Skip to content
FinTech IQ
Get in touch

Canadian financial technology

Everyone has done the training. Not everyone can read the file.

FinTech IQ teaches the Canadian financial-technology sector — how money actually moves, how credit is decided, and where the regulatory lines sit — to the people doing the work: account executives and underwriters, compliance and operations staff, and the product and engineering people beside them. It measures what they can demonstrate rather than what they have clicked through.

Try one questionSee the curriculum
Lessons
48
Simulations
7
Competencies
10
Knowledge entries
97
01

Why this exists

Completion is easy to measure. It is not what you needed to know.

Training in financial services tends to arrive as one of two things. There is compliance content, which exists to be finished and recorded, and there is product training, which teaches one company’s process and stops at the edge of it. Both produce the same artefact: a record that somebody sat through it.

Neither answers the question a manager actually has: whether this person can read a file, price a risk, recognise where a regulatory line sits, and say so out loud to somebody who would rather they did not.

Someone can complete every lesson and still not know that a loss ratio compared across cohorts of different ages is not a comparison, that a debt-collection rule differs by province, or that the number in a policy is a fact about a regulator’s current guidance rather than a fact about the world.

None of that is exotic. It is the ordinary content of the work, and it is invisible to a completion percentage.

02

The approach

Evidence, dated — and allowed to go stale.

FinTech IQ holds no score. It holds a dated, append-only record of what somebody has actually done: a lesson finished, an assessment passed, a simulation worked through, a review session completed. From that record it computes a competency every time you look at it.

Different evidence weighs differently

Working a simulation is stronger evidence than reading a page, and the model says so, instead of treating both as one completion. A score is a summary of the record; the record is the thing that is kept.

Freshness decays

Evidence loses weight as it ages, on a half-life. Something demonstrated two years ago and never since is not the same claim as something demonstrated last month, and a model that treats them alike is telling a manager a comfortable untruth.

Coverage is not the same as score

How much of a domain has been touched is tracked separately from how well. High marks across a narrow slice are a different situation from steady marks across the whole of it, and collapsing the two hides exactly the case worth knowing about.

The result is a shape across 10 domains, not a number: strong here, thin there, ageing somewhere else. That profile is what the backdrop at the top of this page is drawing. A flat chart can show strong and thin. It has nowhere to put ageing, which is the axis that decides whether the other two still mean anything.

  • SMB FinanceReading a small business — revenue, margin, working capital, seasonality — and knowing what its numbers do and do not tell you.
  • Commercial FinanceFinancing products, structure, pricing, and the discipline of matching a product to the shape of a need.
  • Credit & UnderwritingAssessing capacity and character from evidence, and writing a decision another professional could review in a year.
  • RiskNaming, sizing and owning exposure — credit, operational, fraud, concentration, and third-party.
  • Compliance & Financial CrimeThe Canadian AML regime, identification obligations, privacy, and the escalation habits that go with them.
  • PaymentsHow money actually moves — the card chain, settlement timing, cost of acceptance, and the Canadian rails.
  • Financial TechnologyAPIs, data infrastructure, identity, and model governance — enough to work confidently alongside engineering.
  • Customer CommunicationConsultative discovery, explaining cost honestly, and the line between persuasion and pressure.
  • Operations & ServiceThe originations funnel, verification, servicing, and the controls and handoffs that hold them together.
  • Partnerships & GrowthDistribution through partners, embedded finance, and the unit economics that decide whether a channel is worth having.
03

One question

Is this book getting better or worse?

Six origination cohorts from a synthetic lending book. Loss to date is cumulative: each cohort measured at whatever age it happens to be. Answer before you scroll.

Synthetic portfolio. Every figure is invented for teaching and is not any lender’s experience.
CohortAgeShare of bookLoss to date
V118 months on book8.2%7.8%
V215 months on book9.1%7.2%
V312 months on book11.4%6.9%
V49 months on book15.8%5.6%
V56 months on book21.3%4.9%
V63 months on book26.7%1.9%

Loss to date has fallen from 7.8% to 4.9% across the book. What is happening?

