The loud cheers that followed Kenneth Baker’s promise in the late Eighties to give teachers five non-contact days in every school year almost drowned out the second part of his announcement - that the school year was to become five days longer. In the end, these five days are hardly noticeable when added to the other 190, and as schools have considerable leeway in how they use this time, in-service training days often feel like a welcome break.

But a recent report by the National Foundation for Education Research found that Inset days are often spent “giving out information or on administrative matters”. It asked whether we should be making better use of this “substantial investment of scarce resources”.

John Harland, head of the NFER’s northern office, says that although schools have improved their use of the days over the past 10 years, they could do better. His team observed Inset at primary, secondary and special schools, and talked to teachers and managers.

The biggest problem is differentiation. Teachers want to update their own specialist knowledge, or learn new teaching skills and styles that suit their differences. Senior staff lack the resources to identify the right training for each member, then find trainers to provide it. Thus, all the staff end up on the same course.

The Government is concerned about the state of Inset. Ministers know its delivery is often desultory. It’s time for a rethink. Teachers are contracted to work 195 days per year, of which 190 are teaching days. So each teacher could receive five or more days of individualised professional development a year.

This is the general idea - a teacher makes the case for receiving a particular piece of professional development and identifies where to get it. For each day the teacher is out of school, one class has a day of “Inset leave”. Common sense can be applied to choose a class that the Inset teacher would have taught, or one from that teacher’s year or subject area. Teachers released as a result cover the classes of the teacher on Inset. If several teachers want to attend the same course, more than one class would be sent home.

The training would be ongoing, on-demand and on-the-job. Professional development would be on the basis of immediate need.

Teachers would take part in a much wider range of courses, as well as structured inter-school co-training. This could involve exchange, coaching and work-shadowing. Mentoring by and for teachers could become the norm. Teachers would share knowledge and skills, with parents and paraprofessionals acting as co-trainers, with separate, parallel roles. “In-line training for pupil support” would routinely define the teacher’s special needs with respect to the pupil’s special needs. This model would allow a teacher to take several days together for more sustained training.

Although co-training costs would be low, the increased cost of formal courses would need to be met by small DfES grants and other creative budgeting (course costs would fall as demand increased). The Government might want teachers to contribute some of their own time in exchange for accreditation. A balance would need to be struck between individual professional development and the continued need for some school-wide training.

The last words should go to John Harland: “In the few places where I’ve seen a more individualised approach, it’s clear that teachers like the sense of being in control, and they appreciate that it addresses their own needs better.”

Adam Abdelnoor is a senior research fellow at London University