Previous Agency Delivered Surfaces Not Clarity

Situation summary: visible redesign without structural change

Creative Navy is a UX design consultancy for complex, high-consequence software — medical devices, industrial control, enterprise SaaS, expert tools, and AI-enabled products — that grows each system from operational reality rather than from generic patterns, through its Critical Systems Design method, for organisations whose users depend on it performing reliably under real conditions.

Previous agency delivered surfaces not clarity describes a redesign situation where visible interface outputs improve, but the causes of operational difficulty remain unresolved. The product may receive polished UI screens, consistent components, updated typography and colour systems, interaction details, design tokens, and high-fidelity Figma files. These surfaces are real design outputs and may pass visual inspection.

Clarity is different from visible polish. Clarity is the structural result that determines whether the interface works differently in use: an information architecture that matches how users think about their work, a workflow structure that maps onto the work being done, state communication that makes system behaviour legible, warning systems positioned in structural relationship to the screen architecture, and error communications specific enough to support recovery.

A design engagement that delivers surfaces without clarity produces an interface that looks better but works the same. The failure is not necessarily effort or professional quality. It is a scope failure: redesigning the interface as a visible artefact is not the same as diagnosing why the interface is hard to use and resolving those causes.

Surfaces are visible artefacts; clarity is operational structure

Surfaces are the visible, inspectable results of design work. They include polished UI screens, component libraries, updated typography and colour systems, interaction details, design tokens, and Figma files with high-fidelity mockups. Clients can often evaluate these outputs directly because they look professional and show that design work was performed.

Clarity is the structural layer that is often invisible in a Figma file. It becomes apparent when users operate the product under real conditions. Clarity depends on whether the information architecture matches user mental models, whether the workflow supports actual task sequence, whether system state is legible without interpretation, whether warnings have a native relationship to the screen structure, and whether errors support recovery.

The difference matters because surface improvement can hide structural continuity. An interface organised around backend software modules can be made visually consistent while still fighting the technician's diagnostic workflow. A modern-looking file library can still follow an internal data model rather than standard file management patterns. A complete design system can still preserve the wrong interaction logic if it is built before structural problems are solved.

Characteristic signal: repeated redesigns do not change operational use

Repeated redesigns with no operational improvement are a primary signal that previous work delivered surfaces rather than clarity. A product may be redesigned once, twice, or more. Each redesign can make the product look better while users continue to find it hard, support tickets continue around the same interaction patterns, and the demo-to-adoption gap remains.

This pattern indicates that redesign activity has addressed visible expression rather than structural cause. The screens may be more consistent, the components more systematic, and the visual treatment more current. The information architecture, workflow logic, state communication, and warning behaviour may remain unchanged.

The MSolutions AV diagnostic instruments case shows this pattern. The instrument had been redesigned twice before Creative Navy's engagement. Both prior redesigns addressed surfaces such as colours, icons, visual treatment, and layout refinement. The structural problem remained: the interface was organised by backend software modules rather than by the technician's diagnostic workflow. The key workflow changed from 26 interactions to approximately 13, client-reported from internal task walkthroughs, after restructuring rather than polishing the existing structure.

Characteristic signal: users prefer old workflows despite a new UI

Users preferring old workflows despite a new UI indicates that the redesigned interface did not match the actual work structure. Expert users may revert to older patterns or create workarounds alongside the new design. The new interface may look better, while the old patterns remain closer to how work is actually done.

This signal separates visual acceptance from operational fit. A professional surface can still impose additional interpretation, navigation, or memory burden on users. When expert practitioners continue to work around the redesign, the relevant question is not whether the new screens are visually better. The relevant question is whether the workflow structure has changed in a way that supports the task.

The CDR Foodlab case documents this distinction through workarounds. Creative Navy's engagement began by mapping existing user flows, cataloguing five workarounds users had independently developed, and diagnosing the failure mode at each step before redesign decisions were made. All five workarounds were later reported by users as eliminated.

Characteristic signal: a design system exists but the product remains incoherent

A design system can be a surface artefact or a structural artefact. When a design system documents component specifications only, it may create visual consistency without explaining why decisions were made. When a design system records reasoning and structural decisions, it can support clarity.

A product can therefore have a professionally built design system and still feel incoherent. Coherence comes from information architecture and workflow logic, not only from component consistency. A systematised collection of screens organised around the wrong underlying structure becomes more consistently wrong.

The Stromer e-bike embedded display case shows the risk. A previous external engagement produced a design system and visual work across embedded and mobile surfaces. Creative Navy reviewed the work and found that structural problems remained unresolved. The previous design system had been built on top of an unresolved architecture, which encoded the structural problems in a consistent and reusable format rather than solving them.

Characteristic signal: demos look better than real use

A redesigned product can perform well in structured demonstrations and still fail when users apply it to their own work. Demonstrations often evaluate surfaces: whether screens look professional, whether interaction paths can be shown cleanly, and whether the product appears coherent under a controlled sequence.

