Personalised adaptive learning technologies have emerged at the forefront of the education technology field. These software systems provide learning activities for pupils — and what sets them apart is their capacity to use machine learning to target subsequent learning activities, and the rigour of those activities, towards the learning level of the individual pupil. Organisations like X-PRIZE winner onebillion have coupled this software with corresponding hardware to create an adaptive, integrated learning platform for pupils.
The limits of traditional instruction
Let's contrast this with a traditional learning sequence. The night before a lesson, a teacher plans the objectives, procedures and assessments to be used during the class. During a lesson, the teacher facilitates the learning activities and assesses comprehension. But ultimately, these predetermined activities, even those designed to reach as many learners as possible, fail to respond to the actual performance of an individual pupil in real time.
Imagine one pupil who easily completes the first six problems in a problem set on perimeter, while another pupil struggles to answer the first question of the same set. Regardless of their ability levels, both pupils continue on the same problem set — Pupil 1 learning very little because the problems are too easy, and Pupil 2 learning very little because the problems are too hard.
The challenge in developing contexts
It's not hard to see why school systems in developed contexts are moving towards blended learning, in which pupils spend part of their day learning using personalised adaptive programmes that align content rigour to pupil ability levels. But this approach is not universally accessible.
Pupils in developing contexts are often learning in environments with limited internet connectivity, a lack of power sources, and poor security infrastructure to protect valuable technology. Universal adoption of personalised adaptive technologies is not an immediate reality, especially in developing countries. Yet this presents a false choice between tech-enabled personalised learning on one hand, and traditional instruction on the other.
Reading Club: structured adaptive learning
Reading Club is designed to bridge this gap. Rather than relying on hardware and connectivity, it uses structured lesson design and teacher-mediated grouping to approximate the benefits of adaptive learning within a standard classroom. Pupils are grouped by reading level, and lesson activities are matched to each group's current proficiency.
The teacher moves between groups during a lesson, delivering targeted instruction to each. This gives every pupil access to content calibrated to their level — the core promise of adaptive technology — without requiring a device for every child.
What teachers and pupils experience
For teachers, Reading Club provides clear lesson plans for each reading group, removing the cognitive burden of simultaneously planning across multiple ability levels. Teachers report a richer understanding of each pupil's progress than traditional whole-class observation allows.
For pupils, the experience is one of being neither stuck nor bored. They are working at the edge of their competence — the zone where learning is most efficient. Early results show significant improvements in oral reading fluency and comprehension across diverse classroom contexts.
The combination of structured pedagogy and thoughtful grouping represents one of the most scalable frontiers in foundational literacy education — particularly in under-resourced settings where technology-led personalisation remains out of reach.