When people talk about improving a health information service, the conversation usually jumps straight to medical content. But a big share of what makes a service feel trustworthy is plain administration: how quickly a question gets an acknowledgement, how clearly instructions are written, how easy it is to find the next step. None of that is clinical, and a lot of it has been refined for decades by hotels, airlines, banks and delivery companies. This article is an informational checklist for auditing the administrative side of a health-related website or practice desk by borrowing from those industries. Medical advice is out of scope here, and so is anything about treatment; the subject is purely how information and requests are handled.
How to use this checklist
Block out about three-quarters of an hour. Open the pages and messages your visitors actually see, in the order they see them, and tick each item below as pass, partial or fail. Write one line of evidence next to every answer, such as a page address or a screenshot filename. Each partial or fail then moves onto a short fix list, ranked by how many visitors the problem touches.
You can run it alone, though it works better with a second person who has never used your service. Fresh eyes spot the places where an insider fills in missing information without noticing.
Section one: the pre-arrival message, borrowed from hotels
A hotel sends a message before you arrive so you know what to bring, how to find the entrance and what happens on the first day. Health information services can do the equivalent for their own processes.
- Is there a short message sent right after someone submits a request, stating what happens next and roughly when?
- Does it say what information the visitor may need to have ready?
- Does it say clearly what the service does not do, such as giving emergency help?
- Is there one obvious way to reply if something was entered incorrectly?
Score the message from the visitor’s side. If you received it without any context, would you know what to do next?
Section two: signage, borrowed from airports
Airports are masterclasses in wayfinding. Large type, consistent icons, and signs placed exactly where decisions are made. For a website, the signs are menu labels, button text and headings.
- Do menu labels use everyday words, not internal terms?
- Does every page make the next step visible without scrolling?
- Do buttons say what they do, such as “Send your question”, and not just “Submit”?
- Can a visitor get back to the start in one click from any page?
- Are urgent-situation notices placed prominently and kept identical on every page?
Try the airport test: ask someone to find a specific page in under thirty seconds while you watch silently. Note every hesitation.
Section three: queue status, borrowed from banks
Banks and post offices reduce frustration by showing queue information: your number, how many are ahead of you, an approximate wait. Waiting feels shorter when a person can see what is going on behind the counter.
- When someone submits a request, can they see that it was received?
- Do you state typical response times honestly, including when they are longer?
- If you are running behind, is there a notice rather than silence?
- Is there a reference number or equivalent so a follow-up can be matched to the original?
Section four: the receipt, borrowed from retail
A good receipt records what happened. For an information service, the equivalent is a summary of what was submitted and what was provided in reply. It protects both sides and reduces confusion later. Check whether visitors receive a copy of their own submission, whether replies quote the original question, and whether the language of the summary matches what the visitor wrote.
Section five: the shared rulebook, borrowed from household finance
Households that split costs fairly often write down a few clear rules and keep a simple tracker, so nobody relies on memory. The same discipline applies to a small team handling visitor messages: who answers which kind of query, how long before an escalation, and where decisions are logged. A practical example from a different subject, this note on splitting shared bills with a tracker and plain rules, shows the logic in everyday terms.
- Is it written down who handles which type of message?
- Is there a trigger for passing something to a colleague?
- Is there a place where decisions are recorded so the next person can see them?
Section six: learning from other disciplines
The checklist above is itself a borrowed form. Engineers use audits like it to review systems, and learning sites in that field often discuss them. One place that points toward software, AI/ML and DevOps study is the Gyan School landing page, which presents a menu of Blog, Forum and LMS and lays out four short tracks. It is unrelated to health information, and nothing here relies on it for facts, but it reflects a general habit of learning methods from neighbouring fields. For readers who want a gentle introduction to how that kind of self-directed learning is structured, this roadmap for people learning DevOps remotely gives a readable example.
Section seven: the tracking page, borrowed from couriers
Delivery companies learned that most calls to their help lines were people asking where something was. Their answer was a tracking page showing each step. A health information service has no parcel, but it often has a request moving through a few stages: received, read, answered, closed.
- Can a visitor tell which stage their request has reached?
- Are the stage names plain, with no internal shorthand?
- Does each change of stage trigger a short, polite update?
- If the request cannot be handled, is the visitor told where to go instead?
Even a simple email at each stage is enough. The aim is to turn a long silence into a short series of small, clear messages.
A sample audit in miniature
To make this concrete, imagine a small service run by three people. On the first pass, the reviewer finds that the pre-arrival message exists but never says what the service does not do, so it is marked partial. Menu labels use an internal term for the contact page, which fails the signage test. Submissions get an automatic acknowledgement, which passes the queue-status check, but no reference number, so that is partial. There is no written rule about who answers what.
The fix list that comes out of it is short: add one sentence to the pre-arrival message, rename one menu item, add a reference number to the acknowledgement and write a half-page ownership note. All four fit in an afternoon. A month later the same audit shows three passes and one partial, and the team has a clear sense of progress without having bought anything.
Scoring and prioritising
Count your fails and partials per section. A section with three or more fails deserves attention first. Within a section, prioritise by reach: a confusing menu affects everyone, while an unclear receipt affects only those who have submitted something. Choose no more than five fixes for the first round. Finishing a short list beats abandoning a long one.
Fixing the common failures
Three problems keep recurring. The first is silence after submission, solved with an immediate acknowledgement. The second is insider vocabulary, solved by rewriting headings in everyday words and asking a newcomer to confirm. The third is invisible ownership, where nobody knows who is responsible for answering, solved by a one-page rulebook. None of these needs new software. All of them need a person with an hour and a willingness to look at the service as a visitor would.
A reminder about boundaries
This checklist is about administration, and it must not drift into clinical territory. Pages that describe symptoms, diagnoses or treatments belong to qualified professionals and their own review processes. The audit simply asks whether a visitor can find the right place, understand what the service does and does not do, and know what happens after they get in touch. If your service offers any urgent-help guidance, keep that wording exactly as your own professionals supplied it, and do not change it during an administrative tidy-up.
Repeat it on a schedule
Repeat the audit once each fix round is done, and in any case twice a year. Services change, staff change and wording drifts. A short, regular audit catches slow decline before visitors do. Keep the previous scoresheet so you can see progress, which is pleasantly motivating, and share the results with the whole team.