It makes a case from our own record at DreamSpace Academy, sets out the argument against it, and marks how well each claim is backed by evidence. For what we actually run, the main pages are the description to rely on. This essay gives the reasoning behind them.
Education is how the ecosystem gets built
One of three essays on how our education work and the rest of DreamSpace Academy fit together.
In short. DreamSpace Academy is an Innovation & Entrepreneurship Ecosystem Builder: we help a place grow the people, links and support that new ideas and businesses need. An organisation like ours has to run education itself rather than buy it in. Section 5 sets out the five ways we do this.
Evidence marks. Accounts of what happened come from our own reports. What we have decided comes from our written plans. Each claim carries a mark: 🟢 backed by original documents or by more than one source; 🟡 supported, but only in part; 🔴 stated, and still being tested.
1. Why an ecosystem builder runs education
People expect an ecosystem builder to bring people together, make introductions, run an incubator, find money for new businesses and put on a demo day. Running a robotics class for eleven-year-olds in a rural government school is not on that list. It looks like another organisation’s work.
We see it the other way round. In an underserved community, education is the only part of the job that produces what everything else needs: people who will build. Every other part of the job assumes there are already founders, makers, trainers and partners to work with. Where there are not, someone has to bring them through, and teaching is the only way to do that.
Two facts pull in opposite directions here. First, the shortage of such people is what holds the ecosystem back. Second, education on its own does not produce income. This is well evidenced, and our own written analysis says so. We accept both.
2. What ecosystem theory says, and its limits
Researchers who study entrepreneurial ecosystems are open about the idea’s weak points. Its leading writers call the concept “under-developed and under-theorized”. The most common complaint is that the idea goes round in a circle: an ecosystem is whatever produces successful businesses, and wherever there are successful businesses there must be a good ecosystem [Stam 2015; Stam & van de Ven 2021]. A review of 181 studies found that many use the term “in name only” [Wurth et al. 2023].
The review of entrepreneurial ecosystem sets out the full definition and its limits.
Two points from that research hold up despite the criticism, and both explain why an ecosystem builder should teach.
The parts only matter when they work together. Isenberg, who first set out the idea in 2010, says that leadership, culture, capital markets and customers are “in isolation… conducive to entrepreneurship but insufficient to sustain it. That’s where many governmental efforts go wrong: they address only one or two elements.” So a list of the universities, mentors, investors and events in a place does not show that an ecosystem exists.
Entrepreneurs build the system as well as coming out of it [Stam 2015]. This puts education at the centre. If people build the system, then having enough of them is something needed at the start, not a result counted at the end. Where the ecosystem is missing, those people are missing too.
The research also carries a warning that governments often ignore: stop copying Silicon Valley. Reviews find that governments commonly copy a model from a successful ecosystem elsewhere without asking whether it fits local conditions [Spigel 2017; Brown & Mawson 2019]. An accelerator in a district with no founders has nobody to accelerate. The problem there is a shortage of people, and education is how that shortage is filled.
3. The constraint we found: identifying changemakers
Our hardest constraint was never the labs, the funding or the curriculum. It was identifying changemakers.
Reaching people was not the hard part. Anyone could apply, and many did. Field visits, open calls for applications and our own bootcamps brought in large numbers, and most of the changemakers so far arrived that way. One open call in October 2022 ran across four districts and drew 360 applications, of which 30 were selected through district workshops and one-to-one interviews 🟢.
The hard part was that the old approach waited for people who were already motivated to come forward. Such people are rare, and finding them cannot be planned or grown. The supply is unpredictable and tiring to keep up: every search has to be run again the next year, from zero.
So we changed our approach:
Co-creation changes the supply from found to built. The Changemaker Journey (Explorer, Maker, Innovator, Changemaker) develops the motivation rather than screening for people who arrive with it. Finding changemakers stops being a search and becomes something we teach.
