Designing a clearer Learn Hub for microbiome testing
Salient Bio — a UK biotechnology company specialising in microbiome testing — commissioned a redesign of its Learn Hub to help patients and clinicians navigate complex information and build confidence in the value of its services.
The team worked collaboratively across research, information architecture, prototyping and interface design. My strongest contributions were discovery research, quantitative analysis of usability testing, information architecture, and the exploration of search and filtering patterns.
We worked to reorganise the hub into a connected information system combining guided testing journeys with direct access through search, terminology support and clinician resources.
The challenge
The Learn Hub needed to support people before, during and after microbiome testing. Users could arrive with a specific task, question or report term rather than following one continuous journey from beginning to end.
The challenge was to provide enough structure to orient users at each stage while preserving direct access for patients and clinicians with different goals and levels of knowledge.
-
Multiple stages and entry points
Users could arrive while deciding whether testing was relevant, registering a kit, preparing a sample, interpreting results, researching terminology or deciding what to do next.
-
Different user contexts and goals
Patients and clinicians approached the same information system with different responsibilities, questions and expectations of the service.
-
Accessible without losing credibility
Information needed to be understandable to patients while retaining the methodology, limitations and specialist detail clinicians required.
Discovery findings → system requirements
Discovery combined Salient Bio’s commissioned patient research, stakeholder interviews and our exploratory interviews with patients and clinicians. Two requirements shaped the Learn Hub.
Users needed support before, during and after testing
“I’d be interested, but I’d want to know whether it’s actually useful or just another weird gut bacteria thing.”
Patient participant
System requirement
Organise guidance around each testing stage, with direct routes for specific questions and report terms.
Patients and clinicians needed different levels of detail
“They are not interested in technical details; they want to know what it means to them, personally.”
GP participant
System requirement
Start with plain-language guidance, then layer definitions, evidence, limitations and professional resources.
Designing one information system for different routes and levels of detail
Instead of organising the Learn Hub solely around internal content categories, we structured it around the questions users were trying to answer and the stage of testing they were in.
Guided journey routes provided orientation, while search, contextual definitions and clinician resources gave users more direct access when they arrived with a specific question or term.
How to interpret the testing evidence
Small task samples
Most individual tasks were completed by six participants. The next-steps task had five participants.
No time-on-task data
Timing was not captured, so conclusions should focus on route clarity, completion, confidence, comprehension and observed friction.
Directional evidence
Some confidence and reassurance signals were supported by participant comments. The results should guide design decisions, not be presented as population-level proof.
Design decision 01
Strengthen entry points into decision support
Design intention. The homepage needed to help people decide whether testing was relevant without allowing one access method to overpower the guided routes.
Prototype behaviour. Search was prominent, but suitability guidance competed with other destinations and one intended “Is this test right for me?” route was not connected in the prototype.
- 4 of 6completed the task
- 2 of 6stayed close to the intended route
- 50%first-click success
“It felt buried. Expected a CTA and product/service route.”
Users found relevant content through plausible but indirect routes. The issue was route clarity at a decision point, rather than evidence that the content itself lacked value. The missing prototype link also means the result reflects both information architecture and prototype completeness.
Design decision 02
Make search results easier to interpret
Design intention. Search gave users a direct route when they arrived with a specific question, term or testing concern.
Prototype behaviour. Relevant content was usually reachable, but active filters, topic chips, suggested searches and article results were not always easy to distinguish.
- 83%search-task success
- 92%weighted usefulness
- 5 of 6said results matched expectations
- 50%experienced a filter or topic-chip issue
- 0.5average path deviation
Search worked well as a direct access route, but the results page still needed to explain its own hierarchy. The revision separates the search term, proposed quick answer, primary article results, active filters, suggested searches, related topics and routes back into the wider Learn Hub.
- 01Search term
- 02Clearly labelled quick-answer concept
- 03Primary article results
- 04Visibly active filters
- 05Suggested searches and related topics presented as separate groups
- 06Clear routes back into the wider Learn Hub
Design decision 03
Keep definitions within the user’s context
Design intention. Glossary support was intended to make specialist terminology understandable without removing the scientific detail that gave the content credibility.
Prototype behaviour. Glossary terms looked similar to surrounding text, and opening a separate glossary made it difficult to return to the original reading position.
- 33%strict glossary success
- 67%experienced a clear loss-of-place issue
- High confidenceonce definition support was located
“Link looked the same as surrounding text; could not click back from glossary to article.”
The proposed revision makes technical terms visibly interactive and opens definitions in an in-place popover or side panel. Users retain their reading position while specialist detail remains available without interrupting the main explanation.
A connected support system across the testing journey
The final concept combines guided journey routes with direct access to specific information. Patients and clinicians can move through the Learn Hub according to their stage, question and level of knowledge without being forced through one prescribed pathway.
Group 01
Patient journey
Guided routes connect deciding whether testing is relevant, kit registration, understanding results and deciding what to do next.
- 92% procedural clarity
- 67% step-order comprehension
- 100% next-step success
- 4.75/5 average next-step confidence
Group 02
Search and contextual support
Search, active filters, article hierarchy and the proposed quick-answer concept provide direct access, while contextual definitions keep technical support inside the article.
Group 03
Patient and clinician shared support
Plain-language result explanations connect to clinician follow-up, appointment questions, related Health Library content and professional resources.
- 100% result-meaning success
- 100% comprehension accuracy
- 4.5/5 average result confidence
- 100% next-step clarity
- 40% wanted more practical follow-up support
These task results describe navigation and comprehension within this usability study. They are not evidence of clinical effectiveness, long-term behaviour change or population-level outcomes.
What the design enables
For patients
Makes support across the testing journey easier to find, understand and act on.
For clinicians
Provides clearer educational material for explaining reports and discussing appropriate next steps.
For Salient Bio
Turns educational content into a connected support system rather than a larger collection of articles.
Reflection and next steps
The strongest contribution was not an individual interface feature, but an information system capable of balancing flexible access, guided journeys, scientific complexity, different levels of user knowledge and shared patient and clinician support.
The proposed refinements now need further evaluation. They should be treated as evidence-led directions rather than validated outcomes.