Good EdTech rarely fails on quality — it fails on the journey from purchase to classroom. The DfE's June 2026 EdTech market assessment, formally titled Assessment of the education technology market in England, makes this point sharper than any market-size headline by mapping how schools actually adopt technology. EdTech — the use of technology to support and enhance education — is a £6.5bn sector of 1,123 UK companies according to that report (p.7), yet much of what they build never reaches pupils because it stalls somewhere along a predictable, eight-stage path. For headteachers, deputy heads, school business managers, and multi-academy trust (MAT — a group of schools under one governance structure) procurement leads, understanding that path is more useful than any feature comparison.
This article walks through the eight stages, the four system-level constraints that derail them, and what changes when a tool can be trialled for free — with a practical checklist you can apply this term.
What are the eight stages of EdTech adoption in schools?
The DfE's most practical contribution is an eight-stage implementation journey (Fig. 5, p.43; full breakdown in Appendix 7, p.103). It reframes adoption as a sequence rather than a single buying moment — and a tool can die at any step, regardless of how good it is.
| Stage | What happens | Where it commonly stalls |
|---|---|---|
| 1. Identify need | A problem or gap is recognised (e.g. marking workload). | Need is felt but never formally articulated. |
| 2. Scope and evaluate | Options are researched and compared. | Reliance on informal, word-of-mouth evidence. |
| 3. Secure budget | Funding and approval are obtained. | Competing priorities; no budget line. |
| 4. Procure and contract | Purchasing, terms, and data agreements. | Procurement friction and approval delays. |
| 5. Deploy and integrate | Tool is set up and connected to systems. | Infrastructure and IT integration constraints. |
| 6. Train staff | Teachers learn to use it. | Varying digital and AI literacy; no time. |
| 7. Embed in practice | It becomes part of routine teaching. | Initial enthusiasm fades; reverts to old habits. |
| 8. Measure impact | Outcomes are evaluated; renewal decided. | No measurement, so impact is never proven. |
Where do most EdTech procurement decisions go wrong?
The DfE groups the failure points into four system-level constraints (p.10): strategic capability gaps; resource and prioritisation pressures; procurement and infrastructure constraints; and sustained adoption and impact measurement. These map neatly onto the eight stages — capability gaps wreck scoping (stage 2), resource pressures stall budget and training (stages 3 and 6), procurement friction blocks contracting and integration (stages 4 and 5), and weak measurement undermines the final embed-and-sustain decision (stages 7 and 8).
Crucially, these are structural, not attitudinal. Leaders generally want good technology; the system around them makes adoption hard. That is why a tool which removes friction at one of these constraints — rather than simply being better — often wins. The same workload pressure driving demand is covered in our guide to teacher marking workload in the UK, and the strategic context for leaders is unpacked in our school leader's guide to the DfE 2026 report.
What does the DfE mean by "strategic capability gaps"?
Strategic capability gaps describe a shortage of the time, expertise, and planning headroom needed to make sound technology decisions. As the quotation above sets out, schools face limited capacity for long-term planning, varying levels of digital and AI (artificial intelligence) literacy, and a reliance on informal sources of evidence — meaning a colleague's recommendation often carries more weight than independent evaluation. The report also notes that "reliable evidence on how EdTech is used in practice in schools and colleges remains limited" (p.8), which compounds the problem: even diligent leaders struggle to find trustworthy data.
The practical consequence is that tools win or lose on visibility and ease of evaluation, not just merit. If a leader cannot quickly see a tool working on their own pupils' work, the capability gap means it is unlikely to clear stage 2.
How does free-to-try change the procurement maths?
When a tool is free to trial, several stages of the gauntlet collapse. There is no budget to secure (stage 3), no purchase order to chase, no contract to negotiate before testing (stage 4), and often no IT installation to schedule (stage 5). A single teacher can generate first-hand evidence within existing approval limits, turning the "informal evidence" weakness into verifiable, local proof.
The Assistive Technology lending library precedent. The DfE's own Assistive Technology lending library pilot (p.12) lets schools borrow equipment before buying it — an explicit try-before-you-buy model. It signals the policy direction: lowering the cost of finding out whether something works is increasingly seen as the right way to de-risk adoption.
