Teaching abacus classes without buying a franchise licence

You can teach mental abacus without buying a licence. That sentence is hard to find online, because much of what ranks for abacus teacher training is published by the parties selling the training — and, as the Jordanian example below shows, a free trainer course can end in an employment contract with a network’s own centres. This post separates three things that are usually sold as one: the skill of teaching bead work, a certificate that says you have it, and a commercial licence to trade under somebody else’s brand.

What a trainer certificate gets you — and who is issuing it

Two documented pathways show the range. Oxford Academy in Jordan — a private academy wholly unrelated to the University of Oxford — advertises a 180-hour track to qualify as an ACMAS mental-arithmetic trainer: 60 hours of instruction plus 120 hours of field training over three months, delivered as free training with a certificate and employment contracts for top performers, aimed at recent graduates in mathematics and education. At the other end, SMV Academy sells an Arabic-language professional training diploma covering soroban mental arithmetic in three levels and Vedic mathematics in four, priced on its page at USD 830 with a discount advertised alongside it, and states that it is registered in London and appears on the UK Register of Learning Providers. Read that last part slowly: a register of learning providers is an administrative listing, not academic accreditation, and the two are easy to blur when they sit in the same sentence. Prices move, so re-check any figure before you pay it.

• Find the body actually named in the page body, not the one the page title gestures at. Where a title implies one accrediting institution and the body copy names private commercial academies, the body copy is the accurate part.
• Ask whether the named body accredits the programme or merely lists the provider. Company registries and provider registers are administrative records; they assess nothing about your teaching.
• Check whether a university certificate is part of the course or a paid optional add-on sold on top of it. An add-on accredits nothing.
• Ask what the certificate lets you do that you could not do without it — then check the actual rules where you teach, because private tutoring is regulated very differently from one country to the next.
• Ask who employs the graduates. Where a trainer course is free and ends in an employment contract, the certificate is a recruitment step for that network’s centres — which is a fair deal, as long as you know that is the deal.

The four things you genuinely need to teach a class

• A published ladder, so the sequence is decided before the first lesson rather than improvised weekly. Kani uses nine levels from Foundation to Grand, with level set as an explicit placement rather than derived from a child’s age.
• Graded material that already matches the ladder — worksheets and drills per rung with answer keys — so your prep time goes into watching hands, not typesetting sums.
• A way to set homework and see whether it was actually done. This is the entire job between weekly sessions, and it is where informal classes quietly fail.
• A per-child progress record you can show a parent in thirty seconds: which level, what was practised, how accuracy moved. Parents renew on evidence, not on enthusiasm.

What a franchise licence adds — and what it costs

A licence buys four real things: a brand some parents already recognise, a defined territory, an exam and certificate pathway your students can sit, and a supplied curriculum you do not have to build. It also comes with a fee structure, and that structure is ultimately funded by the families in your class. Two notes before the table. First, local pricing is inherent to franchising rather than a quirk of any one network: Kumon UK’s own FAQ says the registration and monthly fee varies slightly from centre to centre and points parents to their local study centre, and Brazil’s SUPERA answers its own page headed with the monthly-fee question by explaining that the fee varies with location, package and current offers, then inviting readers to contact a nearby unit. Second, no operator in this category publishes audited figures for centres, students or outcomes — Kani included. Everything below is as published by the party named, retrieved 30 July 2026.

Your first cohort: group size, session length and the daily loop

• Group size, as published: ALOHA Ireland caps classes at a maximum of 10 students, a SIP Abacus franchisee publishes a maximum of 15, and Mastermind Abacus teaches its live online programme at a 1:5 teacher-to-student ratio. Start nearer five than fifteen — correcting finger technique happens one child at a time.
• Session length, as published: one long weekly session is the category norm. SIP Abacus’s own FAQ describes weekly classes lasting around two hours; Mastermind Abacus publishes twice a week for one hour each.
• One teacher’s view of that norm, from a teaching blog in July 2024: she observes that abacus classes usually run for about 75 minutes, which can be quite a stretch for young kids, that five- and six-year-olds’ focus lasts about 10–15 minutes before they start chatting or asking for a break, and that among her own young students a few left after just one day. That is one practitioner describing her own classes with no retention data — but it is the failure mode to design your timetable against.
• The daily loop moves children, not the weekly session. SIP Abacus’s FAQ puts daily practice at approximately 10 minutes, and the Barner et al. 2016 trial in Vadodara, India ran three hours a week for three years. Ten minutes at home, six days a week, is the thing you are really selling to parents.

