Link Digital is lucky to have an excellent team member, Anastasiia Shevtsova, who works on the user interface/user experience (UI/UX) aspects of our data portal projects. I had a conversation with her to understand the importance of having a user-friendly data portal, especially for teams who don’t have a dedicated designer, and to share insights on best practices to improve portal usability through simple, actionable steps.
Who is Anastasiia Shevtsova as a UI/UX Designer?
“I’m a UI/UX Designer at Link Digital, and I work mostly on data portals, especially CKAN-based portals.
My work can be on different levels. Sometimes I design a portal from scratch for a new project. Sometimes, I redesign an existing portal to make it more modern and easier to use. I also help improve consistency.
For example, making sure the portal follows a design system, so everything looks and works in the same way. I also design new features, like better search, filtering, previews, dashboards, or map views. And very often, I focus on improving user flows, so people can find the right dataset faster and reach their goal.“
What are the most common UI/UX mistakes you see on many of the data portals worldwide?
“I would say that the biggest mistake is that most portals are built around the data structure and not around what people want to do on the portal.
People can come with a simple goal, but sometimes the portal doesn’t guide them properly. Sometimes search and filters are not so easy to use, and data discovery can be confusing.
Another common issue can be that the portals don’t have correct UI stats (or metrics). It’s very simple to develop those states, but I’ve seen a lot of portals that don’t have error states, empty results, errors, or loading.
There’s a common example: when a user searches for information, he would open several data sets. Sometimes it’s more than ten (10) data sets. And when the data set page is overloaded with information, and this page isn’t structured, [the] user has to spend a lot of time reading loads of text before they understand simple points, such as location, publication date, and the time period of the data. So this discovery process becomes difficult for users.“
A simple improvement is to restructure the data set page, highlight key details at the top, and help users understand the data set in a few seconds.
What about organisations running a data portal without a UI/UX expert on their team, what best practices would you recommend?
“The first thing I would advise is to focus on the top user tasks. We call them representative tasks. In simple words, a team can focus on user goals and ask about the five common tasks users do on a portal. Usually it can be searching, filtering, understanding the data set, and downloading flow. Then when we refine these representative tasks, we can make quick usability checks.“
Anastasiia shares that you don’t need a big study. Ask people to try a task and watch where they struggle.
“I like one phrase from one UX designer, Vitaly Friedman. He says small tests can reveal big issues, and that’s really true.
Also, keep the interface simple and predictable. Clear labels, consistent buttons, and good spacing can improve UX a lot.“
What are some accessible tools that organizations with a limited budget or no on-staff designer, can use to improve their site’s UI/UX?
“A lot can be done with simple tools. For quick markups of pages or portals, the best tool is Figma. It’s great for creating layouts with some clickable prototypes and it really helps to test it fast.“
For user testing, Anastasiia recommends a few tools that work really well even for small teams.
- Optimal Workshop: great for testing navigation, menus, and information structure
- Maze: good for testing Figma prototypes, with recordings and heatmaps
- Sondar and Wondering: these are newer tools, and they use AI to help analyze feedback.
They can also include real user testing, recordings, and follow-up questions.
Even if a team doesn’t have a big budget, doing just a few small tests like this can quickly show what users find confusing.
What are your tips for better accessibility?
Anastasiia offered the following tips:
- Use good contrast and readable font sizes so text is easy to read for everyone.
- Make sure the portal works with a keyboard, so users can tab through the page and clearly see where they are. This is called a focus state.
- Also, even if you’re not an accessibility expert, Figma can help. There are plugins like Able, which are accessibility assistants that quickly check contrast and basic accessibility issues. So teams can catch problems early, even before development.
But what if a team needs to urgently do one thing to improve their portal design this week?
“I would improve the search function. The search and discovery flow is the most important user flow on the portal.
Everything can be perfect, the design can be perfect, and the dataset page can be perfect but if users can’t find the right information or the right data set, everything becomes frustrating without the working search flow.
I would focus on such improvements as filtering, sorting, and adding the predictive search suggestions. That shows relevant datasets while the user is typing. It helps users move faster and reduces ‘no results’ situations.“
According to Anastasiia, these basic improvements can take a few days to one or two weeks, depending on the portal and what’s already in place.
What signals should portal teams look at to understand if their design is working?
Anastasiia likes to look at signals that show if users can complete their tasks.
For example:
- Are people finding data sets and downloading them?
- How often do users refine search results with filters?
- Do users leave quickly from data set pages?
- Do people return to the portal?
She also mentioned that qualitative feedback is important.
“Analytics can tell you what happens, but user comments tell you why. Vitaly Friedman often talks about designing for real behavior, not assumptions, and metrics help you see that behavior. Google Analytics is good for basic tracking. Hotjar provides recordings and heatmaps.“
What’s one piece of advice you’d give to someone who knows their data portal needs design work but doesn’t know where to start?
“My advice is don’t try to redesign everything at once. Usually I start with a simple UX review. Focus on the key user journeys, like search, data set pages, and downloading, and identify where users get stuck. Then our team forms an improvement roadmap to work on the most important tasks first. This approach we use at Link Digital helps our clients to get quick wins and build a further improvement strategy.“
Improving a data portal doesn’t require a total overhaul overnight. As Anastasiia points out, focusing on your top five user tasks and making small, iterative improvements, like refining your search filters, can dramatically improve the user experience. The key is to step back from designing for the data and start designing for the people who use it.
