i keep writing the same sentence

[image 1 — see image brief at the end]

i went back through what i have written this summer and found the same argument four times, wearing four different coats.

[website build: these four lines are the scrolling cards. see the note in the image brief.]

delivered is not adopted.

planned is not the same as received.

lived experience of the market, not of the barrier.

and this week: published is not reached.

i did not plan that. i wrote each one because something specific annoyed me that week. it took going back through them to see it was one sentence, not four posts.

here is the sentence.

we measure the sending. we do not measure the landing.

the four coats

a device rollout reports ninety-two percent delivered. four thousand units shipped. those are logistics numbers. they tell you a lorry arrived. they do not tell you whether a lesson changed, or whether anything changed for a child in a way that was still there in the summer term.

an inclusion strategy describes what is planned for a child. from september, ofsted's case sampling tests something different. inspectors follow individual pupils who face barriers and check what actually reached them. (ofsted, what you need to know about the september 2026 updates to education inspections, 12 june 2026.)

the dfe convened a lived experience board for its assessment of the edtech market and gave the seats to market expertise. my perspective: nobody was sitting on that board because a piece of technology had failed them.

and last week the dfe published a back-to-school guide for parents. eight strategies for the hardest week of the year, written by a psychologist with twenty years of family work behind her. it is good. it is also filed inside guidance addressed to school leaders, with no editable version, no translation and no easy-read. (dfe, the back-to-school reset guide, 19 august 2026.)

the same collection already carries ready-to-use letter templates, text message templates and phone scripts. so the department knows precisely how to build something a school can adapt and send. it did not do it here.

which leaves one kind of reader. a parent who went looking for dfe attendance guidance on gov.uk in august. that parent is mostly already doing what it says.

[image 2 — see image brief at the end]

i do not think anyone is being careless

i want to be fair here. i have sat in the room while this decision gets made, and nobody in it has stopped caring. the money has a date on it. the person asking for the update has a date on it too.

delivery is the only thing you can count before the money runs out. units shipped. pages published. seats filled on a board. all of it is finished, defensible, and available on the day someone asks how a programme is going.

arrival has none of that. it happens quietly, in a room you were not in, weeks later, to only one person at a time. it does not fit a reporting cycle and it will not close a budget line.

so the official number that gets reported is the one that exists. not because anyone stopped caring, but because that is what is actually countable at the time, with the instruments they are holding.

give someone a better instrument and they will use it. nobody has.

the part that is harder to say

there is a second reason, and, for me, it is less comfortable.

arrival can only be confirmed by the person it was supposed to arrive at.

for most of this work, that person is a child. and asking a child is the one step that can come back and tell you the money did not work.

an adult account of the week tells you what was planned, what people believed happened, and what they believe did not. all of it from an adult perspective.

only the child can tell you what actually arrived, or didn't.

that is not a cynical reading of anyone's motives. it is just that the question nobody asks tends to be the one with the answer nobody wants.

the evidence caught up this month

i have argued the belonging half of this for a while, and i have always attached a caveat to it. the strongest study i could find was small, american, and about undergraduates. (walton and cohen, science, 2011.)

the dfe published english evidence last week. (dfe, school belonging and attainment at key stage 4, 19 august 2026.)

the question was whether pupils agreed with a single statement. i feel like i belong at school.

that is it. one line, lifted from the pisa survey.

pupils who agreed scored roughly a full grade higher across eight subjects than pupils who disagreed. pupils with send reported lower belonging than their peers, and more bullying than the oecd average.

one question, asked of the child, worth about a grade.

that is not a soft metric. and the caveat can come off.

the measurement gap is documented too. the dfe's own survey found 22% of leaders have a formal plan to evaluate whether their technology is working. counting informal methods, 35% — down from 41% two years earlier. (dfe, technology in schools survey report 2024 to 2025.)

we are spending at scale and measuring less than we were.

[image 3 — see image brief at the end]

what changes in september

case sampling is not the first time inspectors have talked to pupils. that has always happened.

what is new is that it becomes the method rather than a by-product of it. inspectors pick pupils who face barriers, follow what actually happened for them, and talk to those pupils about it. inclusion has its own grade, and this is how it gets evidenced.

so the question moves. not whether the provision is good, but whether it turned up.

plenty of schools will find that uncomfortable. i think it is the right question, and it is the one this whole argument has been circling since july.

what i would actually do

none of this needs permission or a budget.

pick one child at random. not the one you already know the answer for. randomise it, or you will choose the reassuring case.
follow them for a week. what was planned for them, and what reached them.
ask them four things. what helps you in lessons. what do you do when you get stuck. is there something meant to help you that you do not use, and why not. and what do you wish i understood.
write down what you find, including what did not happen. the record of noticing is the evidence. a school that can show a year of looking at arrival is in a completely different position to a school that can show a policy.
keep it blameless. the moment it becomes performance management, staff stop telling you the truth and you lose the only data worth having.

the third one is where it pays. children routinely do not use the thing. because it makes them visible. because it was set up once in september and nobody came back to it. because nobody explained what it was for.

none of that shows up in a plan. all of it shows up in a sample.

the sentence, one more time

delivered is not adopted.

planned is not the same as received.

published is not reached.

four posts, one argument, and i did not notice until i lined them up. every one of them counts the sending and not the landing.

the only person who can confirm a thing arrived is the person it was sent to.

in this work, that is usually a child. and we have built an entire sector that measures everything except them.

sources

earlier posts in this argument: delivered is not adopted · lived experience of the market, not of the barrier · planned is not the same as received.