FinTech IQ is running here: the real application, not a cut-down preview. Access is arranged rather than self-served; the form at the bottom of this page is the way to it, and a walkthrough is usually faster than a login.

What an account opens on is a worked record, not an empty one, and a deliberately uneven one: strong in small-business finance, thin in credit, weak in payments, the compliance work old enough to have started losing weight, and one domain carrying no evidence at all. A flattering profile would demonstrate nothing.

Nothing here is stored or sent. This is a demonstration, not an assessment.

04

The obvious question

Why not use what you already have?

Often you should. These are categories rather than products, because what can be said fairly without inspecting somebody else’s software is what a category is for, and because the real choice is between ways of working, not between two logos.

  • A compliance catalogue

    Exists to be finished and recorded. The output is a completion date, which answers whether somebody sat through it.

    Holds a dated record of what was actually done and computes a competency from it every time you look. Completion is one input among several, and the weakest one.

  • A learning management system

    Hosts and tracks whatever content you give it. The content, and whether it is right, is somebody else’s problem.

    Is the content: 48 lessons, 97 knowledge entries and a competency model, authored against Canadian regulatory ground with jurisdiction as a required field.

  • A generic fintech course

    Usually written for the largest market and silently assumes its rules. A reader cannot tell which parts travel.

    Canada-first. Jurisdiction is declared on every regulatory lesson, and the build fails on one that is missing it.

  • Building it internally

    Produces material that fits your process exactly, and carries your assumptions with it, including the ones nobody has re-examined.

    Is independent of any single institution’s method, and every figure in it is invented, not borrowed. An enterprise deployment can add its own authored content on top and owns it.

  • A quiz bank

    Measures recall. A reader who has seen the question before scores the same as one who understands it.

    7 simulations where the obvious move is reasonable and still wrong, and every figure in every debrief is re-derived from the same data the exercise ran on.

05

What is in it

48 lessons, 12 tracks, 3 pillars.

Every lesson is the same shape: three reading sections with takeaways, a scenario with one defensible answer and feedback on every option, and eight questions. That is 144 sections, 432 takeaways, 48 scenarios and 384 questions — about 25 hours of taught material.

01

The System

2 tiers · 4 tracks · 16 lessons

What financial technology actually is, how money moves through the Canadian payment rails, the infrastructure and data underneath it, and the operational machinery that turns an application into funded capital.

Tier 1 — What Fintech Is and How Money Moves

Fintech Foundations4 lessonsreg
  1. What Financial Technology Actually Is
  2. The Canadian Financial-Services Ecosystemreg
  3. Fintech Business Models
  4. Fintech Economics and What Kills a Fintech
Payments and Money Movement4 lessonsreg
  1. The Card Payment Chain
  2. Authorization, Clearing and Settlement
  3. Interchange and the Cost of Accepting Cardsreg
  4. Canadian Payment Rails Beyond Cardsreg

Tier 2 — The Infrastructure Underneath

Technology and Data4 lessonsreg
  1. APIs and API-Driven Financereg
  2. The Data a Lending Decision Runs On
  3. Identity, Authentication and Encryptionreg
  4. Models, AI and Model Governancereg
Operations and Customer Experience4 lessonsreg
  1. The Originations Funnel
  2. Verification and Funding Workflows
  3. Servicing, Delinquency and Collectionsreg
  4. Operational Controls and Service Levels
02

The Money

2 tiers · 4 tracks · 16 lessons

The financial life of a Canadian small business, the commercial financing products built to meet it, the credit decision that stands between the two, and the risk the lender carries afterwards.

