If you've spent any time in Further Education or Skills, you'll likely have sat through a graded observation. You know the drill: someone sits in, takes notes, and at the end, gives a number. 1, 2, 3, maybe a 4 if things really didn't go well. It’s a model that’s lingered in our sector for years, but let me say it plainly: I think it’s time we moved on.

I’ve never found a single-digit grade to be useful. And I’ve never seen one truly improve teaching.

Let me explain.

It’s just one opinion

Grading a lesson, based on one snapshot in time - is an opinion. That’s all. A potentially well-informed one, yes, but still an opinion. And if we’re being honest, the outcome often says more about the observer than the observee.

Was the observer confident in the subject area? Did they come in with bias from prior knowledge or experience? Were they in a bad mood? Had they just observed an exceptional session before yours?

All of these things matter. Because grading is subjective, and that subjectivity can undermine trust. I've seen too many brilliant tutors demoralised by a '2' that they were told was 'still really good' while being compared to someone else’s '1'. I've seen '3s' handed out in the name of standardisation, and then watched as that number eclipsed everything else the tutor did that year.

It reduces people to a moment - when the job is built on consistency over time.

It doesn’t reflect what Ofsted actually do

Ofsted moved away from grading individual lessons years ago. Yet many providers still cling to the old model, attempting to mirror inspection practices that no longer exist. Inspectors observe to inform judgements about quality overall, not about a single session or a single tutor.

They look for trends. They triangulate. They talk to learners. They follow the thread.

That’s what our internal quality assurance should be doing too. Not dishing out grades that reduce complex practice into something overly simplistic and, frankly, unnecessary.

Expectations, not evaluations

Instead of grading, I believe in clear expectations.

Every organisation should have a shared understanding of what good teaching looks like - in context, in subject, and in line with learner needs. Observations then become about whether practice meets that expectation, or whether it doesn’t yet.

That approach keeps things developmental. It allows for nuance. It prompts better conversations.

And most importantly, it reminds everyone that we’re here to support improvement, not hand out scores.

Grading masks the real issues

Here’s the problem with numbers: they end the conversation.

You give someone a 2, and they breathe a sigh of relief. No need to go any deeper. You give someone a 3, and they focus on defending themselves instead of reflecting on their practice. Either way, the grade becomes the focus - not the learning, not the learners, not the impact.

When you remove grades, you remove the defence mechanism. You make space for curiosity.

What worked? What didn’t? What could we try next?

Those questions matter far more than what number we land on.

The data doesn’t help you

Let’s be real. What do most providers do with their observation grades?

Tally them up. Put them in a dashboard. Track trends over time.

But what does that data really tell you?

It doesn’t tell you about the journey of a tutor who’s made huge progress over the year. It doesn’t capture innovation, risk-taking or responsiveness. And it certainly doesn’t help your QIP or SAR stand up to scrutiny.

What you need is narrative, not numbers. Evidence, not estimates.

An effective observation record should tell the story of teaching in your provision - its strengths, its sticking points, and its strategies for improvement.

What works better?

In my experience, what works is:

• Co-observation and dialogue

• Clear, shared standards of what good looks like

• Constructive, evidence-based feedback

• Regular follow-up and support

Not everyone will get it right every time. But if every session is treated as a chance to grow, not to score points, then your quality culture improves. And so does your teaching.

Final thoughts

Teaching is not a performance. It’s a practice.

Observations should support that practice - not reduce it to a number on a scale.

If we want providers to be inspection-ready by default, then our internal QA must reflect what inspection actually looks like. And if we want our teams to thrive, then we need to ditch the grading and get back to professional dialogue.

The best observations I’ve ever done didn’t end with a grade. They ended with:

"That made me think."

And isn’t that the point?