There is a particular kind of digital product that becomes more difficult to use as it becomes more capable. It usually begins with a relatively clear purpose, a defined audience and a manageable set of features. Over time, more data becomes available, new requirements emerge, additional customers bring different expectations and the product gradually expands to accommodate them. Each addition makes sense in isolation, but the cumulative effect can be an interface that asks people to understand far more than they should need to.
This is especially common in products built around data. Analytics platforms, operational dashboards, insurance systems, fleet management tools and enterprise applications often contain genuinely sophisticated technology. They can process enormous amounts of information, reveal patterns that would otherwise be invisible and support decisions with significant commercial consequences. Yet the experience of using them can become increasingly fragmented as the capability grows.
I find this one of the most interesting challenges in digital design because the answer isn't necessarily to make the product simpler. The underlying complexity may be precisely what makes it valuable. The challenge is to create an experience that gives people access to that sophistication without requiring them to navigate all of it at once.
More information doesn't necessarily create more understanding
Data-rich products have a natural tendency to display what they know. If a platform can measure something, there is often an argument for making that measurement visible. Dashboards accumulate charts, tables, filters, alerts, comparisons and performance indicators until the screen begins to resemble a demonstration of the technology rather than a tool designed around a person.
The difficulty is that information and understanding are not the same thing. A dashboard might contain everything somebody could possibly need and still make it unnecessarily difficult to answer a relatively straightforward question. What has changed? Is something performing as expected? Where should I focus my attention? Does this require action?
Those questions are much more useful starting points for design than deciding how many metrics can fit on a screen. They shift the emphasis from displaying data to supporting decisions, which changes how information should be organised and presented.
A fleet manager investigating unusual vehicle activity needs a different experience from an executive reviewing performance across an entire operation. Both may rely on the same underlying information, but they approach it with different responsibilities, levels of expertise and expectations. Designing one enormous dashboard for everyone can appear efficient from a development perspective while making the experience less useful for both.
The most important design decision may therefore happen long before visual design begins: understanding who is using the product, what they are trying to achieve and which information actually helps them achieve it.
Start with the decision, then work backwards
When approaching a complex interface, I find it useful to begin with the decisions people need to make rather than the data the system is capable of producing. This requires understanding the work surrounding the product: the tasks people perform, the questions they regularly ask, the decisions they are responsible for and the consequences of getting those decisions wrong.
For example, somebody monitoring risk may need to identify an emerging issue quickly, understand whether it is significant and decide what action to take. A historical trend might provide useful context, but it shouldn't necessarily compete for attention with something requiring an immediate response. Equally, an executive reviewing performance may need a reliable overview before exploring the reasons behind a change.
These distinctions create a hierarchy. Some information needs to be immediately visible, some becomes useful when investigating further and some only matters in particular circumstances. Once those relationships are understood, the structure of the product can begin to reflect how people think and work rather than how the underlying databases happen to be organised.
This principle is consistent with the GOV.UK Service Manual's guidance on designing around user needs and the complete task somebody is trying to accomplish. A good service should not require people to understand the internal structures of the organisation or technology delivering it. The same thinking is valuable in commercial software, where the complexity of systems, teams and data sources can otherwise become visible to users who simply want to do their jobs.
Hierarchy is more important than decoration
When a dashboard feels overwhelming, there is often an instinct to improve it visually. Introduce more space, refine the typography, simplify the colours or redesign the charts. These things can help, but they won't resolve a fundamental problem with the information architecture.
If everything has equal prominence, nothing has priority. A beautifully designed interface can still be exhausting if every metric, notification and control appears to demand the same level of attention.
Good hierarchy begins with understanding significance. A critical change in performance shouldn't compete visually with a routine statistic. A primary action should be distinguishable from a secondary control. Information that establishes context should support the decision rather than obscure it. The objective is to make the structure of the interface correspond with the structure of the user's task.
This doesn't mean that every screen needs one enormous headline number and a collection of smaller charts beneath it. Complex users often need to compare multiple pieces of information simultaneously, and oversimplifying that relationship can make the product less useful. The challenge is to create enough order that somebody can understand the screen quickly while retaining the depth required for serious analysis.
This is where visual design and product strategy become inseparable. Typography, spacing, alignment, colour and composition aren't simply aesthetic decisions. They establish relationships between information, influence where attention goes and help people understand what matters.
Reveal complexity when it becomes useful
One of the most valuable principles in interaction design is progressive disclosure: presenting the information or controls people need initially while allowing them to explore more specialised functionality when it becomes relevant. Nielsen Norman Group has written about this approach as a way to make sophisticated applications easier to learn without removing advanced capabilities.
