How to Research a Stock in Canada, Step by Step
Researching a stock means finding out two things: what the business actually does with the money it takes in, and what you would be paying for a share of it. Everything else is detail hanging off those two.
That sounds simple, and the reason it is hard in practice is that the numbers which answer it are not the numbers that get printed in the headline. A company publishes hundreds of figures every quarter. Two or three of them decide the answer, the rest are context, and the ones that decide it are usually a few pages deeper than the ones that lead.
This guide is the process, worked start to finish on one real company listed on the Toronto Stock Exchange. Every figure in it comes from that company’s own published results, and by the end you will have a sequence of questions you can run on any Canadian company, in order, using documents that are free and that the company is legally required to file.
The company is Dollarama Inc. It is used here because it is a business most Canadians have been inside, because its filings contain four different traps that a beginner falls into, and because working through it produces a genuinely interesting answer. It is not a recommendation. Nothing on this page is advice to buy or sell anything, and the point of the exercise is the method, not the verdict.
The habit that separates research from reading
Start with the single most useful thing on this page.
On September 16, 2026, Dollarama reported that sales for its second quarter rose 17.6%. That is the number that leads the release, and the number that would appear in almost any summary you read.
Sales in the Canadian business, which is more than nine tenths of the company, rose 8.5%.
Both figures are correct. The difference between them is a business in Australia that Dollarama bought part way through the previous year, so the comparison quarter contains thirteen days of it and the current one contains ninety-one. The headline is not wrong. It simply is not the growth rate of anything you could extend into next year, and the company publishes the number that is, in a table further down the same document.
The number that decides the answer is almost never the headline number, and every time it is not, the company has already published the one that is. The work of research is knowing which document to open and which line to find in it. That is what the rest of this guide teaches.
Everything a Canadian public company must tell you, and when
Before any analysis, know where the raw material is. This is the part of stock research that differs most between Canada and the United States, and most of what you will find written about it online describes the American system.
Canadian public companies file through SEDAR+, the system run by the Canadian Securities Administrators on behalf of every provincial and territorial regulator. It is free, it needs no account, and it holds the complete filing history of every reporting issuer in the country, and it is the public record the rules themselves point at. Search a company name at SEDAR+ and you get the documents themselves rather than somebody’s summary of them.
Two things are worth knowing before you go there. The first is that a company’s own investor relations page usually carries the same documents in a friendlier layout, and it is where the company posts the release itself. Dollarama’s investor relations page is where the release used throughout this guide was published. The second is that the filings are not optional and not at the company’s discretion about timing. National Instrument 51-102, the rule that governs continuous disclosure, specifies each document and the deadline for it.
| Document | What it answers | Rule | Deadline |
|---|---|---|---|
| Results news release | The quarter’s headline figures and how management frames them | NI 51-102 s.7.1 | Immediately on a material change, with a material change report within 10 days |
| Interim financial report | The actual statements for a quarter, with notes | NI 51-102 s.4.3 and s.4.4 | 45 days after the quarter ends, 60 for a venture issuer |
| Annual financial statements | The audited statements for the year | NI 51-102 s.4.1 and s.4.2 | 90 days after the year ends, 120 for a venture issuer |
| Management’s Discussion and Analysis | Management explaining the statements, including eight quarters of summary data | NI 51-102 s.5.1, Form 51-102F1 | The same deadline as the statements it relates to |
| Annual Information Form | The business described in full, and the risk factors | NI 51-102 s.6.1 and s.6.2, Form 51-102F2 | 90 days after the year ends. Venture issuers are not required to file one |
| Insider reports | What directors and officers bought and sold, and at what price | NI 55-104 s.3.3 | Within five days of the change, filed on SEDI |
Sources: National Instrument 51-102 Continuous Disclosure Obligations, unofficial consolidation, sections 4.2 p.16, 4.4 p.18, 5.1 p.35, 6.1 and 6.2 p.39, 7.1 p.40; National Instrument 55-104, s.3.3 p.9. Insider filings are searchable at SEDI.
Three practical notes follow from that table.
A venture issuer files less, later. A company on the TSX Venture Exchange gets 120 days instead of 90 for its annual statements, 60 instead of 45 for a quarter, and does not have to file an Annual Information Form at all. If you are researching a small company and cannot find its risk factors, that is usually why. Less disclosure is itself information about what you are buying, and the difference between the two exchanges runs deeper than the filing calendar. Our guide to how the stock market works covers what each exchange requires of a company before it can list there.