Tier 1 — The Business and the Capital

Small Business Finance4 lessonsreg
  1. The Canadian Small Business Landscapereg
  2. Revenue, Margin and Operating Expenses
  3. Working Capital and the Cash Conversion Cycle
  4. Seasonality and Cash-Flow Forecasting
Commercial Lending and Financing4 lessonsreg
  1. The Commercial Financing Landscape
  2. Cash-Flow-Based and Merchant Finance
  3. Pricing, Cost of Capital and Risk-Based Pricingreg
  4. Matching the Product to the Need

Tier 2 — The Decision and the Exposure

Credit and Underwriting4 lessonsreg
  1. What Credit Actually Is
  2. Capacity and the Five Cs
  3. Reading Bank Activity as an Underwriter
  4. Credit Bureaus and the Defensible Decisionreg
Risk4 lessonsreg
  1. The Risk Taxonomy
  2. Portfolio and Concentration Risk
  3. Risk Appetite and the Three Linesreg
  4. When a Loan Goes Bad
03

The Practice

2 tiers · 4 tracks · 16 lessons

Conduct. Anti-money-laundering and the FINTRAC regime, fraud detection and defensive security, the consultative customer conversation, and the partnership economics that distribute a financing product.

Tier 1 — Trust, Crime and Control

Compliance and Financial Crime4 lessonsreg
  1. The Canadian AML Regime and FINTRACreg
  2. KYC, KYB and Beneficial Ownershipreg
  3. Suspicious Transactions and Sanctionsreg
  4. Privacy, Records and Consumer Protectionreg
Fraud and Security4 lessonsreg
  1. Identity Fraud and Synthetic Identities
  2. Application and Document Fraud
  3. Account Takeover and Social Engineering
  4. Escalation and Incident Responsereg

Tier 2 — The Customer and the Market

Sales and Account Management4 lessonsreg
  1. Consultative Discovery
  2. Explaining Cost Without Hiding Itreg
  3. Objections, Pressure and the Linereg
  4. Account Management and Renewal
Partnerships and Growth4 lessonsreg
  1. Channel and Referral Partnershipsreg
  2. Embedded Finance and Banking-as-a-Servicereg
  3. Unit Economics: CAC, LTV and Payback
  4. Managing a Partner After the Integration

reg marks regulatory content — 20 of 48 lessons. Each of those declares its jurisdiction, and the product’s build fails if one does not.

8 routes through it, by job

Nobody takes 48 lessons in order. A path is an ordered run of lessons, simulations and an assessment, assembled for one role, and an enterprise academy adds its own on top of these without replacing them.

  • Entering Fintech8 stepsPeople moving into financial technology from another sector.
  • Account Executive / Account Management15 stepsPeople who talk to small-business owners about financing.
  • Credit and Risk9 stepsAnalysts and underwriters, and anyone whose recommendation reaches a credit decision.
  • Compliance and Financial Crime7 stepsAnyone whose role touches identification, monitoring or escalation — which is most front-line roles.
  • Payments and Infrastructure6 stepsProduct, engineering and operations people who need the topology in their heads.
  • Operations and Servicing6 stepsThe people who turn an application into funded capital, and who hold the relationship afterwards.
  • Technology and Data6 stepsEngineers, product and data people, and anyone who has to evaluate a technical claim without being able to check it themselves.
  • Partnerships and Distribution9 stepsPeople who build and manage the channels that bring customers in.
06

Where it stops being reading

7 simulations, each with a defensible wrong answer.

A simulation is not a longer quiz. Each of these puts a learner inside a situation where the obvious move is reasonable and still wrong, and every figure in every debrief is re-derived from the same data the exercise ran on, so the teaching cannot drift from the arithmetic.

  • The Financing Desk

    Assess a financing request from an incomplete file.

    A fictional Canadian restaurant wants capital for a second location. You have their numbers, some documents and a gap where the important information should be. Decide what matters, what is missing, what you would ask, and what you would recommend.

  • Cash Flow Explorer

    Watch a business run out of money while staying profitable.

    Drive a fictional business month by month. Move revenue, margin, payroll, supplier terms, receivable timing, seasonality and financing, and watch the cash position respond. The lesson that everyone nods at and nobody believes until they see it happen.

    Best on a larger screen.

  • The Payment Journey

    Follow one payment through every party that touches it.

    A $48.50 card payment at a fictional café, traced from tap to settlement. Step through authorization, clearing and settlement, inspect each participant, see what information moves and what money moves, and find out where it can fail.

  • The Customer Conversation

    A branching conversation where the fast answer is the wrong one.

