The things that matter most to children, feeling seen, feeling heard, feeling that their voice produces something real, do not show up in the data. That is not evidence they do not matter. It is evidence the data is incomplete.

School leaders implementing mental health programmes face an uncomfortable question: how do you know if it is working?

The honest answer is that the measurement tools available are not good enough to tell you with confidence. Wellbeing surveys capture a child's reported state on a given day, shaped by what happened at breakfast and who they sat next to at lunch. Standardised screening tools were designed for clinical populations and translate imperfectly to school settings. The data looks rigorous. The picture it represents is genuinely uncertain.

This creates a real dilemma for schools operating under accountability frameworks that demand evidence. You are being asked to demonstrate impact on something that resists easy measurement using tools that were not designed for your context. The result, too often, is that schools choose programmes with tidy outcome metrics over programmes that actually change something in children's lives.

There is a better question than "how do we measure this programme's impact?" It is: what would we expect to see in children if this was working, and are we seeing it?

The answer does not live in a survey. It lives in what teachers observe. A child who begins to describe themselves differently. A child who stays after class to keep writing. A child who, for the first time, puts their hand up to share something they created. A child whose story reveals something their teacher did not know was there, and who now has language for it. These are not anecdotes. They are the signal that measurement tools are too blunt to detect.

When 318 children were asked "what was it like to be the author of your own story?", the methodology used was Classic Grounded Theory, designed specifically to let patterns emerge from lived experience rather than from predetermined categories. Seven consistent themes appeared across nine schools in two geographic regions. That consistency, across diverse contexts and demographics, is itself a form of evidence that standardised surveys cannot produce, because it captures what children actually experienced rather than what researchers expected to find.

The question for school leaders is not whether you can produce a clean impact metric for your mental health provision. It is whether you are creating the daily conditions in which children feel seen, develop a positive sense of who they are, and experience themselves as capable of overcoming difficulty. Those conditions are observable. They show up in behaviour, in language, in the stories children tell about themselves. You do not need a better questionnaire. You need adults who are paying attention to the right things.