Set expectations before the first invoice, because the research on short timelines is genuinely mixed and a private class has no brand to hide behind. Barner and colleagues’ 2016 randomised controlled trial in Vadodara — 204 children enrolled, 183 completing, three years at three hours a week — found mental-abacus students outperformed controls on arithmetic (Cohen’s d = .60, 95% CI [.30, .89] on the study’s own measure), but reported that the training did not alter basic cognitive abilities, and that baseline spatial working memory moderated how much children learned, with those below the median not differing from controls. A separate one-year classroom-randomised trial by the same group in 2017, with 180 children, saw no evidence of differential change on either its in-house arithmetic measure or the standardised WJ-III, found only a marginal place-value trend in second graders (p = .052), noted that only 21% of first graders could accurately decode multi-digit abacus representations by year end, and concluded by suggesting caution in adopting mental abacus as a short-term educational intervention. Chunjie Wang’s 2020 review in Frontiers in Neuroscience notes that most studies detecting benefit did so after around three years and describes the gains as near transfer — improvement on tasks tightly related to the trained tasks. So promise a habit, clean bead technique and arithmetic fluency over years. Do not promise anything about intelligence.

Assessment when you have no exam board behind you

You do not need a franchise to give a child something to sit for. Japan splits its soroban proficiency exam across two bodies: grades 1–3 are organised by the Japan Chamber of Commerce and Industry with the Japan Abacus Federation co-sponsoring, while quasi-grades 1–3, grades 4–6 and grades 7–10 are organised by the Federation with the Chamber co-sponsoring. Pass marks are 240 of 300 for grades 1–3 and the quasi-grades, 210 of 300 for grades 4–6, and 120 of 200 for grades 7–10 except grade 10 at 60 of 200, with fees effective April 2024 running from JPY 1,000 to JPY 2,800. A second body, the League for Soroban Education of Japan, runs a parallel ladder of 15 kyu grades plus dan ranks at JPY 700–3,500 and — useful for anyone teaching flash work — a formal flash-anzan certification with 20 kyu grades and 10 dan ranks. PAMA Global, a Taiwan-based association, publishes downloadable sample exam papers across levels 1–12 plus Advanced and runs an annual competition. Between an external sitting once a year and a mock paper you print yourself each term, your cohort has milestones without a licence.

Setting up a class in Kani

Kani is the independent-route option we build, so here is its shape plainly. Nine levels from Foundation to Grand; level assigned as an explicit placement rather than inferred from age; a worksheet generator that prints graded papers with answer keys in English, Arabic, Spanish, French and Portuguese; the 200-question competition papers above; and native Android and iOS apps for the daily practice half. Pricing is about USD 7 per month with a free tier and self-serve checkout — cheaper than the centre tuition in the table above, and more expensive than several apps: Abacus4Everyone publishes INR 399 per month for students and INR 999 per month for teachers with a teacher dashboard that assigns homework and issues PDF belt certificates, and Indian Abacus hosts a worksheet generator whose free use is capped by a question limit, with unlimited access sold as a paid upgrade. Compare the teacher dashboard, the homework flow and the papers you would actually hand out, rather than the headline number.

ALOHA, Brainobrain, SIP Abacus, Mastermind Abacus, Kumon, SUPERA, Smart Brain, Abacus4Everyone, Indian Abacus, SMV Academy, Oxford Academy, ACMAS and PAMA Global are trademarks or trading names of their respective owners. Flash Anzan is a trademark of its rights holder. Kani is not affiliated with, endorsed by, or licensed by any of them, and every figure above is as published by the party named and retrieved on 30 July 2026.