The MD&A is the document people skip and should not. The rule that governs its contents describes it as “a narrative explanation, through the eyes of management, of how your company performed”, and instructs management to give a balanced discussion “openly reporting bad news as well as good news”. The same form tells preparers to “Explain the nature of, and reasons for, changes in your company’s performance. Do not simply disclose the amount of change in a financial statement item from period to period. Avoid using boilerplate language.” When an MD&A reads like boilerplate anyway, that is worth noticing, because the form specifically forbids it. Source: Form 51-102F1 Management’s Discussion & Analysis, Part 1, p.3.
The MD&A also contains a free eight-quarter history. Item 1.5 of the same form requires a summary table of total revenue and profit, in total and per share, for “each of the eight most recently completed quarters”, with a discussion of what caused the variation and of the seasonality of the business (p.8). Two years of quarterly data and an explanation of the seasonal pattern, in one table, in a document most readers never open.
The seven questions, in order
Research is not reading everything. It is asking a fixed set of questions in an order where each one can end the process, so that you stop early when the answer is already clear.
| # | The question | Where the answer is | When to stop |
|---|---|---|---|
| 1 | What does this business actually do, and where does the money come from? | AIF, segment note, results release | You cannot describe it in two sentences without using the company’s own marketing words |
| 2 | Is it growing, and is the growth real? | Segment table, five years of annual releases | The growth is an acquisition, a 53rd week or an accounting change rather than the business |
| 3 | Does the growth reach the bottom line? | Gross and operating margin, by segment | Revenue rises and margins fall every year with no explanation you find convincing |
| 4 | Does it reach you? | Diluted weighted-average share count, five years | The share count rises as fast as profit does |
| 5 | What does it do with the money it keeps? | Capital expenditure, buybacks, dividends declared | Cash goes out and nothing you can point to comes back |
| 6 | What would you be paying? | Trailing earnings from the filings, against a price | The multiple only works if a growth rate continues that nothing in questions 2 to 5 supports |
| 7 | What would prove you wrong? | AIF risk factors, guidance, the MD&A | You cannot name a single figure whose arrival would change your mind |
The order matters. Question 6 is where most people start, because a price and a multiple are easy to find. Starting there is how you end up owning a business you cannot describe.
Question 1: What does it do, and what does the revenue line leave out
Dollarama is easy to describe, which makes it a good place to see the trap. It runs discount stores in Canada. At the end of the second quarter of fiscal 2027, on August 2, 2026, there were 1,734 of them.
That is not the whole company. The results release reports three distinct things:
| Part of the business | Size at the second quarter of fiscal 2027 | How it appears in the financial statements |
|---|---|---|
| Canadian stores | 1,734 stores, $1,841.9 million of quarterly sales | Consolidated line by line |
| Dollarama Australia | 414 stores, $184.8 million of quarterly sales, a $17.1 million operating loss | Consolidated line by line since July 21, 2025 |
| Dollarcity (Latin America) | 781 stores across Colombia, Guatemala, Peru, El Salvador and Mexico, counted at June 30, 2026 | Not in revenue at all. One line: the share of its net earnings |
That third row is the point. Dollarcity’s sales appear nowhere in Dollarama’s revenue. It is an equity-accounted investment, which means Dollarama reports only its share of the profit, as a single line near the bottom of the income statement. In the second quarter of fiscal 2027 that line was $49.9 million against consolidated net earnings of $349.3 million. Fourteen per cent of the profit came from a business contributing zero dollars of reported revenue.
It has been growing that way for years. Dollarama’s share of Dollarcity’s net earnings ran $33.2 million, $45.4 million, $75.3 million, $129.9 million and $191.5 million across fiscal 2022 to fiscal 2026, which is 5.0% of consolidated net earnings rising to 14.6%. Dollarcity’s own store count went from 440 at the end of fiscal 2023 to 732 at the end of fiscal 2026.
There is a further detail in that row, and it is exactly the kind that separates careful reading from quick reading. Dollarcity reports on a calendar-quarter lag. The $49.9 million above is Dollarama’s share of Dollarcity’s earnings for April 1 to June 30, 2026, not for Dollarama’s own quarter ending August 2, and the 781 stores are counted at the earlier date. The company states this. A model that lines the two periods up without noticing is out by a quarter.