    A fictional SMB owner needs money quickly and does not understand the options. Your choices change what she tells you next. Scored on discovery, honesty, suitability and respect — including whether you noticed the thing she said in passing.

  • The Fraud Desk

    Four applications. Some are fine. Find what does not add up.

    Review synthetic applications, documents and transaction patterns. Flag inconsistencies, identify what is missing, and decide what warrants escalation — and what does not, because over-flagging is its own failure.

  • The Portfolio Desk

    The headline is improving and the book is getting worse. Both are true.

    Six origination cohorts, five origination channels and a concentration table for a fictional book. The whole-book loss ratio has fallen; measured at the same age, comparable cohorts have got substantially worse. Find what the numbers support, resist what they only suggest, and decide what to do about the largest partner.

    Best on a larger screen.

  • The Broker Desk

    Eleven broker relationships, one week, and time for five calls.

    A week on the broker desk of a fictional Canadian lender. Eleven relationships want something, four items are waiting in the inbox, and there is time for five calls. The scarce resource is not information but attention: decide where the week goes, work the inbox, and take the call from your biggest broker.

    Best on a larger screen.

7 assessments, 105 items to sit

Assessments draw from a typed bank, not a fixed list, so a second sitting is a different paper. Passing one issues an educational certificate with a checksummed identifier, and that certificate says on its own face what it is not.

What each assessment covers, and what it takes to pass7 assessments
  • Fintech Foundations15 items · 75% to pass · 25 minThe sector, the Canadian ecosystem, how money moves, and the machinery that turns an application into funded capital.
  • Small Business and Commercial Finance15 items · 75% to pass · 25 minReading a small business, the working-capital cycle, the financing product landscape, and matching a product to a need.
  • Credit and Risk15 items · 75% to pass · 25 minCapacity, the five Cs, defensible decisions, and the risk taxonomy a lender actually runs on.
  • Compliance and Financial Crime15 items · 80% to pass · 25 minThe Canadian AML regime, identification obligations, beneficial ownership, and the escalation habits that go with them.
  • The Customer Conversation15 items · 75% to pass · 25 minConsultative discovery, explaining what capital costs in a way the customer can check, the line between persuasion and pressure, and the account relationship after the money moves.
  • Partnerships and Growth15 items · 75% to pass · 25 minChannel and referral relationships, embedded finance and who carries which obligation, the arithmetic of acquisition and retention, and governing a partner after the integration ships.
  • Technology and Data15 items · 75% to pass · 25 minWhat an API actually promises, the three data sources a lending decision runs on and what each cannot see, identity and encryption claims worth testing, and the governance a model that decides about people has to carry.
07

In a room

Advancing a slide is not evidence.

Every lesson can be presented rather than read: projected, driven from the keyboard, with facilitator notes and a clock. The deck is generated from the lesson itself, so there is no second copy of the content to drift out of step with the first. What changes in a room is not the type size; it is who is asserting the material and what the session is allowed to count for.

  • The session records nothing

    No competency evidence, no progress, no attempt.

    A facilitator clicking through slides has demonstrated nothing, and neither has the room. Presentation mode writes no evidence, no experience, no progress, no achievement, no assessment attempt, no simulation run and no streak: seven write paths it is forbidden to touch, each one named in the product’s own checks, which fail the build if any of them appears. Evidence earned by pressing the right-arrow key would turn the competency model into an attendance register.

  • The provenance gets louder, not quieter

    Jurisdiction and review state, pinned to every regulatory slide.

    A learner reading an unreviewed regulatory lesson alone can weigh it. Thirty people being told it by somebody standing at the front cannot. So on a regulatory lesson the jurisdiction and review state sit in a persistent bar on every slide, not in a strip at the foot of a page, and it is the one piece of chrome with no setting to hide it.

  • The answer is not on the screen

    Scenarios stay unanswered until the facilitator reveals them.

    A scenario whose correct option is already visible is not an exercise, it is a statement. The room argues first; the answer and the feedback on every other option arrive when the facilitator decides they should, which is the difference between asking the room a question and telling them the answer.

