Blog
How to Choose Occupational Health Software: The Complete Buyer's Checklist
Your occupational health system still works, but keeping it working takes more effort each month. Administrators copy details between tools. Clinicians search for earlier assessments. A cancelled appointment can leave a case with no clear next action. The wrong replacement could disrupt clinical work and make years of records harder to use.
Choosing occupational health software is a procurement and change project, not a feature contest. This guide offers a nine-step evaluation framework, 30 supplier questions, a weighted comparison scorecard and an implementation checklist. Use the same requirements, scenarios and evidence requests for every supplier. Record what matters before a polished demonstration changes the discussion.
Step 1: Understand why you are changing systems
Start with evidence from a normal working week. Count duplicated entries, referrals waiting for review, cases with unclear ownership, late reports and manual steps outside the system. Ask clinicians, administrators, client-facing staff and information governance leads where the process breaks. Include problems with surveillance follow-up, reporting, integrations, configuration charges and exporting data.
Write each problem as an observable outcome: “We cannot tell which accepted referrals have no booked assessment” is more useful than “We need a better dashboard”. Establish a baseline for the measures you will revisit after go-live. Separate a genuine constraint from a habit inherited from the old product. A replacement need not reproduce every old screen to preserve a sound clinical process.
Step 2: Define essential requirements before demonstrations
Classify each requirement as essential (the service cannot operate safely or effectively without it), important (a material improvement) or desirable (useful if the trade-off is reasonable). Name an owner and a test for every essential requirement. A vendor's “yes” is not evidence until someone can demonstrate the behaviour or provide a verifiable answer.
Cover referral intake and triage, case ownership and tasks, clinical assessments, configurable forms, surveillance cycles, employee history, appointments, report review and release, client and employee access, integrations, permissions and export. Describe required volumes, different client pathways and what staff must be able to configure. Keep speculative future ideas separate from today's service obligations.
Our modern occupational health software guide explores the underlying records. Here, the task is to turn those records into acceptance criteria. For example: “A clinician can view a previous surveillance result while an employer user sees only the authorised outcome.”
Step 3: Test real clinical workflows, including exceptions
Give each supplier the same short, anonymised script. Ask the person who would use the product to perform it, and record where they leave the system or need supplier intervention.
- Management referral: A manager submits a request. The service asks for missing information, accepts it, opens a case, completes an assessment, reviews the report and releases only the authorised employer-facing output.
- Health surveillance: A team identifies employees due for surveillance, issues questionnaires, records findings and assigns follow-up. Ask how a later assessment relates to the earlier one. See our health surveillance solution for the service context.
- Existing employee history: An employee changes role. A clinician needs relevant history, while the manager must not gain access to clinical notes simply because the employee is in their organisation.
- Operational exception: An appointment is cancelled or an assessment is incomplete. Show who owns the next action, what remains visible and how the work is resumed.
A tidy demonstration usually shows the happy path. Ordinary exceptions reveal whether a case has a real lifecycle or whether staff must maintain a spreadsheet beside it. Request a trial using safe test data where practical, and ask for written answers to anything the demonstration cannot prove.
Step 4: Assess security and clinical information governance
Clinical notes and an employer-facing fitness outcome serve different audiences. Ask the supplier to show access as a clinician, administrator, manager and support user. Test whether clinical detail stays separate in screens, APIs, document downloads and exports, and whether permissions are enforced when someone requests data directly, beyond the user interface. Examine individual accounts, multifactor authentication, audit history and how client or tenant boundaries are enforced. Ask how supplier staff obtain access, where data is held, and for evidence of a tested backup restoration.
The ICO's workers' health guidance says health information needs appropriate protection and access should follow a need-to-know approach. The HSE distinguishes health records from confidential medical records in health surveillance. These are legal and clinical governance considerations. Requesting a role demonstration, a recovery test result or an export sample is a procurement method, not a legal requirement in itself.
Review controller and processor roles, contract terms, retention and any relevant impact assessment with your own governance team. Check that historical surveillance records can be retained for the applicable periods, and demonstrate that report approval, release and disclosure follow your clinical procedures. An independent certification may help verify part of a supplier's practice; it cannot show that your proposed role configuration will prevent an employer seeing clinical detail. Our security overview gives further context for questions to ask Occentra.
