Education is the only industry in the world where the fundamental delivery mechanism has remained essentially unchanged for five hundred years. A teacher. A student. A transfer of knowledge from one to the other, mediated by language, through a scheduled encounter in a defined location.
The internet was supposed to change this. E-learning was supposed to make education borderless, self-paced, and infinitely scalable. Massive Open Online Courses promised to democratise access to world-class instruction. Hundreds of billions of dollars have been invested in digital education platforms over the past two decades.
The results, measured against the original promise, have been mixed. Completion rates on online courses are stubbornly low — typically between 5 and 15 percent. Engagement metrics consistently underperform expectations. Learning outcomes, when rigorously measured, often fail to demonstrate meaningful advantage over traditional instruction.
The problem is not that digital education was a bad idea. The problem is that it replicated the wrong thing. It took the lecture — the broadcast of knowledge from one expert to many passive recipients — and put it on a screen. The format changed. The pedagogy did not.
AI does not replicate the lecture. AI makes the personalised tutor possible at scale.
The Size of the Opportunity
The global edtech market was valued at approximately $340 billion in 2023. HolonIQ projects it will reach $600 billion by 2030. This growth is driven not by a single innovation but by the convergence of several: near-universal smartphone penetration, the maturation of cloud infrastructure, the growth of skills-based employment pathways, and — most significantly — the emergence of AI capabilities that can finally deliver what digital learning always promised but never achieved: genuine personalisation.
The COVID-19 pandemic accelerated institutional acceptance of digital learning by approximately five to ten years in most markets. Governments, corporates, and educational institutions that were resistant to digital delivery were forced to adopt it. Many discovered that the learner experience and outcomes, under the right conditions, compared favourably with traditional formats. The scepticism collapsed in a single year.
Why the Old Edtech Model Failed
The first generation of edtech — the LMS platforms, the MOOC providers, the video course marketplaces — made the same fundamental error. They optimised for content volume rather than learning outcome. The measure of success was the number of courses available, the number of enrollments, the hours of content produced.
But the number of courses available is not a learning outcome. Enrolling in a course is not a learning outcome. Hours of content consumed is not a learning outcome.
A learning outcome is measurable improvement in the learner's knowledge, capability, or performance in a defined area. Very few digital learning platforms have historically been able to demonstrate this with rigour, because very few have been designed with this as the primary optimisation target.
The best teacher you will ever have is one that knows exactly where your gaps are. AI makes that possible at scale.
This is the precise capability that AI brings to education that no previous technology could provide. A human tutor, working one-to-one with a student, can observe comprehension in real time, identify the specific concepts where understanding breaks down, adapt the explanation approach, test retention at the right moment, and return to weak areas with targeted reinforcement. This is why private tutoring, where it is available and affordable, dramatically outperforms classroom instruction and self-directed online learning.
AI can do this. Not perfectly, and not yet with the full nuance of the best human teachers. But at a level of personalisation that is orders of magnitude beyond what broadcast digital learning can achieve, at a cost per learner that makes it viable at scale.
The Xpandlabs Platform
Xpandlabs was built to exploit this convergence: the newly accepted infrastructure of digital learning, the unmet need for genuine personalisation and measurable outcomes, and the AI capability to deliver it.
The platform is designed around three core capabilities. Adaptive learning pathways — curricula that adjust in real time based on learner performance data, moving faster through areas of strength and slowing down, retesting, and reteaching in areas of weakness. This is the fundamental departure from the fixed-pace, fixed-sequence course format that characterises most digital learning. Competency mapping — a structured framework for assessing where a learner sits on a defined competency pathway and surfacing that assessment in a form that is useful to the learner, to their employer, and to educational institutions. And outcome tracking — rigorous, longitudinal measurement of learning outcomes that allows Xpandlabs to demonstrate the platform's impact in terms that go beyond engagement metrics.
The initial category focus is professional skills and career development — areas where the demand for upskilling is acute, the willingness to pay is established, and the outcome measurement (employability, promotion, salary progression) is clearly definable.
The MENA Expansion
Sweden is Xpandlabs' home market and operational base. But the expansion thesis is centred on MENA, and specifically on the structural education challenge facing the region.
The Arab world has a youth population of approximately 200 million people under 30. Unemployment among this cohort, even in the relatively prosperous Gulf states, runs at rates significantly higher than in comparable developed markets. The skill gap between what young people in MENA are being educated for and what employers in the region actually need is well-documented and widely acknowledged by governments, development institutions, and the private sector.
The demand for effective upskilling and reskilling solutions in MENA is not aspirational. It is urgent. Governments from Saudi Arabia to Egypt to the UAE are spending billions annually on education reform and skills development programs, largely because the economic and social cost of mass youth underemployment is understood at the highest levels of policy.
An AI-powered, outcome-focused learning platform with genuine personalisation capability — positioned correctly, localised for the Arabic language context, and partnered with regional governments and employers — has an addressable market in MENA that dwarfs the European opportunity.
The Investment Thesis
Edtech investment has been through a significant correction since the heights of 2021. This correction has been healthy. It has eliminated the companies that were riding category momentum without genuine product differentiation. What remains — the companies with real technology, measurable outcomes, and genuine product-market fit — are raising at reasonable valuations with serious institutional interest.
Xpandlabs is positioned at the intersection of two durable investment themes: the global upskilling imperative, which is both a policy priority and a commercial need across virtually every market in the world, and the AI productivity revolution, which is transforming every category it touches.