# Account audit — Healthcare

The account audit with the checks a provider needs: whether every live claim is covered by a current clinical review, whether targeting has drifted into what platforms do not permit, whether spend still matches capacity, and what proportion of enquiries became attended appointments.

Everything else is the general audit, unchanged. One row per finding, evidence attached, honest severity, one action, one owner, one date.

## How to use it

1. Copy this file, or the CSV version if you would rather work in a sheet.
2. Work area by area rather than campaign by campaign, so the same check is applied everywhere.
3. Write what was found separately from what should be done.
4. Attach evidence to every finding, with the date range you looked at.
5. Keep severity honest: if everything is critical, nothing is.
6. Agree owners and dates with the people who will do the work, not in their absence.
7. Close findings as accepted where somebody has decided to live with them.

## The columns

**Area** (Text). Which part of the account this belongs to: claims, targeting, capacity alignment, attendance, booking, pages, access. Claims and targeting are first because they are the two that can be wrong without anybody editing anything.

**Check performed** (Text). What was looked at, as an action somebody could repeat: every live ad matched to the clinical review record, audience settings and copy read together, spend compared with availability. Repeatable checks make quarters comparable.

**What was found** (Text). The observation with no recommendation in it: these ads name a practitioner who has left, this service received spend while its book was full. Stated plainly it can be agreed by the practice manager before anybody discusses the fix.

**Evidence** (Text). Where it can be seen, with the day it was read: the list of ads with review references, the availability screen, the attendance export from the practice system. Availability in particular is a moving number and needs its date.

**Severity** (Text). Spending wrongly now, distorting decisions, or untidy. Advertising a full service is the first; judging a service on enquiries rather than attendance is the second; an inconsistent campaign name is the third.

**Impact** (Text). What it is costing, in appointments rather than in clicks. Where the enquiry-to-attendance rate for the service is not known, say so: that gap is itself a finding and a better one than an estimate built on a guess.

**Recommendation** (Text). One action a person can start on Monday: remove the ads naming a departed clinician, move this week's weight to a service with availability, start recording response times. Not improve lead quality.

**Effort** (Text). Minutes, a day, or a project. Pulling an ad is minutes; getting new wording through clinical review depends on somebody else's week, so write the expected turnaround in the notes rather than a date nobody controls.

**Owner** (Text). The person who will do it, by name, and for anything touching claims that is the clinical reviewer rather than the marketer. Several rows here will be owned outside marketing, which is normal in a practice.

**Due** (Date (YYYY-MM-DD)). When the fix is expected. Anything about a claim or a departed practitioner is dated for this week; a measurement improvement can sit against the next quarter without anybody pretending otherwise.

**Status** (Text). Open, done, or accepted. Accepted is legitimate for structural findings — a service whose attendance the system cannot record yet — and it should close the row rather than leaving it to be reported again.

**Reviewed on** (Date (YYYY-MM-DD)). The day this finding was checked. Capacity, waiting times and rosters all change weekly, so an availability finding is stale faster than any other kind and should be re-read before it is acted on.

**Notes** (Text). What the next auditor needs: which service each campaign serves, the practitioner whose ads were retired, the review that narrowed a claim, the seasonal pattern that explains a quiet month.

## The account audit

| Area | Check performed | What was found | Evidence | Severity | Impact | Recommendation | Effort | Owner | Due | Status | Reviewed on | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|   |   |   |   |   |   |   |   |   |   |   |   |   |
|   |   |   |   |   |   |   |   |   |   |   |   |   |
|   |   |   |   |   |   |   |   |   |   |   |   |   |
|   |   |   |   |   |   |   |   |   |   |   |   |   |

## Before you commit it

- Every finding separates what was found from what should be done.
- Every live ad's claim has been checked against a current clinical review.
- Targeting and copy have been checked against what the platform permits.
- Spend by service has been compared with capacity by service.
- Enquiry-to-attendance has been measured per service.
- Every row has an owner by name and a due date.
- Findings somebody has decided to live with are closed as accepted.