Our methodology

Every broker we review is scored against the same five categories, each rated from 0 to 10 and weighted to a single overall score out of 10, after first passing a safety gate. The weighting reflects what actually changes a self-directed investor's outcome, so cost and the platform you use every day count for far more than features where every broker has converged. This page sets out the full rubric, the reasoning behind each weight, our data sources, our update process, and exactly how the site is funded. If you ever find a figure you think is wrong, the corrections process is at the bottom.

01Our promise

Independence is the entire point of this site. The rules below are not aspirational, they are the conditions under which we publish.

• No broker pays for placement or a higher score. Rankings are never sold, and no brokerage sees or approves its rating before publication.
• Affiliate revenue is weighted at zero. Whether we earn a commission when you open an account has no input into the rubric and no effect on any score.
• No manufactured differentiation. When an honest comparison shows two brokers are equally good or equally weak on something, we say so rather than inventing a distinction to fill space.
• Weaknesses are reported plainly, including for brokers we have a commercial relationship with. A partnership never buys a broker a pass on a poor platform, slow support, or high fees.

02The safety gate, first

Before any broker is measured or scored, it has to clear one bar: it must be regulated by CIRO and its accounts must be covered by CIPF, with a clean regulatory record and sound ownership. A broker that does not meet this is not given a low score, it is simply not listed or recommended at all.

Safety is a gate, not a score. Every broker that clears it is then judged purely on the five scored categories below, because among brokers that all clear the same bar, safety no longer tells you which one to choose. We still describe each broker's regulator, what CIPF covers, its ownership, and its track record in the review itself, lower in the page. One accuracy note: SIPC, the US protection scheme, does not cover Canadian clients, so it is not a point of difference between Canadian-domiciled brokers. A serious, current regulatory failure does not lower a score; it removes a broker from the site entirely.

03What we measure

Having cleared the safety gate above, every broker is scored on five categories. Each is built from specific, verifiable components, not impressions, and the components and their weights are set out in How we score below.

• Fees and cost. Stock and ETF commissions, currency conversion (FX) markups, account and inactivity fees, margin interest, options pricing, and other transaction fees.
• Platform and usability. A composite of six parts, each scored on its own: mobile app design, web and desktop platform design, ease of use, day-to-day functionality, account-opening friction, and the range of account types offered.
• Market and product access. The asset classes you can trade (stocks, ETFs, options, bonds, GICs, mutual funds, direct crypto, and niche instruments), the markets you can reach including international exchanges, fractional shares, and the ability to hold USD inside registered accounts.
• Research and tools. Fundamental and analyst research, charting, screeners, market data and news, options analysis, portfolio tools, and education inside the platform.
• Customer support. Channels, hours, responsiveness, resolution quality, and self-serve resources.

Three things we show but deliberately do not score: sign-up promotions, app store ratings, and Trustpilot ratings. Promotions change too often and reward marketing spend rather than quality. App store and Trustpilot scores are too noisy and too easily skewed to belong in a quality judgement. You will find all three in our comparison table and on the relevant pages, presented as information to weigh yourself, never folded into the score.

04How we weight each category

Not every category matters equally to a self-directed investor. The weighting below puts cost first and the everyday platform experience second, because those are what compound over years of investing.

Fees and cost, 30 percent. The single biggest long-term driver of returns and the reason this site exists. We hold it at 30 rather than higher because commission-free stock and ETF trading is now standard in Canada. The real cost differences have moved to FX markups, options pricing, margin rates, and USD-account fees, so we score those carefully rather than over-rewarding a headline of zero-dollar trades that almost everyone now offers.

Platform and usability, 30 percent. This is where you actually live as an investor, and it now carries as much weight as fees. Rather than score it on a single impression, we break it into six parts and score each one: the mobile app, the web and desktop platform, ease of use, day-to-day functionality (order types, recurring investing, fractional trading, transfers), account-opening friction, and the range of account types offered. The parts are blended into one platform score, which is why a broker with a beautiful app can still land in the 7s if its functionality is thin or its account lineup is narrow.

Market and product access, 20 percent. This genuinely varies broker to broker, from North-America-only apps to platforms reaching dozens of international exchanges, and it covers whether you can trade options, bonds, or direct crypto at all. Real differentiation earns real weight, so we raised it to 20 percent.

Research and tools, 10 percent. Deliberately modest. High-quality charting, screening, and market data are freely available off-platform, so a thin in-house research suite is a minor penalty rather than a dealbreaker. We separate it from usability because a broker can have a beautiful app and weak research, and that split is exactly what powers our "best for beginners" versus "best for active traders" calls.

Customer support, 10 percent. Easy to ignore until the day you need it. Support quality varies widely across the market and is what you are left with when something breaks, so it carries meaningful weight.

Account types. No longer a category of its own. Because most brokers offer most account types, a standalone score rarely separated one broker from another, so we moved it inside Platform and usability, where it is one of the six scored parts. The cases where it matters, such as needing an RDSP or a corporate account, are also called out directly in each review's "best for" and "not for" sections.

Safety and trust, a gate rather than a score. Regulation is near-uniform among the brokers we cover, so a safety score returned the same number for almost everyone and did nothing to separate them. We now treat safety purely as a pass-or-fail gate, explained next, and discuss it in prose in each review rather than assigning it a weight.

05How we score

Within each category, a broker is rated from 0 to 10 against fixed anchors, so a score means the same thing in every review and from one year to the next.

Each category is itself broken into scored components, each rated 0 to 10 on what it owns and blended at fixed weights. This is what keeps a score defensible: it is built from specific parts, not a single impression. The fixed component weights inside each category are shown below.

Each fact counts once, in the category and component it belongs to. A thin research tool lowers the research score, not the platform score; a high currency fee lowers the fee score, not the market-access score. This is how two brokers with the same strengths and weaknesses end up with the same category scores, and why a score on one review can be compared directly with the same category on another. The published review shows the five category scores; the component scores behind them are recorded in our internal rubric and available on request.

The overall score is the weighted sum of the five category scores, reported to one decimal place out of 10. We publish the score as a clean number rather than as stars, because a number tells you more: it lets a 7.5 and an 8.9 read as the genuinely different brokers they are.

06Where our data comes from

We work from primary sources first and treat everything else as something to be confirmed, not trusted on sight.

07How often we update

• Figures are verified against primary sources before publishing, not pulled from memory or older articles.
• Promotions are re-checked frequently and confirmed before every use, because sign-up offers change more often than anything else on the site.
• Every table and review carries a visible "last reviewed" date so you always know how current the information is.
• When a broker changes its published pricing or features, we re-check and re-stamp the affected pages rather than waiting for a fixed annual refresh.

08Our standard for crypto

We only point readers to crypto trading platforms registered with Canadian securities regulators (the CSA and CIRO). Unregistered platforms may offer lower fees or more tokens, but they sit outside the protections Canadian investors are entitled to, so they do not appear in our recommendations.

09How we make money

Broker Guide Canada is free to read and funded through affiliate and referral relationships. Here is exactly what that means.

• Some outbound links to brokers are affiliate or referral links. If you open an account through one, we may earn a commission at no additional cost to you.
• This never changes a score, a ranking, or a word of our analysis. Affiliate revenue is weighted at zero in the rubric, full stop.
• Not every broker we cover pays us. Some links earn nothing, and we include those brokers anyway when they belong in the comparison.

Full details are on our disclosures page.

10Corrections

We aim to be the most accurate broker comparison in Canada, and accuracy is a moving target. If you spot a figure that looks wrong or out of date, tell us. We will check it against the broker's primary sources, correct it if needed, and update the review date so the change is visible.