Client case study · 5 months Illustrative

4.2 min→55 sec

Documentation time per consultation, over 5 months.

Elapsed: 5 months
How it was measured: Timed across 300 consultations before and after, same doctor, same clinic hours.
Running alongside: No change to patient volume or clinic staffing.
Published with permission: Anonymised at client request

The constraint

Forty patients a day, and the notes written after closing

Documentation was happening after hours, from memory, which is both slower and less accurate than writing at the time.

The alternative — typing during the consultation — costs eye contact, which is the part patients actually notice.

Roughly 2.8 hours of documentation a day, after clinic
Notes written from recall, hours after the consultation
Prescriptions re-written by hand into a register

What we found

The audit, including the unflattering parts

We timed 300 consultations and the documentation that followed.

4.2 minutes of documentation per consultation
2.8 hours a day after closing
Inconsistent structure made records hard to search later

The plan

The 90-day roadmap as it was actually written

Draft the structure, keep the doctor in control of every line.

Consultation produces a structured draft, not free prose
Doctor reviews and approves before anything saves
Prescription flows into the register automatically once approved
Nothing saves without explicit approval

Month by month

What shipped, and what it moved

The line extends as you scroll. Every step below is a real month, including the one where nothing happened.

01Month 1

Baseline

4.2 minutes per consultation and 2.8 hours of after-hours documentation.

Baseline: 252 sec

02Month 2

First drafts

Drafts generated, and accuracy problems surfaced immediately.

-17%

03Month 3

Structured fields only

After the failure below, free narrative was removed entirely.

-44%

04Month 4

Prescription register

Approved prescriptions flow into the register without re-writing.

-39%

05Month 5

Approval flow settled

Review takes seconds because the structure is predictable.

-35%

What went wrong

The part most case studies leave out

Accuracy on drug names and dosages was not good enough in month two. We restricted the draft to structured fields with an explicit approval step rather than free narrative, and the doctor reviews every line before it saves.

Free-text drafting was not accurate enough on drug names and dosages, and in a clinical setting that is not a rough edge, it is a hazard.

We removed narrative generation entirely and restricted the system to structured fields with constrained vocabularies. The doctor approves every line before anything saves, and nothing auto-commits.

The lesson is that in clinical work the right automation is narrower than the impressive one.

The numbers

Where it landed, and how we know

Documentation time per consultation: 4.2 min → 55 sec over 5 months. Timed across 300 consultations before and after, same doctor, same clinic hours.

Timed across 300 consultations before and after, same doctor and same clinic hours.

Documentation per consultation: 4.2 minutes to 55 seconds
After-hours documentation: 2.8 hours a day to under 30 minutes
Records now structured and searchable, which they were not before

Results reflect this client's market, licence status and starting position. We publish the method alongside the number so you can judge whether it transfers to yours.

In hindsight

What we'd do differently next time

Start with structured fields; narrative generation was the wrong ambition
Never let anything commit without explicit approval
Measure accuracy before measuring speed

Before you ask us to do this for you

Does this actually apply to your situation?

This worked because of specific conditions. If yours don't match, say so on the first call and we'll tell you what would change.

Market: High-volume outpatient clinics where documentation happens after hours.
Starting point: A doctor willing to review every draft. This does not run unattended.
Timeline: Four to six months.

Start with the audit. It's free and it's specific.

Send us a URL. You'll get back the technical issues, your AI-search readiness score, what competitors are doing that you aren't, and what we'd fix in the first ninety days.

Get a free audit → WhatsApp us instead