Unannounced CQC inspection: the first thirty minutes

A lot of providers worry that the inspector will arrive cold and start asking for paperwork they should have remembered to print. That is not how it goes.

Before we walked in, we already had what CQC holds on its own systems: the statement of purpose, the registered activities for the location, who the registered manager is, the previous report and its findings, any conditions on registration, the history of statutory notifications submitted by the provider, and the safeguarding referrals the local authority has logged from the service over the trailing period. We were not turning up to ask the team for any of that. Asking would have told them I hadn't done the preparation, which is not the impression an inspector wants to give in the first ten minutes.

What we did not have, and could only learn by being there, was the live shape of the day. How many patients in the service today. What the activity level looks like: busy, quiet, mid-morning lull, end of clinic. How the team is currently configured: who is on shift, who is in charge, who the consultant is, who is the most senior clinician physically present. What's happening right now on the floor. That is what the first thirty minutes were for.

I would show my ID at reception, explain who I was, and ask to speak with whoever was running the service that morning. Sometimes that was the registered manager, sometimes a deputy, sometimes the senior clinician on shift if the manager was off-site. I was not in a hurry. If the person I needed was with a patient, I waited. I did not need them pulled out of clinical work to greet me.

When we did sit down, the first conversation was deliberately not about documents. It was about the day. How many patients are booked in. Is anyone particularly unwell or at risk today. Is there anything happening on the unit right now that I should be aware of: an incident in progress, a complaint being handled, a member of staff off sick that has stretched the roster. Have there been any changes since the last inspection I should know about: new lead clinician, new service line, change of premises layout. Is the registered manager the same person CQC has on record. Are there any restrictions or limitations on what I can see today, and if so, why.

That conversation served three purposes at once. It established that the day's operational reality mattered to me more than the documents. It let me read how the person in front of me described their own service when they had not had time to prepare. And it let them set expectations: if there was a particularly unwell patient in a room, we would not be walking into that room without warning. If there was a complaint meeting at 2pm, we would work around it. Inspections do not improve services by disrupting them.

I would also use the opening to be transparent about my own day. Roughly how long I expected to be there. What I was going to want to see, generally, not the document-by-document list. Whether I would be on my own or whether a second inspector was joining later. Whether I would want to interview specific people and when. That predictability matters. A team that knows the shape of the day relaxes; a team that thinks the day might lurch in any direction stays defensive, and defensive teams answer questions worse than relaxed ones.

We were not auditing yet. We were forming a first impression of the operational rhythm of the service. The questions on my mind:

How does the senior person in the room describe their own service when they have not had time to think about it. Does the description match the reality I can see around me on the way in: the reception area, the corridor, the noticeboards, the staff body language, the patient flow. Is the team behaving normally around the manager or are they suddenly on best behaviour because an inspector is in the building. Does the manager know what's happening on the floor today in specific terms: names of patients of concern, status of the consultant rota, current bottlenecks; or are they describing the service in the abstract.

None of that was being recorded as a finding. It was setting the working hypothesis I would then test for the rest of the day against the records, the cases, the observations, the interviews with staff. The first thirty minutes were the question-setting phase. The rest of the day was the answering phase.

The documents came after that. Once we'd had the opening conversation and I'd had a chance to walk through the relevant clinical areas, I would ask to see specific things: the incident log for the trailing twelve months, the safeguarding log for the same window, the current dashboard or quality report the service uses to run itself, the training matrix covering the staff on shift, and the action plan from the last inspection if there had been one. By that point I had a sense of the service to test the records against, rather than starting with the records and trying to read the service from them.

Two things I was reading inside those documents that providers often miss. First, training currency for the staff in the building today, not training compliance in the abstract. A 92 per cent training compliance rate matters less than whether the two clinicians on shift this morning are signed off on what they are actually doing. Second, evidence of learning from things that had gone wrong. An incident log full of entries with no linked action plan, or action plans with no audit or training event closing the loop, tells the same story as a thin log: the service is recording, not learning.

By the time we sat down with the registered manager for the first formal conversation, usually within an hour of arrival, we had a working hypothesis. Three patterns shaped how the rest of the day went.

Pattern one: documented service, working culture. The records were timely, the logs were complete, the actions tracked. The team behaviour we observed in the corridor matched what the records said. The first interview confirmed the working hypothesis: this was a service we needed to test on specifics rather than test on systems. The rest of the day went into case-by-case sampling. The report tended toward Good or Outstanding.

