Globe with pictures of people, global communication network

Microsoft XC Research

Stop making assumptions about language: How user research can support better language choices

Share this page

By Sheila O’Hara (opens in new tab), Holly Ambler (opens in new tab), and Veronika Hanson (opens in new tab)

Illustration of a tablet with a man popping out of a browser, and a cellphone with a female waving hello. In between are chat bubbles with varying symbols and artworks

When a customer uses one of our products to complete a task, they interpret visual and textual cues in the user interface (UI) to “make meaning” so they can understand and use the tools we provide. It’s our job as writers and designers to help them do this efficiently and successfully.

If we learn how our customers think about what they’re trying to do and reflect that in how we structure the content and the language we use, we make it easier for them to understand and use our products. This means doing research rather than relying on assumptions and instincts. After all, we are not our customers.

Over the last year, our writing and research teams partnered to conduct surveys, moderated and unmoderated studies, and roundtable discussions to understand how language affects user experience and success across several Microsoft admin products.

The role of language in product design

First, a quick review of five things language can do as part of UX design.

  1. Support ease of use. Consistent terminology, style, and content patterns create UI that is scannable and predictable.
  2. Add clarity and readability. Plain, descriptive language makes what we’re saying clear and easy to read. Plain language means using words that are in most people’s common vocabulary or that are in the vernacular for your audience. It means not making up a new term for a feature or functionality if we can use a common word to identify it and a description to clarify what it does or means.
  3. Create an information scent that supports wayfinding. Language guides customers with structure (chunking and ordering information) and labeling that matches how customers think about a task or activity.
  4. Build trust. When we describe features from the perspective of how they benefit customers and what will happen when they take action, we build trust and confidence.
  5. Support and convey our brand. Language that reflects the brand voice feels consistent and reassures the customer. Conversely, off-brand language creates dissonance and erodes trust.

While the studies we conducted around these topics were focused on IT admins’ use of our products, we believe these insights are pretty universal. We learned that:

  • Consistent terminology is the most important element in making terms understandable. Users can infer the meaning of and relationships between visual elements based on gestalt principles* and prior knowledge, but unfamiliar terms—or worse, the same term used in a different way in different locations—are much harder to interpret.
  • Unambiguous labels, headers, and descriptions are critical to user success. In one test task, participants were asked to choose between the synonymous terms “Retire” and “Wipe”. Forty-five percent of participants made the wrong choice. In the real world, that mistake would erase all data from an employee’s personal phone.
  • Industry-specific and relevant language drives satisfaction and engagement. Avoid jargon and brand names. In another study, we provided users with examples and anti-examples of our naming principles, to validate them. For example, we compared terms like Organization profile vs. Tenant profile, and Bills & payments vs. Invoices. In both cases, we saw a significant number of participants voting for plain language terms—in this case, Organization profile and Bills & payments.
  • Internal terminology does not always match customer mental models. Even when you’ve applied all the principles, you’ll still want to check naming assumptions with customers. They are the best judges of what terms will provide the greatest clarity.

*Gestalt principles: Principles/laws of human perception that describe how humans group similar elements, recognize patterns, and simplify complex images when they perceive objects.

When less isn’t more

Users have also given us valuable feedback on UX content that revealed key principles around what provides clarity and what leads to confusion:

  • Tooltips that repeat a label, that use opaque labels, or are simply numeric error codes frustrate customers and impede task success.
    • Customers consult tooltips because they’re seeking more context and explanation; repetition of labels does not add clarity or improve understanding.
    • Error messages need to go beyond a numerical value and tell the user what went wrong, where it went wrong, and ideally, how to recover.
    • Opaque, rather than descriptive label names, introduce ambiguity and increase cognitive load.
  • Just how valuable is a descriptive label? We’ve learned that customers prefer descriptive labels, even if the descriptive labels are longer. Verbose is better if it provides clarity. In one example, users preferred “allow users to change settings” over “setting modification” for this reason.

The importance of proximity

A recent customer roundtable session demonstrated how important it is to provide customers the information they need in the places they need to access it—namely, within the console. Anticipating users’ questions and answering them inline plays a vital role in the user experience. One of our discoveries based on these discussions:

  • Pain point: Leaving the workflow to consult documentation increases time on task and decreases satisfaction. (It might prevent users from completing a task on the first try).
  • Solution: Provide the information needed to complete the task in the UI. (Tooltips and information in the interface are two ways to do this.) If the information can’t be presented in the UI, put it in proximity through means such as contextual help.

All these points contribute to the idea that the language is the UX—that the content is inseparable from the user experience.

By sharing what we’ve learned in this space, we hope to build empathy for customers who must make sense of our word choices. We encourage you to partner more with research partners to understand the language that best matches mental models. The findings will equip you with the knowledge you need to design delightful, cogent user experiences.

How does language play a role in the user interface of the products you design? Has your organization conducted research to uncover insights that you’d like to share? Tweet us @MicrosoftRI (opens in new tab) or like us on Facebook (opens in new tab) and join the conversation.


Shelia O’Hara leads a team of passionate, dedicated, customer-centric UX writers who design content experiences that empower Microsoft 365 admins with clear, coherent, and helpful information when and where they need it—to remove barriers and help them get their jobs done. Her work as a content designer has spanned consumer and commercial products in a wide range of industries, including non-profit and government, productivity software, small business tools, and financial apps. The common thread winding through these experiences is the importance of language in supporting comprehension, building confidence, and inspiring loyalty and trust.

Holly Ambler is a senior UX researcher on the Microsoft Endpoint Manager Intune team. She began her career as a technical writer, editor, and content strategist, but after tiring of writing documents that helped users recover from errors born of bad design, switched paths and entered the Human-Centered Design and Engineering program at the University of Washington. Holly earned her master’s in 2010 and joined Microsoft in 2018. She’s delighted to contribute to empathetic, thoughtful user experiences as part of the Intune research team. 

Veronika Hanson is a user researcher on the Microsoft 365 admin center team. She uses both qualitative and quantitative research methods to understand IT admins’ needs and evaluate their patterns of interaction with our product. Veronika is passionate about learning how customers approach their work and observing behaviors that lead to “aha” moments. Through daily collaboration with UX designers and content writers, her work ensures that customer insights are reflected in design, and their experiences become more efficient and joyful.