A search has no schedule, and nobody knows how many people it will find. Teaching has a curriculum, a timetable, a cohort (a group that starts and finishes together) and a rate that can be measured. Education is what lets an ecosystem builder make that change. Nothing else on the usual ecosystem-builder list does.
4. The main argument against: education does not produce income
The strongest objection to this argument is “education does not produce livelihoods, and the data says so.”
The data does say so. Our own analysis of why young people in these regions struggle to earn rests on these national statistics:
- Unemployment rises with education in Sri Lanka: 0.7% at Grade 5 and below, 3.4% at Grades 6 to 10, 6.0% at O/L, 8.0% at A/L and above, and the gradient holds in every province 🟢.
- 67.6% of Sri Lanka’s unemployed are already computer literate, 75.5% of those aged 20 to 24, 81.2% of those aged 25 to 29 🟢.
Two out of three unemployed Sri Lankans can already use a computer. Computer literacy alone does not guarantee work; market access and other support matter as well.
Studies of training programmes reach the same answer from the other side. Training on its own does little or nothing for income, and what it does often fades:
- only about 29% of short-term earnings estimates are positive;
- a review of 37 evaluations finds “no immediate translation into business set-up or increased income”;
- the most careful randomised trials of skills training show the effects fading by 12 to 15 months 🟢.
Training plus something else is a different matter. A job referral raised earnings by 40% where training alone did nothing. Programmes that combined several kinds of help produced gains that lasted, across six countries. In post-conflict Uganda, cash plus training plus supervision doubled incomes 🟢. Peer-reviewed research on FabLabs (shared workshops with digital making tools) finds that making builds some entrepreneurial skills, and needs dedicated programmes and business support on top 🟢.
We call these extra inputs couplings: the money, access to buyers, introductions and continued support that turn a skill into a livelihood. Our analysis concludes that what breaks the link between skill and income is the lack of these couplings, not a lack of training.
Our position is therefore narrower than “education works”:
Education is the mechanism that produces builders. It is not the mechanism that produces income. An organisation that teaches without the couplings will watch the gains decay on a schedule the literature can already predict.
This is why the teaching sits inside an ecosystem builder rather than inside a school. A school can offer only the first half. Our education depends on what comes with it: the labs, the bridging into partners’ work, the seed money, the mentors and contact that lasts for years.
Several of these exist today:
- an impact-loan fund and co-founding of ventures, both held by DreamSpace (Pvt) Ltd rather than DreamSpace Academy;
- bridging and access to buyers;
- contact that lasts for years.
But they have never been put together as one planned package, or measured as one.
5. How we use education to build the ecosystem
Our teaching builds the ecosystem in five separate ways. Lumped together as “programmes”, they look like a training centre. Taken one at a time, each one builds a different part of the ecosystem.
① Empowerment Inside School: the school programme
Empowerment Inside School (EIS) is our main programme inside schools. It teaches ICT that follows the Sri Lankan government syllabus, plus Robotics & STEAM, at no charge to underserved schools. It is planned to stay with the same pupils for six years, from Grade 6 to Grade 11. Pupils learn by working on challenges rather than by memorising.
The programme follows a written procedure with 13 phases 🟢. The main steps are:
- choosing schools that have no ICT teacher, no computer lab, low O/L ICT results, rural or marginalised;
- getting Zonal Education Department approval;
- testing pupils at the start, so there is a baseline;
- teaching;
- testing again at the end and comparing with the start;
- closing, which trains teachers and leaves a student ICT club behind.
Our programme reports give these figures:
- Batticaloa pilot, July to December 2025, 5 schools: “Reached 194 students across Grades 6–9”;
- Hatton, across Dickoya and Ireby, late 2025: 306 students;
- four Nuwara Eliya schools, April to May 2026: 114 students, 13 sessions, 21 projects 🟢.
② The Changemaker Journey: where it runs, including the MakerSpace
The STEAM programmes are one part of what we teach. We also run business and entrepreneurship training, digital security training, and workplace and teacher training.
