Some usability problems are writing problems. Others only look like writing problems because the underlying workflow is asking too much of the user.
My approach to UX content starts with the decision or task someone is trying to complete. I look at the situation from the user's perspective, then consider what the system can do to help them move forward confidently.
That can mean rewriting a confirmation message. It can also mean redesigning statuses, turning technical error codes into actionable guidance, or replacing a document-based request process with a guided workflow.
| The goal isn't simply clearer language. It's reducing the work required to understand what to do next.
The existing confirmation message required users to interpret vague language before deciding whether to continue. I rewrote the message so the consequence of the action was explicit, simplified the final question, and separated explanatory information from the decision itself.
| UX decision: Tell users what will happen before asking them to commit to an action.
Workflow statuses serve two audiences at once: the people doing the work and the people waiting for it.
I revised status language to distinguish between actions and resulting states, using active language where someone needed to act and passive language where the status represented what had already happened.
| UX decision: Status language should tell people where work stands without requiring them to understand the system behind it.
An internal production application returned numeric error codes that were meaningful to Engineering but not to the operators responsible for completing the work. When Engineering wasn't available, an unexplained error could delay processing.
Rather than remove the technical information, I designed the message for both audiences. The error code remained available for Engineering, while operators received an explanation of the problem and guidance they could use to troubleshoot it themselves.
| Design principle: Preserve the information experts need while making it useful to everyone else.
Not every usability problem can be solved by rewriting the interface.
Sometimes the difficulty comes from the structure of the task itself. When the process makes users do work the system could handle for them, clearer instructions can only do so much.
The problem isn't the wording.
The interaction itself needs to change.
Customers requesting custom work were given a lengthy document containing questions for many different types of requests. They had to determine which questions applied, provide the relevant information, and return the completed request by email.
I redesigned the process as a guided Typeform experience using conditional logic. Customers selected the type of work they needed and were shown only the questions relevant to that request. The workflow also prevented unrelated requests from being combined into a single submission and provided links to relevant knowledge-base guidance when additional explanation was useful.
Completed submissions were connected through Zapier to create Jira tickets, moving structured information directly into the team's operational workflow instead of requiring employees to interpret and manually transfer information from an email.
Conditional logic presented customers only with questions relevant to the type of work they were requesting.
THE LOGIC BEHIND THE EXPERIENCE
Branching logic handled the complexity behind the interface, routing customers through the appropriate questions
while keeping the visible experience focused.
| Design principle: Move complexity into the system when the system can handle it better than the user.
Across these examples, the scale of the solution changed, but the starting question remained the same: What does the user need in order to move forward?
Sometimes the answer was clearer language. In other cases, improving the experience meant changing what the system communicated or asking the system to carry more of the burden itself.
| Good UX content doesn't stop at the words when the words aren't the real problem.
_____________________________________________________________________________________