Anyone who researched Dollarama by dividing profit by revenue and comparing the result to other retailers was measuring a margin that includes the earnings of a business whose sales are not in the denominator. The number is not wrong. It just does not mean what it appears to mean, and one sentence in the segment note is enough to see it.
The general lesson is worth keeping: an income statement shows you the businesses a company consolidates, not the businesses it owns. Joint ventures, associates and minority stakes turn up as a single line, and finding them is question 1, not an afterthought.
Question 2: Is it growing, and is the growth real
Growth is the easiest thing to measure and the easiest thing to measure wrong. There are two traps in Dollarama’s numbers, and both of them are common enough across Canadian companies to be worth learning as patterns.
The acquisition inside the comparison
Dollarama acquired its Australian business on July 21, 2025. The comparison quarter for the second quarter of fiscal 2027 therefore contains thirteen days of Australia, and the current quarter contains all of it.
Consolidated sales went from $1,723.8 million to $2,026.6 million, which is the 17.6% the company reports. Strip Australia out of both sides and the Canadian business went from $1,698.1 million to $1,841.9 million, which is 8.5%. The headline overstates the growth rate of the underlying business by 9.1 percentage points.
Both numbers are in the same release. The segment table is not hidden, it is simply further down than most people read.
The Canadian 8.5% also decomposes into two pieces the company states directly. Comparable store sales, meaning sales at stores open in both periods, grew 5.4%, made up of 3.7% more transactions and a 1.7% larger average basket. The store count grew 4.1%, from 1,665 to 1,734. Those are different kinds of growth: one is customers choosing the stores more, the other is building more stores, and the second one has a natural ceiling that the first does not.
The 53rd week
Dollarama’s fiscal year ends on the Sunday nearest January 31. A fiscal year defined by a weekday rather than by a date is common among retailers and grocers, and because 52 weeks is only 364 days, the calendar drifts until a 53-week year is needed to correct it, which works out at roughly one year in six.
Fiscal 2025 was one of those years. Here is what that does to a five-year growth series:
| Fiscal year | Weeks | Reported revenue growth | Growth per trading week | Difference |
|---|---|---|---|---|
| 2023 | 52 | +16.67% | +16.67% | 0.00pp |
| 2024 | 52 | +16.12% | +16.12% | 0.00pp |
| 2025 | 53 | +9.30% | +7.24% | +2.06pp |
| 2026 | 52 | +13.14% | +15.31% | -2.18pp |
Source: Dollarama fourth-quarter and full-year results releases for fiscal 2023, 2024, 2025 and 2026, Selected Consolidated Financial Information; week counts as stated by the company.
The extra week flatters fiscal 2025 by about two points and then penalises fiscal 2026 by about two more, because fiscal 2026 is compared against an inflated base. The two years swing 4.2 points between them, and the headline points the wrong way in both. On the reported numbers, growth decelerated from 9.3% to a recovery at 13.1%. On a per-week basis, growth slowed to 7.2% and then accelerated to 15.3%.
Neither story is a fabrication. The first is just the arithmetic of a calendar. Any time you see a retailer’s growth rate move sharply in a year with no obvious cause, count the weeks. The company states the count itself.
Question 3: Does the growth reach the bottom line
Growing sales is not the same as growing profit. The bridge between them is margin, and margin is where segment reporting matters most.
| Second quarter of fiscal 2027 | Consolidated | Canadian segment | Australian segment |
|---|---|---|---|
| Gross margin, prior year | 45.5% | 45.6% | 37.1% |
| Gross margin, this year | 44.5% | 45.7% | 32.4% |
| Change | -1.0pp | +0.1pp | -4.7pp |
| SG&A as a share of sales, prior year | 14.0% | 13.8% | 25.3% |
| SG&A as a share of sales, this year | 15.1% | 13.8% | 28.1% |
| Operating result | $517.3 million | $534.5 million | -$17.1 million |
Source: Dollarama Inc., second-quarter fiscal 2027 results release, September 16, 2026, Selected Consolidated and Selected Segmented Financial Information.
Read the top three rows again. Consolidated gross margin fell a full percentage point. The Canadian business, which is 91% of sales, saw its gross margin rise. The consolidated line and the business it mostly represents moved in opposite directions in the same quarter.
