From free to paid
How a free project turns into a paying customer
Free is not a trial. Free is the entire discovery phase, and it costs almost nothing to serve.
- 1Open a free labTwelve labs are live. The first full project in each one is free forever, with its research question, ten written steps, the people to interview, the defence questions and the marking rubric.
- 2Do the workThree or four projects a year alongside school, in different industries. High-Tech uses only free browser tools. No-Tech uses paper and conversations, so a classroom with no electricity runs the identical curriculum.
- 3Submit for assessmentMarked out of 20: artifact 6, process and its failures 6, defence 8. Ten logged relationships are required per project. AI marks and written feedback returns within two weeks.
- 4The conversion momentThe student names the industry that is theirs and asks for a second project in the same lab. Willingness to pay appears here, not at signup, and no amount of free content substitutes for it.
- 5Pay at achievementEAD Plus is a low flat monthly or annual family subscription covering up to three children, and Level 1 certification is charged at the assessment rather than at the start. Rates are set per country.
Year 1 plan
What we are aiming at by July 2027
These are the Year 1 numbers from the plan, alongside one sponsored lab and roughly US$13,000 to US$16,000 per month by month 12.
Measured in by July 2027 (targets)
Targets from the plan, not traction. Nothing on this chart has been achieved yet.
See the numbers as a table
| Line | by July 2027 (targets) | Note |
|---|---|---|
| Students completing a free project | 20,000 | at least one project |
| Paying families | 600 | EAD Plus |
| Professionals certified | 100 | entering at Level 1 |
| Paying campuses | 25 | School Edition |
Revenue lines
Six lines, three payers, real prices
Parent, school and sponsor buy different things, so one paying does not remove the reason for another to pay.
| Line | Who pays | Price |
|---|---|---|
| EAD Plus | Parent | Flat monthly or annual family rate, up to 3 children |
| Certification, Levels 1 to 3 | Parent, school or sponsor | Charged at the assessment, rising by level |
| School Edition | School | Flat monthly rate per campus, up to 300 students |
| Guaranteed Term Pilot | School | Single campus, group and international rates, refunded in full if the EAD class does not outperform a matched comparison class |
| Teacher Certification | School | US$2,320 per cohort of 30, 2 days onsite plus 4 weeks guided practice |
| Sponsored Niche Lab | Sponsor | US$15,000 to US$50,000 per year for one lab |
The moat, ranked honestly
What is copyable and what is not
Ordered from weakest defence to strongest. The business sits on the last three cards, not on the first.
Content and method: copyable
Twelve labs, 240 written steps and the two-route design are public and can be cloned by anyone with a writing team. Treat this as distribution spend, not as an asset.
Assessed evidence: costly to copy
Every certified project carries a marked artifact, a documented failure record, a recorded defence and ten logged relationships. A competitor has to build the marking operation, not just the pages.
A recognised credential: hard to copy
Value comes from institutions accepting the levels. Curriculum is already mapped to SNC and PCTB, Cambridge and KHDA. Formal recognition is a Year 5 target of three institutions; today it does not exist.
The network: not copyable
Students, mentors, industry sponsors and alumni, where posting, peer review and mentor circles open only after one completed project. Membership is earned by work, so the network cannot be bought or cloned.
Unit economics
Where the cost of one paying family goes
About US$24 a year to serve one paying family, against an annual EAD Plus subscription plus certification when a level is passed — a contribution margin that holds in every market we price for.
100% = 100% is the planned US$24 annual cost to serve one paying family
- Human certification — 50%about US$12
- Platform, storage and support share — 33%about US$8
- AI marking and defence questioning — 9%about US$2
- Parent reports — 8%about US$2
Planned costs from the model, not audited accounts. AI marks and humans certify, so the cheap step absorbs the volume and the expensive step carries the value.
See the numbers as a table
| Line | Share | Note |
|---|---|---|
| Human certification | 50% | about US$12 |
| Platform, storage and support share | 33% | about US$8 |
| AI marking and defence questioning | 9% | about US$2 |
| Parent reports | 8% | about US$2 |
Built versus planned
The line between what exists and what is intended
Read the right column as intent only. Everything on the left can be opened, downloaded or self-hosted today.
Shipped and verifiable today
- 12 labs live, with the opening project in each written out in full
- 240 written project steps across both the High-Tech and No-Tech routes
- 41 free video lessons for ages 3 to 8 and 35 booklet missions for Grades 5 to 9
- Community Edition LMS core, open source and self-hostable
- Assessment rubric marked out of 20 and a submission template
- Delivery in real Lahore classrooms, and curriculum mapped to SNC and PCTB, Cambridge and KHDA
Still a plan
- Every revenue and user number from Year 1 onward, including 20,000 free learners and 600 paying families
- Free-to-paid conversion at any rate, which has never been measured at scale
- Projects P03 to P12 in each lab, which exist as titles and outlines rather than written steps
- Institutional recognition of EAD levels, a Year 5 target of three institutions
- Sponsored labs and campus revenue, with one sponsored lab targeted in Year 1
- Longitudinal outcome data, which cannot be checked before 2036
The reasons to say no
There is no longitudinal outcome data, so the ten-year claim is checkable in 2036 and not today. Conversion from free to paid is unproven at scale, human certification does not scale linearly and is deliberately the expensive part, only the opening project in each lab is written step by step, and the market's willingness to pay for a new credential is untested. For this to be large, one thing has to be true: that a meaningful fraction of students who finish their first free project start a second one in a different lab.
The plan, labelled
Year 1, Year 5, Year 10
Every figure below is a target from the plan. None of it is achieved, and the last one is a claim that settles publicly.
- Year 1 · to July 2027Target: prove the second project20,000 students completing at least one free project, 25 paying campuses, 600 paying families, 100 professionals certified, 1 sponsored lab, roughly US$13,000 to US$16,000 per month by month 12. Targets, not results.
- Year 5 · 2031Target: recognition and scale150,000 free learners, 500 campuses, 6 sponsored labs, 3 institutions formally recognising EAD levels, US$1.8M to US$3M ARR. By then Level 2 and 3 learners write projects for those below them as part of their own assessment. All targets.
- Year 10 · 2036Target: the claim gets checkedThe 2026 cohort turns twenty holding 50+ certified projects and a network of about a thousand people, at the age their classmates begin sending CVs. A target. The longitudinal data will settle it publicly, including if it disappoints.
Free download
The EAD Investor Brief
Four documents covering the model, the prices, the planned unit economics and the curriculum map.
- 01A one-page summary of the wedge, the twelve labs and the three payers
- 02The model and pricing detail: EAD Plus, certification, School Edition, the Guaranteed Term Pilot, Teacher Certification and sponsored labs, at real prices
- 03The unit-economics sheet: about US$2 AI marking, US$12 human certification, US$2 parent reports and US$8 platform share per paying family per year
- 04The twelve-lab curriculum map against SNC and PCTB, Cambridge and KHDA, with [ROUND SIZE AND USE OF FUNDS — CONFIRM] stated by the founder on request
Here it is
We have emailed these links as well, so they are easy to find later.
Before you write back
Questions we get asked
If yours is not here, write to us. The answer usually ends up on this page.