What changes when there is no purchase order to chase. Adoption shifts from a top-down procurement project to a bottom-up evidence exercise. As a worked example, ReMarkAble AI — a UK-built marking platform that assesses student work against AQA, Edexcel, OCR and WJEC mark schemes from KS1 (Key Stage 1) through A-Level — is free to try with no card required, so a teacher can mark a real class set and judge the output before any budget conversation begins. That is precisely the friction-removal the DfE's analysis rewards.
A practical eight-step checklist for trialling AI tools in your school
- Name the need precisely. Tie the trial to one stated problem (e.g. feedback turnaround), not a vague interest in "AI".
- Demand classroom-level evidence. Favour tools you can test on your own pupils' work rather than relying on testimonials.
- Start within existing approval limits. Choose a free, no-card trial so no budget line is needed to begin.
- Check integration cost honestly. Ask whether it needs IT installation or works in a browser today.
- Run a small, time-boxed pilot. One department, two to four weeks, a defined cohort.
- Plan for staff literacy. Budget time for training; assume varying digital and AI confidence.
- Define how you will measure impact. Decide the success metric before you start, not after.
- Only then escalate to procurement. Move to budget and contracting once the evidence justifies it.
Trial AI marking before you commit a penny
The fastest way past the procurement gauntlet is to skip it for the trial. ReMarkAble AI marks student work against real UK exam board criteria from KS1 to A-Level and returns examiner-style feedback — grade, strengths, areas for improvement, and next steps. Free to try, no card required, no IT setup. Mark a class set this week and let the evidence make the case.
Frequently Asked Questions
What are the eight stages of EdTech adoption?
The DfE's June 2026 EdTech market assessment sets out an eight-stage implementation journey that runs from identifying a need, through scoping and evaluating options, securing budget and approval, procuring and contracting, deploying and integrating with existing systems, training staff, embedding the tool into everyday practice, and finally measuring its impact and deciding whether to sustain it. The point of the model is that adoption is not a single purchasing decision but a sequence, and a tool can stall at any stage. A product can be excellent on paper yet never reach pupils because it fell at training, integration, or impact measurement. School leaders who map a proposed tool against all eight stages before committing budget tend to make better decisions than those who focus only on features and price.
What are the biggest barriers to EdTech in UK schools?
The DfE groups the obstacles into four system-level constraints: strategic capability gaps, resource and prioritisation pressures, procurement and infrastructure constraints, and difficulties with sustained adoption and impact measurement. In plain terms, schools often lack the time and specialist knowledge to plan technology strategically, are juggling EdTech against many competing priorities, face friction in buying and integrating tools, and struggle to prove whether a tool actually improved outcomes. These are structural rather than attitudinal — most leaders want to adopt good technology but the system around them makes it hard. Recognising which of the four constraints is biting hardest in your setting is the first step to overcoming it.
How can a school trial an AI tool without committing budget?
The most reliable route is to choose tools that offer a genuine free trial with no purchase order, no card, and no IT installation, so a teacher can test them within existing approval limits. This sidesteps the procurement and budget stages of the adoption journey entirely for the trial period. A single teacher or department can run a small pilot, gather evidence on real student work, and only escalate to a formal purchase if the results justify it. The DfE's own Assistive Technology lending library pilot reflects this try-before-you-buy direction of travel, lowering the cost of finding out whether a tool works before any money changes hands.
What is the EdTech Testbed Programme?
The EdTech Testbed approach refers to government-backed evaluation initiatives that let schools trial education technology under structured conditions and feed evidence back to the sector, rather than every school evaluating in isolation. The aim is to reduce the reliance on informal, word-of-mouth evidence that the DfE identifies as a weakness in current procurement. The DfE's June 2026 report explicitly notes that reliable evidence on how EdTech is used in practice in schools and colleges remains limited, which is exactly the gap structured trials are designed to close. For an individual school, the practical takeaway is to favour tools that provide transparent, classroom-level evidence you can verify yourself.
What does the DfE recommend for evaluating new technology?
The DfE's June 2026 assessment does not issue a single checklist, but its analysis points clearly towards evaluating technology against the full adoption journey rather than the purchase decision alone — checking that a tool can be trialled cheaply, integrated without heavy IT work, embedded into routine practice, and measured for impact. It cautions against reliance on informal sources of evidence and highlights limited capacity for long-term planning as a core constraint. The report also flags that investment is shifting towards specialised AI solutions, which suggests leaders should weigh purpose-built tools against general-purpose ones. In short: trial it, integrate it, embed it, measure it — and demand evidence at each step.