Imagine a CQC inspector asks how a missed medication prompt was fixed. You have training records and QA records, but can you show the link between them without opening a spreadsheet? That question sits at the heart of good governance, especially when safe care is under the spotlight. Storing both sets of records is a good start. It is not the same as joining them. In this post, we will walk through a practical three-step traceability test you can run on any care platform. First, trace a QA finding backwards to the training and assessment of the person involved. Next, trace it forwards to the corrective action and the evidence it worked. Finally, check whether anyone had to export data to make that join. We will also look at what a pass and a fail look like, walk through a missed medication prompt and a falls audit failure, and share red flags to ask about on a demo. By the end, you will know how to test traceability before your next CQC inspection.
Nearly every platform you demo will hold training records and quality assurance records, and most are telling the truth. Holding both is table stakes, not a differentiator. The useful distinction is between a system of record, which stores data, and a system of traceability, which connects it. Buying guides rarely test for it; a feature list cannot show the difference.
In our experience working with provider teams, training, audits, incidents and care records often sit in separate modules owned by different teams, so care providers face real obstacles connecting them. The gap is the join, not the storage.
Inspection day exposes it. An inspector asks who was trained and assessed on a task, what the assessment showed, and what changed afterwards. If that takes, say, three days of admin, the platform stored your data without connecting it.
This is our practical argument for judging any shortlisted system, ours included, not a CQC requirement.
In our experience of providers preparing for CQC visits, inspectors want to see governance working in practice, not just described on paper.
Start with Regulation 17 (Good governance), last updated 16 May 2025. Check the current wording before you rely on it. A missed medication prompt or falls audit failure would plausibly sit under Regulation 12 (Safe care and treatment); treat those as illustrations of risk, not law.
The compliance gap we hear about most often is not missing records, it's not being able to surface them quickly when an inspector asks. From what providers tell us, if the link between training, assessed competence and the action taken lives in someone's memory or inbox, you are relying on that memory at the worst moment. An integrated platform built for CQC compliance keeps it in one place.
With the CQC framing in mind, get your evidence ready before you log in.
Pick one real finding from the last three months, ideally one with a person attached: a missed medication prompt, a falls audit failure or a missed observation chart entry. Have the finding reference, date, service location and the role or person involved to hand.
Hold two questions in your head: who was trained and assessed on this task, and what changed afterwards?
Then run it live on a demo and ask to drive the screen yourself. A pre-built report handed over by a salesperson shows you their best presentation, not the test. You want to watch the system answer your questions, using your finding.
With your finding and two questions ready, run the three-part test. The test has three parts, tracing backwards, tracing forwards, and checking for exits, each covered in detail below.
Start with the finding itself, not the training record. Search for a missed medication prompt or falls audit failure and see what the system offers to link from it. If you cannot see who was trained on that task, when, and what their assessment outcome was without already knowing their name, it is not tracing, it is searching.
Check visibility by role and team, because shift patterns, agency staff and new starters all muddy who was actually on duty. Then test the reverse: open a learner's profile and see whether their QA findings and actions appear there. A genuine join works both ways. Finally, watch for the same-system, different-module trap. Two tabs behind one login still means storing, not joining, unless a record in one visibly points to the other. Good Real-Time Reporting should make that path visible.
Once the finding links backwards to training and assessment, follow it forwards. Can you create a corrective action from that same record, without leaving it? Check for a named owner, due date, live status, and the option to reopen if the fix does not hold. Can you attach proof it worked, a follow-up observation, re-assessment, supervision note or repeat audit, directly to the finding? You should also see who signed it off, and when. Does the action connect to learning, such as a refresher, a learning pathway or a competency re-check on the failed task? Finally, test whether the loop really closes. If findings sit open for months because closure depends on someone remembering, it has not closed the loop.
Once the action is closed, the real question is how you got there. Ask directly: show me this without exporting anything. Then stay quiet and watch the screen.
Red flags:
A downloadable CSV.
A spreadsheet opened alongside.
Data copied into a report template.
An email to the training lead to check something.
Count the steps and minutes, then multiply by the findings you handle each quarter. That is your real admin cost, and it never appears in a product brochure.
Other tells: reports that only ever hold one dataset, dashboards that refresh monthly, and a training view and a quality view never seen together.