Pattern two: working service, undocumented culture. The team behaviour was visibly competent. The registered manager was respected by the staff we met. The incident log was thin or the safeguarding log was patchy or the dashboard did not quite line up with what we were seeing. This was the most common pattern in well-run independent services that had not invested in the records side. The rest of the day went into asking the registered manager to evidence what we could already see was happening. The report tended toward Requires Improvement on Well-led, sometimes Good on the four other key questions, which felt unfair to the team and was fair to the regulation.

Pattern three: documented service, brittle culture. The records were impressive at first glance. The dashboard was sophisticated. The incident log was complete. The first conversation with the registered manager showed the records were a layer independent of the actual work; she could narrate the metrics but could not connect them to the practice the team had run that morning. The rest of the day went into testing whether the records corresponded to reality. The report tended toward findings that the registered manager had not seen coming.

The first thirty minutes did not determine the rating. They determined the questions. A team that walked the inspector through the records calmly, knew where each element lived, and had a working answer to the cross-reference question was a team that had already answered most of the Well-led test before the formal conversation started.

If a CQC inspector walks in unannounced on Tuesday, the test is whether the first thirty minutes give them pattern one, two, or three. Three things change which pattern they see.

Make the time-to-produce under fifteen minutes for each opening-request item.The current dashboard, the incident log for twelve months, the safeguarding log for the same window, the training matrix with currency dates against each requirement, and the action plans behind the last six months of incidents showing what changed in practice. CQC already holds the statement of purpose, the registered activities, the previous report, and the history of notifications, so the opening request does not ask for any of that. The items above are the ones the inspector cannot see from outside the building. Each should be retrievable in two clicks. If the dashboard lives in a slide deck on a consultant's shared drive, move it. If the incident log lives in three different spreadsheets, consolidate it. If the training matrix is a printed sheet on the office wall, the inspector will be told it is current and have no way to check. The platform you choose should make the request collapse to a single login.

Test the cross-reference yourself. Pull the safeguarding log and compare to what you would expect the local authority to have on file. Pull the incident log and compare to the actions register. Pull the actions register and compare to the meeting minutes. Pull the training matrix and compare against the staff on shift today: who is signed off on what they are actually doing, and whose currency lapsed last month without anyone noticing. Significant gaps inside your own evidence are the gaps the inspector finds in the first hour.

Test the records-to-reality match by walking two records.Pick two from the incident log at random. Walk the chain of evidence behind each one: the initial record, the investigation, the action, the change to practice, and the audit or training event that closes the loop on the change. If any link in the chain requires a phone call, a different system, or somebody's memory, the chain is incomplete in a way an inspector will find. The question they ask is not “did you investigate?” but “what changed in practice as a result, and how do you know it stuck?” Fix the chain on those two records and extrapolate.

Know how you compare, and to what. In some form, the inspector asks how you know the service is safe and improving. Closure rate on incidents over the trailing six months. Safeguarding referrals per hundred episodes against what is reasonable for the service type. Training currency percentage across the team. Audit completion against the assurance calendar. A team that can name the trend, the benchmark, and the direction of travel in the first hour is a team that runs the service against the records. A team that can only describe the records as a filing exercise is pattern two from the section above.

The first thirty minutes do not catch you by surprise if the records side of the service runs at the same tempo as the clinical side. Verivius exists to make that the default state rather than the project state. The platform records the incidents, the safeguarding referrals, the actions, the audits, the training matrix, and the assurance calendar in one place, and surfaces the dashboard from the records themselves. Closure rate and training currency sit next to the work that produces them, not in a separate quality report a consultant runs once a quarter. The opening-request set lives one login away. The cross-reference test runs on its own. The records-to-reality match, and the “what changed in practice as a result” question that follows it, is built into the workflow rather than reconstructed for an inspection.

The most useful preparation is not what to say to the inspector. It is what you should be able to ask them. A registered manager who opens the first conversation with “what would help you most at this stage?” and listens to the answer puts the rest of the day on a footing that is collaborative rather than defensive. We were not adversaries. We were two people trying to establish whether a service was running safely. Acting like that was true changed how the day went, in both directions.

I have seen registered managers tense up at unannounced inspections and I have seen registered managers welcome them. The second group ran better services and got better reports. Not because they were better managers. Because they had nothing to hide and the records side of the service was at the same tempo as the clinical side. That is what to prepare for.

On the same Well-led pattern from a different angle: Why providers fail Well-led. On the longer journey of moving from Requires Improvement back towards Good: From Requires Improvement to Good. If you are preparing for a real inspection within the next six months, the Verivius Mock Inspection is the consultant-led calibration of where you actually stand against the inspector's questions.