  • One source, so the deck cannot drift

    Generated from the lesson, not authored beside it.

    A deck maintained alongside a lesson is a deck that diverges from it, and the divergence is invisible because both halves stay individually valid. Nothing here is a second copy: the title, the three sections, the scenario and the closing slides are derived from the lesson that already exists, so correcting the lesson corrects the deck.

If facilitated delivery ever needs recording, the thing being recorded is attendance, and it belongs somewhere that is not the competency model.

08

How it is checked

A measurement that flatters is worse than none.

The recurring question behind all of this is whether a check can actually see the fault it claims to guard against. More than once here the answer was no: a contrast sweep reporting zero because it could not parse the colour notation it was looking at, a bound that was never reached, a test whose defect had never been injected. Those are recorded, not quietly fixed.

  • Nine automated checks, each proven against the defect it exists to catch

    Writing a check is not the same as knowing it works. Every one of them has had the fault it exists to catch deliberately injected, and been watched to fail. Several passed a broken build on the first attempt and had to be rewritten, which is the entire argument for doing it.

  • The arithmetic in every simulation is re-derived, not transcribed

    A debrief that quotes a figure the exercise did not produce is teaching something the software does not do. The checks recompute each one from the same data the exercise ran on, including the requirement that a channel table blend to exactly the cohort it describes.

  • Contrast is measured on rendered pixels, in four display modes

    Not read off a stylesheet. Foreground against composited background, on every visible text node, in light, dark, and both high-contrast ramps, then again in the second language. It found faults that twelve rounds of careful authoring had not.

  • Jurisdiction is a required field, not a habit

    A regulatory lesson that does not declare its jurisdiction fails the build. Canadian rules are not presented as universal, because a learner who carries an Ontario rule into Alberta has been given a confident wrong answer.

  • Every lesson carries its provenance

    Owner, author, version, status, jurisdiction, sources, when it was created and when its next review is due, plus the reviewer and review-date fields, which are empty on every lesson because no expert has signed one off yet. That is what makes a review possible at all: an expert can see what they are being asked to check, and the record shows plainly that nobody has.

  • The limitations are audited, not asserted

    A limitation’s justification is a factual claim, and factual claims go stale. The product’s own roadmap records auditing each one and finding that one was false and one was the wrong claim to be making.

09

What you walk away with

A certificate is only worth what a stranger can check.

Passing one of the 7 certification assessments issues a certificate that names the competencies it covers and the score at issue, because one that says only “passed” tells a reader nothing. The harder question is what a reader can establish about it without taking anybody’s word. Three things get asked. Two of them are settled.

  1. 01

    Was this identifier typed correctly?

    Settled

    Every certificate identifier carries a check character. It catches 100% of the look-alike confusions a person reading one off a screen actually makes (0 for O, 1 for I, 5 for S), about 99% of single-character typos and about 96% of two adjacent characters swapped. Those are measured from the shipped implementation on every build, not estimated: measuring them is what found the original multiplier catching only four transpositions in five, and replaced it with the one shipped now.

  2. 02

    Has this certificate been altered since it was issued?

    Settled

    A verification code binds the identifier, the assessment, the date, the score and the holder’s name together into fifty bits of SHA-256. Change the score from 62% to 92% and leave the code alone, and the verifier prints 62% beside the paper that says 92%. Anyone can run that check, on any device, with no account and nothing to look up. The verification page is public, and the code is readable, so a reader sees the name and the score before they paste anything.

  3. 03

    Did FinTech IQ issue this certificate?

    Not established

    This one is open, and no amount of care in the other two closes it. The algorithm that makes a code is part of the product and the product runs in the browser, so anyone can compute a valid code for whatever contents they like. A signing key would not help: it would ship in the same file, readable by everybody it was meant to exclude, and shipping it would be a control that only appears to exist. Proof of issuance needs a registry (a service that says yes, we issued that one), and so does revoking a certificate that should no longer stand. There is no registry here.

Open the verification page . No account, nothing to install. Paste any FinTech IQ verification code into it, including one from a certificate you have been shown by somebody else.