The Changemaker Journey is the path learners follow in the STEAM programmes. It can run in schools, at partner sites, online and in our own spaces.
MakerSpace by DreamSpace Academy is something narrower. It is a design for a physical space (the equipment list and layout) plus a standard for running it. It is built to be copied and franchised in a space that a local partner provides, and it is one of the places the journey can run. It is not the organisation itself, and our education does not depend on it.
So far we have run several curriculum tracks and our own bootcamps 🟢:
- seven numbered robotics batches, two of which split into two groups running side by side;
- a music programme in three batches, with graded exams in two languages;
- AI & Electronics with Scratch;
- Make & Play;
- a programme preparing students for university;
- monthly one-day workshops with prepared agendas and slides.
The essay How DreamSpace Teaches covers the teaching method.
We have no standing labs. Software, media and music work is done by groups of people who share a craft and learn from each other, not by standing units. An older published list of eleven labs describes the past, not the present.
③ Handing over: local organisations run school clusters
We do not run every EIS cluster ourselves. We hand clusters to local organisations that employ their own staff:
| Organisation | What it runs | Source |
|---|---|---|
| Kathira Greens | the Vakarai, Kalkudah EIS cluster (3 schools) | our director, August 2026 🟢 |
| Inverta | the Kantale cluster (5 schools), employing its own trainers | our director, August 2026 🟢 |
| Eegai | incubated 2021, now the EIS managing organisation for Trincomalee | our DreamShot programme papers 🟡 |
| Thiru & team | Hatton estate schools | our director, August 2026 🟢 |
Through the school programme, local organisations gain a contract to run, staff to pay and a reason to exist. A changemaker who came through our pathway founded Kathira Greens. Our teaching produced the founder, and then produced the founder’s contract.
④ Bridging: placing local people in partners’ programmes
Bridging is not teaching, but it is something we offer that outside programmes do not. We know the region and its people, which an international programme cannot buy, and bridging turns that local network into work for local people.
The clearest example is in our director’s own words: “IREX wanted to spread its media-literacy programme across Sri Lanka and they didn’t have a good partner, DreamSpace Academy brought them to the Central and the North… I introduced Ladhushan and Donald as trainers for their training programme”, and they trained about 60 people 🟢.
Bridging takes four forms:
- placement: putting individuals into a funder’s programme as paid trainers;
- devolution: handing school clusters to local bodies (③ above);
- convening: a partner’s premises and reputation host one of our events;
- exposure: we use our international standing to put our own people on a stage.
Bridging is not a revenue model. It is a way of getting work done. We do not publish yet how it is paid for (trainer day rates, the terms for EIS clusters, and anything we earn as the go-between).
⑤ Alumni: returning as teachers and directors
Alumni now lead much of our software, media and music work, and two are directors 🟢. When we took our approach to Erode, Tamil Nadu, it was “delivered by two people who are themselves products of the pathway”. A large share of the people who teach for us came through our own programmes.
The limit: these are Batticaloa-trained people travelling to a new place. That is not yet evidence that a new site can grow its own teachers. The one test of that failed. DreamSpace Mullaitivu closed because nobody took over from the person running it: that person could not continue, no successor with the interest was found, and the search looked only locally. Funding, day-to-day running and skills were not the cause. Any claim that the approach can be copied to new places has to count the site that closed as well.
6. Sri Lanka’s school reform and the gap in rural schools
Sri Lanka’s curriculum reform makes ICT and “Entrepreneurship & Financial Literacy” compulsory essential subjects in Grades 6 to 9 for the first time. It phases in from 2026 and covers all four grades by 2029 🟢.
The means to teach them were never built:
- 19% of type 2 and type 3 schools have IT labs, against 72% of top-tier schools 🟢. (Type 2 and 3 are the smaller government schools that do not teach up to A/L.)
- In the technology stream there are 35.3 students per teacher, against an average of 12.8 🟢.
- At home, 1.5% of estate households own a computer, and 92.6% of estate internet access is by smartphone alone.