Nothing deteriorated. A lower-margin business was added to the mix, and mixing a 32.4% margin into a 45.7% margin lowers the average. That is arithmetic, not performance. The same applies to the SG&A row: Canadian SG&A was 13.8% of sales in both periods, unchanged, while the consolidated figure rose 1.1 points.
The bottom row is the one most worth sitting with. Consolidated operating income of $517.3 million is less than the Canadian segment produced on its own, because the Australian segment lost $17.1 million at the operating line. A reader looking only at the consolidated statement sees a company whose operating income grew 7.0%. A reader who opens the segment table sees a Canadian business that grew its operating income 10.5% and an Australian project currently costing money.
Which of those is the right way to look at it depends on what you believe about Australia, and that belief is now an explicit part of your research rather than something hidden inside an average. That is the whole purpose of the step.
Question 4: Does the growth reach you
A company can grow revenue, grow profit, and deliver nothing to the person holding the share, if it issues more shares along the way. It can also deliver considerably more than it grew, if it retires them.
The relevant figure is the diluted weighted-average share count, and it is in every quarterly and annual results release. If the idea of different share counts is new, our guide to what a stock is and what you actually own sets out the three of them and why the diluted one is the one to use.
Dollarama’s earnings per share went from $2.18 in fiscal 2022 to $4.73 in fiscal 2026, which is 21.4% a year. Revenue over the same period grew 13.8% a year. The gap between those two numbers is the answer to question 4, and it breaks into exactly three pieces.
– Revenue rose from $4,330.8 million to $7,255.8 million, a factor of 1.68. – Net margin widened from 15.31% to 18.05%, a factor of 1.18. – The share count fell from 304.416 million to 276.684 million, down 9.1%, which multiplies earnings per share by 1.10.
Multiply those three together and you get 2.17. Reported earnings per share went from $2.18 to $4.73, a factor of 2.17. The decomposition closes, which is the check that tells you it is a real accounting identity rather than an estimate. In order-independent terms, revenue produced 67% of the growth, margin 21% and the shrinking share count 12%.
The reason this matters more than it looks: those three sources have completely different durabilities. Revenue growth depends on customers and store openings. Margin expansion has a ceiling, because no retailer’s margin rises forever. The share count effect depends on management continuing to spend money on buybacks, which is a choice they can reverse next quarter. A company delivering 21% per-share growth where most of it comes from the third source is a different proposition from one where most comes from the first, and no headline growth rate distinguishes them.
This is the measure behind our ranking of Canadian growth stocks, which compares ten TSX companies on per-share growth rather than revenue growth and finds the order changes substantially. The arithmetic of why small annual differences turn into large ones is in our guide to compounding.
Question 5: What it does with the money it keeps
Once you know how much a company earns, the next question is what happens to it. There are four destinations: reinvest in the business, buy back shares, pay dividends, or pay down debt. The mix tells you what management believes.
| Fiscal year | Net earnings | Share buybacks | Dividends (derived) | Capital expenditure |
|---|---|---|---|---|
| 2022 | $663.2m | not stated | $61.2m | $159.5m |
| 2023 | $801.9m | $689.0m | $64.4m | $156.8m |
| 2024 | $1,010.5m | $655.9m | $80.5m | $278.8m |
| 2025 | $1,168.5m | $1,068.2m | $103.3m | $243.4m |
| 2026 | $1,309.4m | $834.2m | $117.1m | $272.8m |
Across fiscal 2023 to fiscal 2026, Dollarama spent $3,247.3 million buying back its own shares and the equivalent of $365.3 million on dividends. It returned almost nine dollars through buybacks for every dollar of dividend, a ratio of 8.9 to 1. In the second quarter of fiscal 2027 alone it repurchased 1,596,016 shares for $300.4 million, at an average price of $188.23, which is 86.0% of that quarter’s net earnings.
This is the single most common way a screen misleads a Canadian investor. A stock screener sorted by dividend yield shows Dollarama at roughly a quarter of one per cent and drops it near the bottom of any income list. The screener is reporting the shortest bars on that chart and nothing else. Whether you think the buyback was a good use of the money is a separate question, and a fair one, since buying back stock at a high multiple destroys value as surely as buying it cheaply creates it. But a reader who never saw the buyback never got to ask.
