There is a line in CQC’s own guidance that most practices have never read, and it changes what preparing for an assessment should actually look like.

Writing about how it gathers evidence, CQC says: “We do not want providers to prepare specific documents rather we ask for information they already have.”

And just before it: “There isn’t a full list of evidence that fits every service.”

Read those two together and something becomes clear. There is no pack to assemble. There is no definitive list to work through. When CQC arrives, it will ask for what you already hold – and what you already hold is what gets read.

Which changes the question entirely

Most practices preparing for an assessment ask themselves whether they have a policy for something. It is the wrong question, and it is the easy one.

The harder question is this. If an inspector read that policy cold, on their own, with nobody from the practice in the room to explain what was meant – what would they conclude?

In our experience the problem is very rarely a missing document. It is a document that does not say what the practice thinks it says.

Does your safeguarding policy name your practice? Does your training matrix show that a course was relevant to the role, or only that it was completed? Do your minutes record a decision, or a discussion? Does your significant event log show what changed afterwards, or only what happened?

None of that is about the quality of care. It is about the record of it. And it is remarkably hard to see in your own documents, because you already know what they were meant to say.

Why we built something

PCMS has over twenty years of experience in general practice, and our team includes former CQC inspectors and specialist advisors. Mock inspections, remedial work after a poor rating, practices in special measures, practices already Good and trying to work out what Outstanding would take.

Doing that work, the same thing kept coming up. Collecting the evidence was never the slow part. Reading it was. Working out what a document actually established, whether it matched what staff described, whether it would hold up in front of someone who had never met the practice – that is where the days went, and that is where the judgment lived.

So we built something to do the reading.

CQC AI Auditor started as a tool for our own consultants. Then the obvious thought followed: if this helps us inspect a practice properly, it can help a practice inspect itself.

What it actually does

A mock inspection has a method. We built that method into the product.

• It reads the evidence you already hold, rather than checking that it exists.
• It asks your practice the questions an inspector would ask, and follows up on the answers.
• It checks what you said against what your evidence shows.
• It assesses each quality statement on the evidence itself, not on whether a document was supplied.
• It rates and calibrates the findings, so a strong area and a weak one are not treated the same.
• It sets those findings against the patterns that appear in published CQC inspection reports.
• It gives you a view of where you would most likely stand, the reasoning behind it, and what to do next.

The third step is the one that matters most. Plenty of tools will read your documents. Others will ask you questions and score your answers. Checking one against the other is where the real findings come from, and it is what a good inspector does instinctively.

What it is not

Worth being clear, because this is a crowded market and a lot of things get called the same thing.

• It is not a document store. If you need somewhere to keep policies, cheaper things exist.
• It is not a policy generator. It reads what you have rather than writing it for you which matters, given CQC asks for what you already hold.
• It does not replace a consultant, or your own judgment. It gets you to the judgment faster.
• It does not guarantee a rating. Nothing honest could. It gives you an indicative position and shows its reasoning.
• It is not affiliated with the Care Quality Commission.

Why now

Because the gap between assessments is closing.

Since March 2026, CQC has been running focused assessments under its Returning to Good and Outstanding programme, aimed at NHS GP practices rated Good or Outstanding whose last inspection report was published between 2017 and 2022. Those come with at least five working daysnotice.

CQC has also published its assessment priorities for primary care, and among them are services whose rating is more than seven years old.

If your last report was published in 2019, you can work out where you stand from two things you already know: your rating, and the date on that report. And five working days is not long enough to fix anything that has been drifting since then.

Where it is up to

CQC AI Auditor is currently in development.

If you would like to know when it is available, you can join the waitlist at cqcaudit.ai.

And if you would rather a person did it, nothing has changed there – we still run mock inspections, and always will.