Published: 12 August 2026
Most adult social care providers deliver their mandatory training. Very few can produce, within an hour, a defensible answer to a narrower question: which of your staff are not currently compliant, how long have they been in that position, and what did you do about it?
That second question is the one Regulation 18 actually turns on, and it is where a lot of otherwise well-run services come unstuck.
Delivering training is an operational activity. Monitoring training, and acting on what the monitoring shows, is a regulatory duty. Providers tend to evidence the first well and the second badly.
The duty is monitoring, not just delivery
Regulation 18 requires providers to have sufficient numbers of suitably qualified, competent, skilled and experienced staff, and to ensure those staff receive the support, training, professional development, supervision and appraisal necessary for their role. CQC's guidance goes further than the delivery of training: it expects providers to monitor whether required training is completed and to take action where requirements have not been met.
Read that as two separate evidence obligations:
- The training happened. A record per person, with a date.
- You were watching, and you acted. A record showing you knew about the gap and what you did about it.
A folder of certificates evidences the first. Almost nothing in a typical provider's systems evidences the second, because the monitoring happens in someone's head, in a meeting, or in a spreadsheet that gets overwritten every month with no history.
Why the training matrix stops working at scale
The spreadsheet matrix is close to universal in social care, and for a single service with stable staffing it can be perfectly adequate. It fails predictably as the operation grows, for reasons that have nothing to do with how carefully it is maintained.
It has no memory
A spreadsheet shows today's position. Overwrite it next month and the previous position is gone. When an inspector asks how long someone has been out of date, or what you did when you first noticed, the honest answer is that the system does not know.
Turnover resets it constantly
Recruitment and turnover in adult social care mean a service's compliance position changes materially month to month. Every new starter restarts induction, Care Certificate progress and competency sign-off, while renewals keep falling due for everyone already in post. A manual matrix is always describing a workforce that has already changed.
Renewals run on individual clocks
Training renewals are anniversary-based per person, not calendar-based per service. Modelling that in a grid means either recalculating by hand or defaulting to an annual sweep that leaves some people out of date for months.
It cannot answer the follow-up question
Inspectors rarely stop at the headline number. They ask about a named individual, or a specific requirement, or a particular service. A group-level percentage cannot be drilled into, so the conversation moves to files, and the files are in a cabinet at a different site.
Completion is not competence
This is the single most common weakness in social care training evidence, and it is worth being precise about.
A care worker completing an e-learning module on moving and handling has demonstrated that they engaged with content. A senior carer observing that same person perform a transfer safely, and recording it, has demonstrated something else entirely. Regulation 18 is framed around competence. The evidence for competence is an observation, with an assessor, a date and an outcome.
In most services that observation genuinely happens. It is done properly, by someone qualified to do it, and then written on a form that goes into a paper file in the manager's office. It is not reportable, it is not visible above service level, and it cannot be produced quickly. The provider has done the work and cannot show it.
| Evidence type | What it proves | What it does not prove |
|---|---|---|
| Module completion | The person engaged with the content on a date | That they can perform the task |
| Knowledge check score | They understood the content | That they apply it under pressure |
| Observed competency | A named assessor watched them do it safely | That it remains true a year later |
| Re-observation | Competence has been confirmed since | Nothing further — this is the strongest of the four |
The mandatory list is not fixed
Providers often talk about "the mandatory training" as though it were a settled list. In practice it is assembled from several sources that change independently:
- Skills for Care's statutory and mandatory training guidance for the sector
- Local authority and ICB contract requirements, which differ between commissioners
- Role-specific requirements — medication, catheter care, PEG feeding, epilepsy and rescue medication
- The Oliver McGowan training on learning disability and autism, at the tier appropriate to the role
- Service-specific requirements arising from the needs of the people you support
A provider operating across several local authorities is therefore running several mandatory lists at once. Holding that in one spreadsheet with a column per course, applied uniformly to everyone, produces two failure modes simultaneously: people are assigned training they do not need, and people miss training they do.
Pick one care worker at one service. Can you show, in five minutes, what their role requires, what they hold, what is outstanding, when each item expires, and who observed their last competency check? If not, the inspection version of that question will not go better.
What good evidence actually looks like
Strong providers tend to have five things in common, and none of them are about buying software specifically.
1. Requirements are defined per role and per service
Not one list for everyone. A care worker in a supported living service under one contract carries a different requirement set from a nurse in a residential home under another. Getting this right removes most of the noise from the reporting that follows.
2. Some requirements block, and everyone knows which
Items that must be complete before someone works unsupervised should be explicitly flagged as blocking. Rota planning should be able to see who is not yet cleared without asking. Making this visible is usually the single fastest safety improvement available.
3. Competency observations are captured as data
The observation still happens the same way. The difference is that it is recorded against the worker with the assessor, date and outcome, rather than filed on paper. That makes it reportable and it makes re-observation schedulable.
4. Monitoring leaves a trail
The evidence that you acted is as important as the evidence of the gap. A record showing that a shortfall was identified on a date, escalated to a named person, and closed on another date is genuinely strong evidence of the Regulation 18 monitoring duty. It is also the piece almost nobody keeps.
5. Policy changes reach the right people
When a policy is revised, the people whose practice it affects need to be reassigned training and the acknowledgement needs to be recorded against that version of the document. Otherwise you have a current policy and a workforce trained on the previous one, with nothing to distinguish them.
Provider checklist
- Define requirement sets per role and per service rather than one universal list.
- Mark the items that must be complete before unsupervised work, and surface them on the staff list.
- Record competency observations as data with assessor, date and outcome.
- Keep the history — the position last month matters as much as the position today.
- Log the action taken on a gap, not just the gap.
- Calculate renewals from each person's own dates, not an annual sweep.
- Reassign training when a policy changes, and record acknowledgement against the version.
- Be able to filter to a single service, because that is how you will be inspected.
None of this makes an inspection pleasant. It does change what the first hour looks like: opening a record rather than opening a filing cabinet, and answering the follow-up question rather than deferring it.
Frequently asked questions
What does Regulation 18 require in relation to training?
Regulation 18 of the Health and Social Care Act 2008 (Regulated Activities) Regulations 2014 requires providers to deploy sufficient numbers of suitably qualified, competent, skilled and experienced staff, and to ensure staff receive appropriate support, training, professional development, supervision and appraisal. CQC's guidance is clear that providers should monitor whether that training is completed and take action where it is not.
Is a completion percentage enough evidence?
Rarely, on its own. A percentage tells an inspector how many people finished a module. It does not show which individuals are outstanding, how long they have been outstanding, whether anyone acted, or whether the person can perform the task competently. The action taken on the gap is usually the more revealing evidence.
How should competency be evidenced separately from training?
With an observation record: what was observed, who observed it, the date, the outcome and any re-observation. Completing an e-learning module on moving and handling and being observed performing a transfer safely are different pieces of evidence, and inspectors treat them differently.
Does the Oliver McGowan training apply to all our staff?
The Health and Care Act 2022 introduced a requirement for CQC-registered health and adult social care providers to ensure staff receive training on learning disability and autism appropriate to their role. Which tier applies depends on the role, so it should be modelled as a role-level requirement rather than assigned uniformly. Check the current guidance for the position that applies to your services.
Sources & further reading
- Care Quality Commission — Regulation 18: Staffing
- Skills for Care — Statutory and mandatory training guidance for adult social care
- NHS England — The Oliver McGowan Mandatory Training on Learning Disability and Autism