The fourth destination is debt, and it is the one the release says least about. Dollarama’s net financing costs rose from $43.2 million to $51.2 million in the quarter, an increase of $8.0 million that the company attributes primarily to higher average Canadian debt after two fixed-rate note issues in the previous quarter, at 3.940% and 4.576%, plus $2.9 million from Australia. That is the other side of a buyback programme running at 86% of quarterly earnings: when shareholder returns exceed what the business generates, the difference is borrowed.
Notice what the release does not give you. It states financing costs and it states the coupons, but it does not state net debt by fiscal year, so you cannot build a leverage trend from results releases alone. That figure is in the balance sheet inside the interim and annual financial statements, which are a separate filing. Knowing which document a missing number lives in is as much a part of research as reading the ones you have.
There is a tax dimension too, and it runs in the buyback’s favour for most Canadians. A dividend is taxable in the year you receive it, in a non-registered account, whether or not you wanted the cash. A buyback returns value by raising your proportional ownership, and nothing is taxed until you sell. Our guide to how investment income is taxed in Canada works the gross-up and the dividend tax credit through to the dollar, and our dividend income calculator shows what a stated yield actually pays after tax in each account type.
For companies where the dividend genuinely is the point, the coverage and payout tests matter far more than the yield does, and those sit on our Canadian dividend stocks page.
Question 6: What you would be paying
Only now does price enter. And there is a trick here that almost nobody teaches: you can build a company’s trailing price-to-earnings multiple without using any data source except the company itself.
Trailing earnings first. The most recent full fiscal year plus the most recent interim period minus the same interim period a year earlier gives you twelve months of earnings ending at the latest quarter. All three figures are in the company’s own releases.
| Step | Figure | Where it comes from |
|---|---|---|
| Fiscal 2026 diluted EPS | $4.73 | Fourth-quarter and full-year fiscal 2026 release |
| Plus first half of fiscal 2027 | $2.39 | Second-quarter fiscal 2027 release, 26-week period |
| Less first half of fiscal 2026 | $2.14 | Same release, comparative 26-week period |
| Trailing twelve-month diluted EPS | $4.98 | Computed |
This is the standard construction and it is very slightly approximate, because each period’s per-share figure uses its own weighted-average share count. On a company retiring stock steadily, the effect is in the second decimal place.
Now a price. The usual source is a quote screen, and that is fine. But in this case the company disclosed one: under its normal course issuer bid, Dollarama repurchased shares during the quarter at a weighted-average price of $188.23, stated in the release and excluding the tax on share repurchases. That is a price at which the company itself chose to buy its own stock in size.
At $188.23 against $4.98 of trailing earnings, the multiple is 37.8 times, and the earnings yield, which is the same number upside down, is 2.65%. The declared quarterly dividend of $0.12 annualises to $0.48, a yield of 0.26% against that price and a payout of 9.6% of trailing earnings.
Those four numbers are the entire valuation picture, and not one of them came from an aggregator. If you want the version that appears on a broker’s quote screen and how it differs, our guide to reading a stock quote takes the fields apart, including why a trailing multiple and a forward multiple can be twenty per cent apart on the same stock on the same day.
What does 37.8 times mean? By itself, nothing. A multiple is not high or low, it is a statement about expectations, and it has to be read against the answers to questions 2 through 5. An earnings yield of 2.65% on a business growing earnings per share at 21% a year is a different proposition from the same yield on a business growing at 3%, and the honest conclusion of question 6 is usually a sentence of the form “this price makes sense if X continues, and X is the thing I now need an opinion about.”
Note too that the market’s own expectation is a separate number again. Street consensus for that quarter was $1.2556 of earnings per share from nine analysts, and Dollarama reported $1.29, so the result beat the estimate by 2.7%. Consensus is street data rather than company data, and it is worth keeping the two clearly apart. Why a beat does not reliably move a stock upward is the subject of our guide to what moves a stock price, which works through a week when every big Canadian bank beat expectations and the shares went in four different directions.
Question 7: What would prove you wrong
Research that cannot be falsified is not research. The last step is naming the figures that would change your mind, and the filings help more than you would expect.