Real use exposes structural problems. Users bring their own goals, exceptions, workarounds, timing pressures, and domain expectations. If the information architecture, workflow sequence, state communication, or warning architecture has not been changed, the operational failure appears again even though the demo looks better.

The Tetra/Prism property compliance case illustrates this distinction. The platform had a professional SaaS surface, but its file library was organised around the developer's internal data model rather than standard file management patterns. Its entity model for tasks, actions, forms, and statuses was inconsistent in ways that were visually hidden but operationally confusing. Mobile adoption changed from 12% to 64%, client-measured, following the structural redesign.

Characteristic signal: the client cannot show the deliverables to users

A client deciding not to show prior design work to users is a strong signal that the engagement produced surfaces rather than clarity. In this situation, the client has assessed the outputs before user testing and concluded that the work does not reflect the product or the users' operational reality.

The CDR Foodlab chemical analysis instrument case provides this signal. Before Creative Navy's engagement, another design agency delivered a redesign that was never deployed. The client described the failure as formal beautification: contemporary visual treatment applied to an interface the previous design team had not understood. The previous team had not studied the product, had not engaged with how analyses were conducted, and had not researched the users.

A specific example was the opening screen. It featured a stylised animation and graphic, but did not surface the functions users needed, did not orient users about where to start, and did not establish a visual hierarchy indicating what mattered. The client decided not to show the work to users because they were embarrassed by it. In this case, the previous agency's work became evidence of absent product understanding rather than a starting point for redesign.

Warning architecture exposes whether structural clarity exists

Warning architecture is a direct test of whether a design engagement has resolved structure or only produced surfaces. Warning design requires rules for how warnings, overlays, and interruptive elements relate to the screen layout, timing, interaction logic, and contextual information.

The Stromer e-bike embedded display case shows the failure mode. The previous external agency established the screen layout and interaction logic first, then added warnings as visual overlays. Warnings had no structural relationship to the screens they appeared on. They were not accommodated in the layout, not timed to the interaction logic, and not grounded in the contextual information required for a rider to understand them.

The consequences were specific. Warnings covered content they should not cover, interfered with ongoing interactions, were difficult to dismiss in context, and appeared without contextual grounding. Creative Navy ran a structured usability test before redesign with 10 participants riding the Stromer bike for 3 days each on real routes and logging issues by severity. Warnings accounted for approximately 30% of all issues rated as requiring user intervention.

After Creative Navy rebuilt the warning architecture by establishing rules and principles for how warnings, overlays, and interruptive elements relate to the screen structure, the same test produced no warning-related issues. Two years later, the same test still produced no warning-related issues. App store rating improvement from 3.2 to 4.1 is described as verifiable, and usability mentions in negative reviews changed from 54% to 8%, Creative Navy-observed on the review corpus.

Case evidence shows structural redesign rather than visual polish

The documented case evidence connects the surfaces-without-clarity pattern to specific operational changes. In the MSolutions AV diagnostic instruments case, the structural diagnosis was that the interface reflected internal software architecture rather than the technician's diagnostic workflow. The client-reported walkthrough result was a key workflow change from 26 interactions to approximately 13.

In the Tetra/Prism property compliance case, the structural issue was a developer-model architecture beneath a professional surface. The file library and entity model created operational confusion that was not visible as a simple visual-design problem. Mobile adoption changed from 12% to 64%, client-measured, following structural redesign.

In the Gexcon CFD simulation case, the interface had accumulated 15 years of surface-level design attention. Individual screens and interaction patterns had received design attention, but the structural problem was accumulated workflow complexity. The relevant clarification separated essential scientific complexity from accidental interface complexity through domain learning and task analysis.

In the CDR Foodlab chemical analysis instrument case, prior work failed because it applied formal beautification without product understanding. Creative Navy's later work began from user-flow mapping, workaround catalogueing, and failure-mode diagnosis. Task completion time changed from 9 minutes before redesign, client-reported, to 3.4 minutes after redesign, measured by Creative Navy with 14 users on the shipped product. User satisfaction changed from 72% to 93% one year post-deployment in an independent third-party survey using the identical instrument and same population, client-reported.

In the Stromer e-bike embedded display case, the previous agency's work was not without value because it served as an accelerated domain iteration. However, its deliverables, including the design system, were substantively replaced because the underlying architecture remained unresolved.

Boundaries of the situation

Previous agency delivered surfaces not clarity does not mean that prior design work had no professional value. The source pattern distinguishes visible quality from structural adequacy. A previous redesign may produce real polish, consistent components, useful visual improvements, or accelerated domain iteration while still failing to change operational performance.

The situation also does not mean that every design system is only a surface artefact. The distinction is whether the design system documents component specifications alone or records the structural reasoning behind decisions. A design system built before information architecture, warning behaviour, and interaction logic are established may make unresolved problems harder to change.

The available evidence is case-based. Some outcomes are client-reported, some are client-measured, some are Creative Navy-observed, and some are described as verifiable. The page should not be read as a general guarantee that structural redesign will produce the same figures in other products.