The certificate says all of this about itself, in the document and in its footer: it records an educational assessment, it is not accredited, and it is not granted by, affiliated with or endorsed by any regulator, financial institution or professional body. A credential that overstates itself is worth less than one that does not, because the first thing a careful reader does is check.

10

For your IT review

The answers, before you ask for them.

Security and accessibility are the two places where an overstatement does the most damage: one gets tested, the other is a legal representation. Each of these was checked against the source before it was written, and the things that have not been done are named alongside the things that have.

The network boundary, storage, the supply chain, passwords, sessions and accessibility6 answers

  • It cannot phone home, and that is enforced

    The Content-Security-Policy sets connect-src 'none': the page is not permitted to open a network connection of any kind. There is no fetch, no XMLHttpRequest and no beacon anywhere in the source. This is not a promise about what the product chooses to send; it is a rule the browser applies whatever the code does.

  • No backend, and no account on anybody’s server

    Everything runs in the browser and every record lives in local storage, sorted into five scopes: organisation, learner, work, device and tenant. That is a real limitation as well as a privacy property: progress does not follow a learner between devices, and the product says so, instead of implying otherwise.

  • Nothing is loaded from anyone else’s server

    Twenty third-party files (scripts, styles, fonts) are self-hosted with a SHA-384 recorded for each. The build fails if one is edited or if a content-delivery link is reintroduced. That closes a supply-chain dependency and stops a learner’s address being disclosed to a font host on page load.

  • Passwords are hashed properly

    PBKDF2 over SHA-256 at one hundred thousand iterations with a random salt per account, through the Web Crypto API. Failed sign-ins lock the account progressively: a minute after five, fifteen after ten, a day after twenty. The counter resets after an hour of quiet.

  • Sessions expire, and signing out clears the work

    An inactivity timeout ends the session, configurable and sixty minutes by default. Signing out clears the work scope, and which keys belong to that scope is derived from the storage table, not a list somebody maintains by hand. The hand-maintained version is what went stale in sibling products and leaked keys.

  • Accessibility is measured, not asserted

    The structure is built for WCAG 2.2 AA and checked against it. axe-core 4.10.2 last ran on 29 September 2026 and reported zero violations across all fourteen routes, in sixteen combinations: four account roles, two languages, two themes. Four roles and not one, because four routes are unreachable as a learner and had never been audited at all. That sweep is run by hand and is not part of the automated build, so it describes the build it was run against and not every build since. The product’s own contrast sweep goes further on colour, measuring rendered pixels instead of reading a stylesheet, across every route in all four ramps. No formal audit and no screen-reader session has been carried out. Built for it and verified as it are different claims.

Not claimed, because none of it has happened: no penetration test, no third-party security audit, no SOC 2 or ISO certification, and no formal accessibility audit. Local storage is not encrypted at rest.

11

Who made this

Built by one person, on purpose.

The Raspberry Company

FinTech IQ is the sixth product in a family of professional competency platforms, each covering a different sector and each built on the same argument: that what somebody can demonstrate is a more useful record than what they have completed. The engine, the competency model and the checking discipline are shared; the curriculum and the regulatory ground are not.

Independent of the sector it teaches

The curriculum is built from public regulatory material, public industry information and original work. It is not affiliated with, sponsored by or endorsed by any lender, processor or financial institution, and it describes no company’s internal method. That independence is what makes it defensible on its own terms rather than on somebody else’s.

It teaches. It is not counsel.

None of it is advice, and no part of it substitutes for a qualified person inside a regulated institution. The product prints that on every screen, not in a footer, and so does this. What this is and is not is set out in full a section below.

Everything here is checkable

Every figure on this page — 48 lessons, 97 knowledge entries, 105 assessment items, 20 regulatory lessons — is generated from the product, not typed by hand. If the product changes and this page is not regenerated, this project’s own check fails rather than the page quietly lying.

12

Where the line is

What this is not, stated first.

Financial services is a field where a confident wrong answer costs somebody money. So the limits are here, at the top, in the same type size as everything else, not in a footer, and not softened.