With the clock stopped, here is how to read the result.
A pass: one finding, opened once. Backwards it shows the person, their training history and their assessment outcome. Forwards it shows the action, the evidence and the sign-off. All dated and attributable.
A fail: two reports side by side, or a caveat about what the system cannot quite do yet.
The in-between case is common. The finding links to a training module but holds no assessment outcome. You know someone completed a course, not that anyone checked they could do the task. Completion is a weak answer to a competence question. Competence is the ability to apply knowledge and skills to achieve intended results, so assessment has to sit in the same chain as the finding.
The question that separates the two: "If I asked for this on a Tuesday morning, with an inspector beside me, how long would it take?"
Here's how that plays out on a real finding.
Picture a medication audit that flags a missed prompt on one shift. Treat it as a walkthrough, not a breach.
Tracing backwards, you want the training and competency picture for everyone on that shift.
Tracing forwards, you want the corrective action and its evidence of closure.
In a joined platform, that whole story lives in one record: finding, people, training and assessment history, action, evidence, closure date.
Manually, the same question spans three systems and two spreadsheets, needs a call to the training lead, and depends on who is in the office that day.
Same test, different part of the service. This time a falls audit flags that a post-falls observation or a moving and handling task wasn't completed to standard.
Backwards, you want to see who was assessed as competent on that task, when, and when that assessment is due to expire. That includes anyone who has since changed role or moved service, which is where manual records quietly fall apart.
Forwards, you want the retraining, the re-assessment and the repeat audit. Those show practice actually changed, rather than an action being marked complete.
From our work with providers, falls and medication are two areas where training and day-to-day practice most obviously have to line up.
Bring both to the demo and ask plainly. Can you show me a real finding and its linked training without exporting anything? If you get a pre-built report instead of a live path, you have your answer.
Next, where do assessment outcomes sit relative to training completion? If they live in a separate module with no link, someone is doing the join by hand somewhere.
Can a corrective action be linked to one person's learning record, and does that record show the finding it came from? If not, you're looking at an actions tracker, not a quality assurance loop.
Ask who maintains the link when someone leaves, changes role or is redeployed.
Write down what you were shown, keep it separate from what you were told, then ask for that behaviour to be demonstrated again at contract stage or in a pilot.
If the demo passes all five questions, ask to run it again at contract stage using your own records.
Those questions matter because of what happens when the answer depends on something you built by hand. That is our argument, based on what we see in provider teams, not a CQC mandate. Any join you assemble yourself, in a spreadsheet or across two exports, is a join you own, including every error in it.
Manual joins drift. One side gets updated, the other does not. Dates reflect when the spreadsheet was built, not when the event happened. The version an inspector sees may not be the version your team works from.
At inspection, "we can pull that together" is not "here it is". That gap is the difference between describing your governance and demonstrating it. It also eats the time you need for the part that matters: acting on the finding.
So ask any platform one final question: who owns the join?
myAkoTQ closes that gap.
myAko is a UK learning and development platform built for health and social care, used by around 1,500 providers and 100,000 learners across domiciliary, residential and supported living. It covers CPD-accredited eLearning, competency assessments, supervisions and appraisals, learning pathways, face to face events and CQC compliance reporting.
myAkoTQ joins training and quality assurance, so a finding traces back to the training and competence record and forward to the action and its evidence, no spreadsheet in the middle. Built with care providers rather than adapted from a generic HR tool, the workflow matches shift life.
Want to run the three-step test on your own data? Every client gets free UK-based chat and phone support, rated 4.8 out of 5. Ask us to walk through a real finding on your own data, our UK support team is on chat or phone.
Run the test on your own data. Pick one finding from the last three months and run the three-step test on your current system.
If any step needed an export or a second report, bring it to your next platform conversation and ask for a live demonstration on your own data.
Finally, keep CQC in proportion: read the current Regulation 17 and Regulation 12 guidance on the CQC site for the wording that applies to you, then use this test to judge good governance support, not as a rule to obey.
Our Learning Management System is used by hundreds of Health & Social care providers - book a demo now!
One of the team is ready to show you how everything works and answer any questions you might have!
Choose a Time Slot
Join Zoom Call
45 Minute Demo
Ask Questions