The risk factors are ranked, by rule. Form 51-102F2, which sets out what an Annual Information Form must contain, requires a company to disclose risk factors and then instructs: “Disclose the risks in order of seriousness from the most serious to the least serious”, and adds that “A risk factor must not be de-emphasized by including excessive caveats or conditions”. Source: Form 51-102F2 Annual Information Form, Item 5.2 and Instructions, p.10.
That single instruction changes how you read the section. The first three risks in a Canadian AIF are the ones management considers most serious, in their own ranking, written by people who know the business and reviewed by lawyers who did not want to be sued for understating them. Most readers skim risk factors as boilerplate. The ordering is the opposite of boilerplate.
Guidance tells you what management will and will not commit to. Dollarama’s outlook for fiscal 2027 covers comparable store sales, gross margin, SG&A, capital expenditure and store openings, and it is explicitly for the Canadian segment only. The company states that it gives no consolidated revenue, earnings-per-share or net earnings guidance, and none for Australia or Dollarcity. That is a deliberate choice and it is informative: the part of the business management is prepared to be measured on is the Canadian part.
Insider filings are a five-day-old record, not a rumour. Under National Instrument 55-104 a reporting insider must file within five days of any change in their holdings, and the filings are public on SEDI. What directors and officers actually did with their own money, at what price and on what date, is free and current.
Dollarama’s guidance is also specific enough to be tested. On September 16, 2026 it raised its fiscal 2027 Canadian comparable store sales range to 4.0% to 4.5%, from the 3.0% to 4.0% issued the previous March, and raised net new store openings to a range of 65 to 75. It left gross margin of 45.0% to 45.5%, SG&A of 14.1% to 14.6% and capital expenditure of $420.0 million to $470.0 million unchanged. For Australia it stated the opposite of a target: the company “continues to expect a net loss for the Australian segment in fiscal 2027”.
Notice what that last one does to a falsification test. An Australian operating profit is not the near-term test, because management has already said there will not be one. The test is whether the loss narrows.
So a short and testable list for this company would look like: Canadian comparable store sales staying inside or above the 4.0% to 4.5% range, the Canadian gross margin holding near the top of the 45.0% to 45.5% guided band, the Australian net loss shrinking rather than widening, the diluted share count continuing to fall, and the Dollarcity earnings line continuing to compound. Every one of those appears in the next quarterly release, on a known date. If three of them go the wrong way, the case that supported a 37.8 times multiple has weakened, and you will know it from the same document you started with.
The mistakes that actually cost people money
Four of them, all visible in the work above.
Extrapolating a headline growth rate. Projecting 17.6% forward when the underlying business grew 8.5% builds a model on a number that describes an acquisition, not a trend. The fix costs one minute: find the segment table and subtract.
Reading a consolidated margin as a performance signal. Dollarama’s consolidated gross margin fell a point in a quarter when the Canadian business expanded its margin. Mix moves averages. Always ask whether a margin changed because the business changed or because what is inside the average changed.
Screening on yield and never looking further. A screener ranks Dollarama near the bottom on income at a 0.26% dividend yield, while the company was returning nine times as much through buybacks and spending 86% of a quarter’s earnings doing it. A screen is a filter, not a conclusion. Anything a screen eliminates, it eliminated on one variable.
Comparing across a 53-week year. Two of the last four fiscal years carried a growth rate that pointed the wrong way, by about two points each, purely because of the calendar. Retailers, restaurant groups and grocers commonly use a 52-or-53 week year. Count the weeks before you compare.
There is a fifth that belongs here even though it is not visible in the numbers: doing all of this work and then losing part of the result to the mechanics of buying. Commissions, the bid-ask spread and currency conversion are a real cost on a real position, and they are fully knowable in advance. Our guide to what trading actually costs in Canada prices each of them from the brokers’ own published schedules, and how to buy your first stock covers the order types that decide what you actually pay on the day.
The checklist
Everything above compresses to one page. This works on any company filing on SEDAR+.
Before you open anything
1. Write down, in two sentences and without the company’s marketing words, what the business sells and who pays for it. If you cannot, stop.
From the most recent results release
2. Find the segment table. Note every segment and its sales, margin and operating result. Note anything equity-accounted, which appears as a single line and whose revenue is not in revenue. 3. Recompute the growth rate excluding anything acquired inside the comparison period. 4. Compare segment margins against the consolidated margin. Note any that move in opposite directions. 5. Note the diluted weighted-average share count and compare it to the same quarter a year earlier.
