CPD has moved toward AI-powered personalised pathways in 2026. We examine how platforms are using classroom data to suggest specific micro-improvements for teachers.
For decades, professional development in education has been remarkably resistant to the personalisation that we expect in almost every other domain of modern life. We have algorithm-driven music recommendations, personalised health tracking and bespoke consumer experiences — but teachers have largely received CPD determined by school timetables, budget availability and the preferences of whoever booked the training day, rather than their specific developmental needs.
In 2026, that is changing. AI-powered professional development platforms are beginning to deliver on the promise of genuinely personalised learning pathways for educators — using classroom observation data, student outcome data and self-reported reflections to suggest specific, targeted micro-improvements rather than generic training packages.
The most sophisticated AI professional development platforms operate by integrating multiple data sources to build a picture of a teacher's current strengths, development priorities and learning preferences. These sources typically include lesson observation data — often captured through video with teacher consent — student assessment and progress data, teacher self-reflection inputs and comparative data from teachers at similar career stages and in similar contexts.
From this data, the platform generates specific, actionable recommendations. Not "improve your questioning technique" — which is generic to the point of uselessness — but "your wait time after closed questions averages 1.2 seconds: research suggests extending this to 3-5 seconds typically increases the quality of student responses by 40%. Here is a five-minute micro-learning module on implementing longer wait times."
This is CPD that is relevant at 9.00am and implementable by 10.00am. It is a fundamentally different proposition from the generic seminar model.
The evidence for AI-personalised CPD is promising but still developing. The most rigorous evaluation to date — a randomised controlled trial conducted by the Education Endowment Foundation in partnership with The CPD Register — found significant positive effects on targeted teaching behaviours but more modest effects on student outcomes, suggesting that behaviour change in teachers takes time to translate into improved learning (EEF, 2025).
This finding is consistent with what we know about professional learning more broadly. Teaching practice changes gradually, not overnight. The value of AI personalisation is not that it produces instant transformation but that it ensures the incremental improvements teachers make are targeted at the areas where they will have the greatest impact on student outcomes.
It is important to be clear about what AI professional development can and cannot do. It can identify patterns in practice data that human observers might miss. It can deliver content at a time and pace that suits the individual learner. It can track progress and adapt recommendations based on what is and is not working.
What it cannot do — at least not yet — is replace the human relationship at the heart of excellent coaching and mentoring. The most effective professional development has always involved a trusted colleague who knows you as a professional and a person, who can read the emotional subtext of a conversation, who can challenge you in ways that feel caring rather than threatening. AI augments this relationship. It does not replace it.
At Academica Mentoring we are enthusiastic about the potential of technology to personalise professional development. We are also clear that the human expertise, care and relationships we provide remain irreplaceable — and that the best outcomes come from integrating both.