
Secondhand user-reported data and firsthand interviews are great, however, they rely on customers’ recollections and opinions. Usability testing can help round out the overall picture with objective information.
Although this isn’t an exhaustive list, some of the most common usability testing methodologies include:
Session Recording and Heatmaps
Session recording and heatmaps track where users click and hover, giving you insight into behavior and interaction patterns. Heatmaps typically depict more popular parts of a page with red shading and less popular areas with blue. Session recordings lack the colored visualizations included with heatmaps, but they allow for real-time playback of user interactions in granular detail.
Eye Tracking and Biometric Data
Eye tracking and biometric data can reveal insights into how users interact with and feel about products—revealing things that might get missed in interviews or surveys. Eye tracking can reveal where users look and for how long, which can help you understand which features, pages, and content they’re gravitating to. Biometric data can measure physiological responses like heart rate and blood pressure to determine how the software makes people feel emotionally, even subconsciously.
Tree Testing, Task Analysis, and Cognitive Walkthroughs
All three of these techniques ask testers to find things or perform tasks, but they go about it in slightly different ways.
Tree testing asks users to find items within the navigational structure of SaaS tools, helping discover areas of confusion and difficulty.
Task analysis asks users to complete specific tasks using the product, like “add a new customer profile,” or “change a filter on the dashboard.”
Cognitive walkthroughs check how users will complete some of the main tasks they might perform. For each task, you’ll check if the user can find the correct action, notice if it had an effect, and if they feel they’re getting closer to their goals.
3. Fielding Surveys and Gathering Data
Surveys can be a great tool for gathering feedback from end users, with a couple of caveats. First, don’t overwhelm people with constant survey requests, and if you do field surveys, try to keep them as short as possible.
Coming up with clever ways to embed feedback mechanisms can help make your data even more customized. For example, you could have a survey pop up only after a specific user action is taken, and ask only about that particular task. This can also allow for shorter surveys.
Before you launch any research, it’s also a good idea to consider analyses you might run later. Surveys must be structured in specific ways to allow for certain analyses. Whether you’re running complex models or just looking at basic statistics, thoughtful planning in research is just as important as execution.
4. Harnessing SaaS UX Research on Competitors
Surviving in a competitive market requires understanding your unique value proposition, but you can’t do that until you understand your competition. Comprehensive competitive analyses help you identify what makes your product unique and where it can be improved.
SWOT analysis (Strengths, Weaknesses, Opportunities, Threats) provides a helpful framework for comparing SaaS products and entire companies against competitors. It’s high-level and holistic, but it's a great starting point for brainstorms before more detailed investigations. Feature parity assessment, where you check to see how many of the features of your SaaS tool are in common with others on the market, can also be combined with SWOT. For example, having an extra feature, or missing one, could classify as a strength or weakness.
5. Leveraging Analytics and Predictive Modeling
In SaaS UX research, advanced analytics and modeling can be used for everything from uncovering behavior patterns to forecasting trends. Statistical regression models can help you understand things like which feature additions or updates are contributing to customers' NPS (Net Promoter Score) and other ratings. These tests let you evaluate almost any numerical or categorical data about your product by reliably measuring how much a particular variable contributed to the change by controlling for all the others.
In some cases, when you may not have specific hypotheses or questions, machine learning algorithms can uncover surprising insights through data mining. ML can also be used to power predictive analytics, uncovering unforeseen areas where you can leverage aspects of your SaaS product for competitive advantages.
How to Incorporate UX Research Insights into SaaS Design
To make data-driven decisions, you need to build frameworks for feeding insights back into the design process. Crafting personas, harnessing iterative design, and creating meaningful cross-functional collaborations are great ways to ensure that your research is put to good use.
1. Persona Development for User Segmentation
Persona development is helpful for understanding who uses your SaaS and what they need. By breaking your user base into segments via demographic information and other characteristics, you can unveil meaningful groupings that will be useful in design, marketing, and testing. Internalizing these personas can help you hold different hypothetical end users in mind, making sure your work delivers on the unique needs and preferences of each.
Personas can be derived by looking at descriptive statistics or combing through interview transcripts to notice where demographic and behavioral characteristics seem to cluster. For example, you may notice that your power users tend to be later in their careers, reside in cities, and have higher incomes—and you could create a persona from this—“The Seasoned Urbanite,” for example. But if you have survey data, you can use statistical techniques like cluster analysis to create more accurate groupings.
Personas can help inform the development of a product or be used later to mine insights and understand survey feedback. Either way, they’re a great lens for making sense of research results, and essential for understanding how data relates back to real customers.
2. Iterative Design Process for Continuous Improvement
Embracing an iterative design process is key to creating SaaS products that satisfy end users and meet market demands. Your research may reveal that users require software integrations or features that aren’t initially obvious. Or you may discover that your efforts to streamline a tool actually stripped away too much functionality.
Baking an iterative design process into your workflow can help make end products more refined and enjoyable to use. Having a system for prototyping can also help in this regard. It can be a difficult balance, but try to get a feel for developing prototypes with enough fidelity to demonstrate ideas but not so complex that they take forever to create.
3. Feature Prioritization
To create the first version(s) of a product that isn’t bloated or challenging to use, some functions and features will naturally need to be prioritized over others. Early in a project, user needs analysis and consideration of strategic business objectives can help you understand what should be prioritized.
Some of the most common prioritization frameworks in UX are MoSCow (must-have, should-have, could-have, and won’t-have) and Kano. Although they’re not exclusively used for UX, they’re versatile enough to apply to the entire project, a small piece, or even during revision and redesign phases.
4. Cross-Functional Collaboration
Getting quality data from UX research involves leveraging the different talents of team members. Not every designer has analytical skills, and most data scientists won’t understand all of the nuances of design. It’s easier said than done, but effective cross-functional collaborations are key to doing effective research.
In practice, that could mean having product teams sign off on surveys or attend research presentations to create buy-in. If you want the research to have an impact, make sure it is collaborative and actually shared with those who make your product.
The Key to Making SaaS User-Centric
Great SaaS prioritizes users, and quality research is the key to understanding them. Great products cannot exist without great design, but great design can’t exist without understanding your end user.
While there are many tactics for conducting UX research, there’s no one “right” method. In-depth interviews, advanced usability testing, surveys, competitive analysis, and advanced analytics all play a role. But that’s only half of the story. To be truly effective, research findings have to be combined with strategic thinking to make use of any data you gather.
At Neuron, we’ve incorporated research into our design process since our inception, and research is a critical pillar in our SPARC service. We can help you implement a research program quickly and efficiently—you don’t need in-house expertise or additional staff. If you’re serious about using research to uncover UX insights, reach out to the team at Neuron today.
Comments