  • These certificates are not a designation.

    They are educational. No regulator grants them, none is accredited, and none counts toward a professional title. The certificate says so on its own face, in the body and in the footer, not only here.

  • It teaches. It does not advise.

    Nothing in it is financial, credit, investment, legal or compliance advice, and nothing substitutes for a decision made by a qualified person inside a regulated institution.

  • Every figure in it is synthetic.

    No number anywhere is a real lender’s criteria. A learner who left believing they knew a real credit box would have been mistaught, which is a worse outcome than teaching them nothing.

  • It describes no company’s proprietary method.

    No lender’s, processor’s or financial institution’s internal method appears anywhere in it, and none of them endorses it. Where the material does come from is set out under Who made this.

  • A Canadian rule is never presented as universal.

    Jurisdiction is a first-class field on every regulatory lesson, and the product’s own build fails if one is missing. The taxonomy can express 11 jurisdictions; today's lessons span 3 of them, and the rest are declared so the shape exists before the content does rather than being coverage we have.

  • Volatile numbers are kept at the lowest altitude the teaching allows.

    Where a real requirement turns on a threshold, a deadline or a rate, the lesson teaches the obligation and points at the regulator’s current published guidance, which is where the number lives and where it will still be right after it changes.

The regulatory content is awaiting expert review.

20 of the 48 lessons carry regulatory content, and none has yet been verified by a subject-matter expert. The product reports every one of them as unverified on its own administration screen, and its build reports them on every run. Built for accuracy is not the same as checked for accuracy, and until that review happens this is a teaching draft, not a reference.

The interface runs in English and Canadian French. The 48 lessons, 97 knowledge entries and 7 simulations are English only, a gap this states rather than papers over.

13

Questions

The ones worth asking first.

Has the regulatory content been checked by an expert?
Not yet. 20 of the 48 lessons carry regulatory content and none has been verified by a subject-matter expert. That review is the gate before this trains anybody, and the product reports every unverified lesson on its own administration screen and on every build.
Is it available in French?
The interface is, in Canadian French: navigation, forms, the dashboard, the competency taxonomy, all fifty level names, and 11 jurisdictions, with fr-CA number, currency and date formatting. The 48 lessons, 97 knowledge entries and 7 simulations are English only. That is deliberate rather than unfinished: unverified English can still be checked by a reviewer, and unverified French regulatory prose cannot.
Where do the numbers in the simulations come from?
They are invented, and constructed so the arithmetic teaches something. No figure anywhere is a real lender’s criteria or experience. The alternative — plausible-looking numbers a reader might carry into real work — is the failure mode this avoids on purpose.
Is my data sent anywhere?
No. The product runs entirely in the browser and keeps its records in local storage; there is no backend and no account on a server. That cuts both ways. Your data goes nowhere, and neither does your progress: it stays in the browser that made it.
Can it be run for a team?
There is an enterprise layer with tenant branding, assigned paths and an administration console. Team progress currently reports only the signed-in learner, which is a real limitation of a browser-only build and is stated on the page itself, not filled with an invented figure.
What does a certificate from this actually mean?
That somebody passed an educational assessment on a stated date, at a stated standard, and that the identifier on it verifies against a checksum. What it is not is set out in full a section above, and printed on the certificate itself.
What does it cost?
There is no published price. What determines one is scope: how many people, whether an enterprise deployment carries its own authored content, and whether expert review is being commissioned as part of the engagement. Ask, and the answer will be a number rather than a range.
14

Get in touch

See it properly.

The most useful thing is a walkthrough: a lesson, a simulation, and the competency model reacting to evidence that is deliberately stale. Twenty minutes is enough to tell whether this is worth more of your time. Bring whoever in your organisation is most sceptical that any of this can be measured; that conversation is more useful than the demo.

You can try one part of it right now without asking anybody: the certificate verifier is public and needs no account. Everything else is arranged, and a walkthrough is the quickest route to it.

No mail client? Write to matt@theraspberry.ca. This form composes a message in your own mail client: nothing you type here is transmitted by this page.