Anticipated Use Assessment
- Base
- AI/HF Risk and Adoption variant
A use specification reconstructed from your documents, a task analysis with use-error scenarios, and an anticipated-use gap map rated by severity, with hazard candidates delivered in DCB0129-compatible format for your own hazard log, plus an evidence gap statement.
The same, with an AI-specific lens added: automation bias, trust calibration and adoption risk alongside the standard use-error scenarios.
Where the product will be used wrongly, before anyone has used it.
For companies pre-launch or pre-pilot, with no user base yet.
Everyone else’s hazard log is built from work as imagined. This one is built from work as imagined, done properly and adversarially, and it tells you what to go and check once you have users.
Formative Usability Evaluation
Watching real participants use the device, before anyone else does.
A moderated usability study with representative users and the actual device, under controlled conditions: onboarding friction, use-error hazard analysis, and adherence to your instructions for use, aligned with IEC 62366-1 formative usability engineering. Delivered as a Formative Human Factors File.
For companies pre-market, who need real usability evidence in the Human Factors Engineering File before a summative study or a regulatory submission.
Anticipated Use Assessment reconstructs use from your documents, without a user in the room. This puts real users in front of the real device, before launch.
Deployment Assurance
- Evidence
- Safety
- Both
Integration validity findings against your claim wording.
A DCB0129-format hazard log extract and a CSO handover.
Both deliverables, from one fieldwork pass.
Add-onsAny tier. Stackable.
- Workflow and Signals pass
- AI/HF Risk and Adoption lens
A second analytical pass over the same fieldwork data: workflow and human-performance findings, what the integration-validity gaps mean for your actual decision points, and a signals-to-behaviour and load audit.
For a product with an AI or decision-support component already live or piloted. Automation bias and trust calibration findings from observed use, and right-answer-wrong-action failure modes, rated by severity.
Whether the output holds in the setting where it is actually used.
Work as imagined is reconstructed from your documents, then compared against work as done, with every gap rated by severity, from a single fieldwork pass.
For companies live, piloted, or deployed. The follow-on from an Anticipated Use Assessment.
Assessment is remote. A consultant on site for one day sees one day. A protocol run by your own staff across two weeks sees the variation, and it stays yours to rerun at every release.
Cognitive Load Mapping
- Focused
- Multi-role / sensor-rich
A cognitive-load map for a single role or task cluster, built from real tasks, transitions, interfaces, interruptions and decision points, with contributing factors, overload hypotheses, and practical recommendations: redesign, task redistribution, interface changes, or training.
The same, across multiple roles or teams, and/or with sensor-derived data.
Where the workflow is overloading the people running it, and what to do about it.
For companies whose product or workflow depends on a person's attention and judgement under load, in clinical, safety-critical or high-stakes settings.
Deployment Assurance asks whether the output holds in real use. This asks why the person using it is struggling to keep up.