Step 5: Calculate the full cost of ownership
Request a three-year quotation against the same scope from every supplier. Include setup, subscription, user or activity charges, configuration, bespoke development, migration, training, integrations, support, contractual minimums, annual increases and exit assistance. Add your team's time for cleaning data, testing, training and running two systems during cutover.
Three-year total cost = one-off setup and implementation + migration + training + integration work + 36 months of recurring charges + expected change and exit costs + internal project effort.
For an illustration only, a quoted £900 monthly subscription over 36 months is £32,400. If a hypothetical project also needs £6,000 of implementation, £4,000 of migration and £3,000 of internal time, its three-year estimate is £45,400 before any further charges. These figures are not supplier quotes. Compare scope and assumptions alongside the result. Our UK software pricing guide covers pricing models in more depth, and Occentra's pricing shows its current plans.
Step 6: Examine migration and the future exit
Ask for a source-to-target map before accepting a migration estimate. Check clients and employees, referrals, historical cases, assessments, documents and attachments, surveillance results, reports and audit history. Preserve original documents and the provenance, authorship and timestamps of historical records. Relationships matter: an assessment or attachment without its employee and case context may be technically imported but clinically hard to use.
Agree a representative sample containing active, closed and awkward records. Define who checks that migrated clinical records are accurate and retrievable only by authorised users, and how defects are corrected. Some history may need a governed archive instead of conversion. Our migration guide covers that project in detail.
Also inspect the exit clause now. Confirm data ownership, usable export formats, attachments, metadata, timing, assistance, deletion evidence and fees. A promise that “you can export your data” is too vague to price or test.
Step 7: Evaluate the supplier behind the product
Ask who handles implementation, support and incidents; how changes are prioritised; how releases are tested and communicated; and how the service would recover from an outage. Examine service commitments, escalation routes, contract terms and the product's direction. Request evidence proportionate to your organisation's dependence on the system. Company size alone does not establish reliability.
Listen for specific limits. A credible supplier can explain which workflows are configurable, which need development, what is available today and how a requested change reaches production. Put material answers into the contract or implementation statement rather than relying on a demonstration conversation.
Step 8: Plan implementation as a controlled change
Assign a clinical lead, operational owner, data owner and supplier contact. Agree acceptance criteria and decisions for each phase:
- Confirm requirements and the workflows that must be live first.
- Configure roles, clients, forms, pathways and environments.
- Map source data and agree retention and archive decisions.
- Test a representative migration and reconcile counts and relationships.
- Validate clinical and operational scenarios, including exceptions and permissions.
- Train each role on its actual work and record unresolved questions.
- Go live with named support, a cutover checklist, fallback procedures and a list of urgent or outstanding clinical work.
- Review defects, workarounds and measures after the first live period.
A phased rollout can reduce the size of a cutover, but running two systems creates reconciliation and ownership questions. A single cutover may be clearer for some services, but needs stronger rehearsal and contingency planning. Decide from referral volumes, clinical continuity, migration quality and the ability to keep records consistent. Agree how clinicians will access essential information and handle urgent work during an outage, migration and cutover. Specify who can stop go-live if a critical test fails.
Step 9: Measure whether the change worked
Compare post-launch performance with the baseline from Step 1: referral processing time, outstanding cases, report turnaround, administrative handling time, user adoption, manual workarounds, migration accuracy and support incidents. Ask clinicians and operations staff where the new process still requires reconstruction or duplicate entry.
Review at agreed intervals, such as the first week, month and quarter. A completed cutover is not proof of a successful implementation. The test is whether authorised people can complete ordinary and exceptional work with less uncertainty and without weakening clinical safeguards.
30 questions to ask occupational health software suppliers
Use the same questions and test scenarios for every shortlist demonstration. Copy the questions into an evaluation sheet, with one row per question and supplier. Record the answer, evidence seen or requested, verification status (verified, partial or unverified), and any outstanding action with an owner and due date. Close open actions before scoring. Where a requirement is critical, ask for a working demonstration or contract commitment rather than a verbal assurance. You can also print this page for use during a demonstration, without providing an email address.
A. Clinical and operational workflows
These questions test whether the product connects the request, clinical work and authorised output. Strong answers show the full lifecycle and a real exception without relying on off-system notes.