From four or five years of annual releases
6. Build a table: revenue, net earnings, diluted earnings per share, gross and operating margin, diluted share count, dividends declared per share, capital expenditure. Check the week count of each fiscal year. 7. Split earnings-per-share growth into revenue, margin and share count. Ask which of the three you expect to continue. 8. Add up buybacks and dividends. Compare the total to net earnings and to capital expenditure.
From the MD&A and the AIF
9. Read Item 1.5 of the MD&A for the eight-quarter table and the seasonality discussion. 10. Read the first three risk factors in the AIF, which are ranked by management’s own assessment of seriousness. 11. Note exactly what guidance covers and what it does not.
Then, and only then
12. Build trailing earnings per share: last full year plus the latest interim period minus the same period a year earlier. 13. Divide a price by it. Invert it to get the earnings yield. 14. Write down the three figures in the next quarterly release that would tell you the case is breaking.
Steps 1 through 11 are the research. Steps 12 and 13 take two minutes. Step 14 is the one people skip and the one that saves money.
Where to go from here
This guide is the frame, and each question inside it opens into a subject of its own. Three are worth taking up next.
The first is the financial statements themselves. Everything above was worked from results releases and segment tables, which is deliberate, because they are short and they contain more than most people realise. But an income statement, a balance sheet and a cash flow statement connect to each other in specific ways, and a company can report rising profit while cash goes the other direction. That connection is where earnings quality lives.
The second is what to do once you own it. A position needs its cost base tracked from the first purchase, not reconstructed later, and reinvested dividends are the most common reason a Canadian pays tax twice on the same money. Our guide to adjusted cost base works nine quarters of a real reinvestment plan through to the dollar.
The third is the American version of the same process, which most Canadians need at some point. The questions are identical, the documents are different: filings go to EDGAR rather than SEDAR+, the annual report is a 10-K and the quarterly a 10-Q, and the risk-ordering rule quoted above does not apply. If you end up buying the exposure in Canadian dollars instead, our guide to Canadian Depositary Receipts measures what that convenience costs per year.
One closing note on the company used throughout. Our own write-up of the quarter used here covers what the market made of it on the day, which is a different question from the one this guide asks. Dollarama was chosen because its filings illustrate four separate traps cleanly, not because of any view on the shares. The figures here describe a quarter reported on September 16, 2026 and four fiscal years before it. Nothing on this page is a recommendation, and a process that ends in a verdict rather than in a list of testable beliefs has not been followed properly.
Common questions
Do I need to pay for data to research a Canadian stock? No. Every figure in this guide came from documents a company is legally required to file and publish for free, on SEDAR+ and on its own investor relations page. Paid services save time and organise the data, and none of them contain a number the filings do not.
How far back should I look? Four to five fiscal years is enough to see whether margins and share counts trend, and short enough that the business is still recognisably the same company. Going back ten years usually means comparing two different businesses under one name.
Which document should I read first if I only read one? The MD&A. It carries management’s own explanation of the numbers, an eight-quarter summary table required by Item 1.5 of Form 51-102F1, and a discussion of seasonality. The results release is faster, but the MD&A is where the reasons are.
What is the difference between a company’s guidance and analyst consensus? Guidance is what the company itself commits to, published in its own release. Consensus is the average of analysts’ forecasts, compiled by third parties, and it is what the share price is usually measured against on results day. Dollarama’s fiscal 2027 guidance covers its Canadian segment only and includes no earnings-per-share figure at all, which is itself worth knowing.
Why use the diluted share count rather than shares outstanding? Diluted includes shares that would exist if outstanding options and similar instruments were exercised, so it reflects the claim on earnings that is already committed rather than only the one already issued. It is the count earnings per share is reported on, and the one that makes a five-year comparison honest.
Does this process work for a TSX Venture company? The questions do, but there is less to read. A venture issuer gets 120 days instead of 90 to file annual statements, 60 instead of 45 for a quarter, and is not required to file an Annual Information Form, which means no ranked risk factors. Less disclosure is part of what you are buying.
Disclaimer: The content on bestcanadianstocks.ca is for informational and entertainment purposes only and does not constitute financial advice. Past performance is not indicative of future results. Always consult a qualified financial advisor before making investment decisions.