A phone is not enough to learn CAD, coding or a laser cutter.
The government’s STEM reform shows there is demand for this work. The state has promised something it cannot yet deliver in rural schools. That gives us a place as a partner that supports the state system and feeds into it, not as a rival.
Whether that place can be made formal (an MoU, a provincial agreement, recognition as a teaching partner) is still an open question. The need is proven. Our formal right to act on it is not. What exists today is narrower: written Zonal Education Office approvals in three zones (Batticaloa, Kantale, Nuwara Eliya).
National policy makes a related point in its own words: Sri Lanka’s startup ecosystem is “confined to the urban areas especially Colombo”, and “decentralizing support infrastructure” is named as a national policy priority. Our work carries out that priority.
7. BOOKBRIDGE: an education programme that produced an organisation
DreamSpace Academy began in BOOKBRIDGE’s 11th Capability Program in 2018. The programme had four modules, ran for about six months and ended with a pitch to investors. As BOOKBRIDGE publishes it, each centre is set up as a social enterprise run by a local entrepreneur, paid for by a soft loan (a loan on easy terms) that is repaid from the centre’s own trading income 🟢.
So an education programme produced an organisation: DreamSpace Academy itself.
The details of the loan, including whose it was and whether it was repaid, have not been confirmed. This story is not an argument against grants.
We are funded from three sources: grants and partnerships, revenue we earn ourselves, and the founders’ own capital. We are working out how much each one contributes, and will say so once we know. We are not trying to become independent of grants. We are building work that keeps running after the grant that paid for it stops.
8. What we do not claim
The argument above explains why an ecosystem builder should teach. It does not claim that our ecosystem has yet produced what it aims at.
- We do not yet know how many learners become Changemakers. We are setting up records of how many people start the pathway. Until that count exists, there is nothing to compare the Changemakers against, and nothing can be said about how well the pathway works as a whole, only that each part of it is backed by evidence. 🔴
- “Earn” is a hypothesis being tested, not a result. That a skilled maker can sell things they have made for a small income is near-term and low-risk 🟡. That this becomes a lasting livelihood at a reliable rate, in this place, through this pathway, we have not measured yet 🔴. We have no figure for jobs created, and we publish none.
- What the ecosystem has produced is mostly unmeasured. The description Innovation & Entrepreneurship Ecosystem Builder rests on what we have built, not yet on what the ecosystem has produced.
- Reported outcomes so far come from people who finished. We are working on following up learners who leave, and we will publish what we find.
9. Summary of the argument
Entrepreneurs build an innovation and entrepreneurship ecosystem as well as filling it. In a community where those people are not already there, an ecosystem builder has two options: wait for them to appear, or bring them through.
Waiting is a search with no schedule and no promised result. We did that for years and named it as our hardest problem. Bringing people through is teaching, with a curriculum, a cohort and a rate that can be measured.
The evidence also shows that teaching alone does not produce income. So the teaching has to be joined to labs, bridges into partners’ work, money and contact that lasts for years, not just to a certificate. Those links are the ecosystem, and education brings it the people.
A note on sources
This essay draws on our own reports, case studies, accounts from our director and the history of our funders, together with our written plans: our description of who we are, our analysis of why young people struggle to earn, our theory of change, the DreamSpace Lifecycle and the incubation model.
National statistics come from their publishers: the Department of Census and Statistics Labour Force Survey 2023 and Computer Literacy Bulletin 2024, UNDP and OPHI district profiles 2023, the World Bank Sri Lanka Development Update (October 2024), and the WIPO Global Innovation Index 2025. Intervention evidence is cited to the published studies named in the text; the full list is on the references page.
Some figures often quoted on this subject have been shown to be wrong, among them the “90% franchise versus 30% startup survival” statistic and a branded “Five Stages of Scale”. We use neither.
A third party repeating our own words is not independent evidence. Partner pages, directories, award listings and news items that quote us are repeating us, not reporting something they saw, and we do not treat any of them as confirming anything.