- How does the system handle the complete referral-to-case lifecycle?
- Can clinical forms and assessment workflows be configured without bespoke development?
- How does the system support ongoing health surveillance and historical employee records?
- How are clinical reports prepared, reviewed and released?
- What happens when an assessment is incomplete or requires follow-up?
B. Security and information governance
These questions establish who can see sensitive information and how access can be explained later. Ask for a role-by-role demonstration and relevant written evidence. If AI features are offered, establish whether they process clinical information, whether that information is used for model training, what safeguards and data-processing arrangements apply, and who reviews any AI-generated clinical output before use.
- How are confidential clinical records separated from employer-facing information?
- How are user roles and permissions managed?
- What audit history is available for sensitive actions?
- How is data protected in transit and at rest?
- What independent assurance or evidence is available for security controls?
C. Configuration and integrations
These answers reveal the cost and delay of routine change. A useful response distinguishes supported settings, integration work and bespoke development, including who maintains each.
- Which workflows can administrators configure themselves?
- Which changes require paid supplier development?
- What APIs or integration mechanisms are available?
- Can the system support different clients without custom development?
- How are configuration changes tested and released?
D. Migration and data portability
Ask for samples of an import plan and an export. Strong answers preserve record meaning, attachments and relationships, and identify who validates the result.
- What information can be imported from our existing system?
- How are historical clinical records and attachments preserved?
- How will migration completeness and accuracy be verified?
- Can we export our information in usable formats?
- What happens to our data when the contract ends?
E. Commercial and supplier considerations
These questions make quotes comparable and expose dependencies after signing. Look for a scoped price, support commitments and a recovery plan appropriate to the service.
- What is included in the subscription?
- Which services or features incur additional charges?
- What implementation and training support is included?
- What support commitments and escalation processes are offered?
- What business continuity and disaster recovery arrangements exist?
F. Implementation and long-term operation
A strong supplier can describe the work on both sides, show a test environment and agree how success will be measured after launch.
- What does a typical implementation involve?
- What responsibilities remain with our organisation?
- How can we test our workflows before go-live?
- How are software updates introduced and communicated?
- How will we measure whether implementation has succeeded?
Supplier comparison scorecard
Set the weights before scoring demonstrations. The example below totals 100%; change it to reflect your service. Score each area from 1 (poor fit or weak evidence) to 5 (requirements met and demonstrated). The supplier scores are hypothetical and use a five-point scale.
Scroll table to see all columns →
| Category | Weight | Supplier A (1–5) | Supplier B (1–5) |
|---|---|---|---|
| Clinical and operational fit | 30% | 4 | 5 |
| Security and information governance | 20% | 5 | 3 |
| Configuration and integrations | 15% | 3 | 4 |
| Migration and portability | 15% | 4 | 2 |
| Total cost of ownership | 10% | 3 | 5 |
| Supplier support and continuity | 10% | 4 | 3 |
| Weighted total | 100% | 79% | 76% |
Weighted score = (category score ÷ 5) × category weight. Add the six results for a score out of 100%. Supplier A's weighted contributions are 24, 20, 9, 12, 6 and 8 percentage points, totalling 79%. Supplier B totals 76%. These are examples, not assessments of real suppliers.
Keep the evidence and verification status beside each score. A function demonstrated with your test data carries more confidence than an unsupported verbal assurance; request proof and revisit uncertain scores. A high total cannot excuse failure of an essential clinical or security requirement. Treat those as pass/fail gates, document the reason for any exclusion, then compare viable options. The scorecard supports a decision; it should not replace professional judgement about implementation risk.
Where Occentra fits
Occentra connects referrals, cases, forms, appointments and reports in one operational system, so staff can follow the work from request to authorised outcome without reconstructing it across disconnected tools. Supported workflow and form configuration can handle routine operational changes without making each change a bespoke development project. Clinical work and employer-facing information have distinct access boundaries. Ask to see how those boundaries work for your roles and pathways.
Evaluate Occentra against the same 30 questions and pass/fail requirements as every other supplier. Confirm what is available today, and the scope and effort for surveillance pathways, migration, integrations and exit. Explore the platform. For a closer look, Book a walkthrough using one of your clinical workflows and an operational exception.