The idea is particularly useful for data-rich products, although it requires judgement. Hiding information indiscriminately can make expert users slower, particularly when they need to compare multiple variables or move quickly between related tasks. The objective isn't to reduce the number of things visible at all costs. It is to understand which information belongs together and when deeper exploration adds value.
A well-designed product might begin with an overview that identifies changes or exceptions, then allow somebody to investigate the underlying data, compare periods, adjust filters or examine individual records. The experience becomes progressively more detailed as the user's questions become more specific.
This creates a different relationship with complexity. Instead of confronting someone with every possible capability, the product supports a natural sequence of understanding. It helps the user recognise something important, investigate why it is happening and decide what to do next.
The complexity remains available, but it no longer dominates the first encounter.
Data visualisation should explain, not impress
Data visualisation is another area where sophisticated products can become unnecessarily complicated. The availability of charting libraries and interactive components makes it relatively easy to create visually impressive dashboards. The harder question is whether those visualisations help somebody understand the information more accurately or efficiently.
A chart should have a reason to exist beyond making a screen look analytical. Sometimes a simple comparison is more useful than an elaborate visualisation. Sometimes a table is the correct choice because the user needs precise values rather than an impression of a trend. In other circumstances, a well-designed chart can reveal relationships that would be extremely difficult to identify in rows of numbers.
The decision should come from the question being asked. Are we showing change over time, comparing categories, identifying an exception, understanding distribution or revealing a relationship between variables? Different questions require different visual approaches, and choosing the wrong one can make accurate information surprisingly difficult to interpret.
Consistency matters here too. If colours, labels, scales and interaction patterns change unpredictably between screens, users have to relearn the visual language every time they move through the product. A considered data visualisation system can establish recognisable conventions that make information easier to interpret while allowing different views to serve different purposes.
The most successful visualisation is not necessarily the one that attracts the most attention. It is the one that helps somebody see something important that they might otherwise have missed.
Trust is designed into the experience
In data-rich products, clarity isn't only about making information easier to read. It also influences whether people trust what they are seeing.
A beautifully presented number is of limited value if the user doesn't understand what it represents, when it was last updated or whether it can be relied upon. The same applies to alerts, predictions and recommendations. If a system identifies a potential issue but provides no useful explanation or context, the user may struggle to decide whether to act.
Trust often depends on relatively unglamorous details: clear terminology, consistent units, understandable calculations, appropriate explanations, visible data freshness and honest treatment of uncertainty. In products where information can affect financial, operational or safety-related decisions, these details become particularly important.
This is also where the distinction between simplicity and simplification matters. Removing qualifications, hiding uncertainty or compressing a complicated situation into a single score can make an interface look cleaner while making it less useful. The better approach is to communicate complexity in a way that people can understand, including where the information has limitations.
A confident interface doesn't need to pretend that every answer is simple. It needs to help the user understand what is known, what remains uncertain and what the information means for the decision in front of them.
Design systems create room for better decisions
As digital products grow, inconsistency can become another source of complexity. Different teams introduce slightly different tables, filters, forms, navigation patterns and visualisations to solve similar problems. Over time, the product begins to feel like several applications sharing the same login.
A design system can help establish consistency, but its value goes beyond maintaining a library of components. It creates a shared approach to how the product behaves, how information is structured and how common problems should be solved. When teams have agreed principles for hierarchy, interaction, accessibility, data visualisation and language, they can spend less time repeatedly resolving familiar questions and more time improving the parts of the experience that genuinely require new thinking.
The danger is treating the system as a collection of rules that must be applied regardless of context. Complex products often contain specialist workflows that need flexibility. A useful design system provides a strong foundation while allowing appropriate variation where the task demands it.
The objective is coherence, not uniformity. People should recognise the product's underlying logic as they move between different areas, even when the work they are doing changes significantly.
The real measure is what people can do with the information
It is tempting to judge a digital product by the sophistication of its technology, the number of features it contains or the visual quality of its interface. Those things can all contribute to its value, but they don't necessarily tell us whether the experience is successful.
A more useful measure is what the product enables somebody to do. Can they recognise a problem sooner? Understand a complicated situation more confidently? Complete a task with fewer unnecessary steps? Make a better-informed decision? Explain what they have found to somebody else?
These outcomes require more than polished screens. They depend on product strategy, user research, information architecture, interaction design, visual communication and an understanding of the real environment in which the product is used.
For businesses developing sophisticated technology, this creates an important opportunity. The underlying capability may be difficult for competitors to reproduce, but that advantage only becomes meaningful when customers can understand and use it. A clearer experience doesn't diminish the technology. It makes its value more apparent.
The most effective digital products don't eliminate complexity simply to create an impression of simplicity. They organise it, explain it and reveal it when it becomes useful, allowing people to benefit from sophisticated systems without having to think like the people who built them.
Complexity can be valuable. Using it shouldn't feel complicated.
Gareth.