The headline wants seven bargains that will also grow. The tape does not have them, not as a set you can buy from a list. On October 8, 2026, spot gold was about $4,131. A compilation of company-reported second-quarter all-in sustaining costs put the median near $1,926 an ounce. Subtract, and the typical producer in that set still had an implied margin above $2,200 against the morning price. Implied margin is not cash in the bank. It is a sketch. The sketch says the business of digging gold is not distressed. A business that is not distressed is a hard thing to call cheap just because the share fell.
That is the idea to hold. Undervalued is a word about the share. A low cost is a word about the mine. High growth is a word about ounces that are not yet being sold. The three do not travel together unless a filing forces them to. At $4,130, almost every serious gold miner still clears its cost by a wide gap. The gap is why earnings have been large. It is also why a list of "cheap" gold stocks, written after a pullback from a January high near $5,400 to $5,600, is usually a list of shares that fell with the metal. Falling with the metal is not a discount. It is the gear.
The seven names below are Canadian gold mining companies, or Canadian-listed ones, chosen because their latest reported costs span the range from very low to merely fine. They are not a ranking. They are not gold stocks to buy. They are seven ways to ask whether the cash, the cost, and the unbuilt ounce can sit in one story without the word undervalued doing the work. This is not a recommendation to buy or sell any security.
What the word would have to mean
People ask what investors should look for in an undervalued gold stock. Start with what they should not look for. They should not look for a small share price. A dollar share can be expensive if the company will have to issue more shares to finish a mill. They should not look for the stock that fell the most in October. The stock that fell the most may be the one whose cost is rising, or whose ounces were cut. They should not look for a headline that says high growth, if the growth is a drill result that is not a reserve.
Look for four things, and make each one survive a gold price below the one on the screen. First, a cost you can read in the company's own unit, and that still clears at a price you are willing to live with for a year. Second, cash or a funded plan, so the next ounce does not depend on a hopeful issue of stock. Third, ounces that are in a reserve or a mine plan, not only in a slide. Fourth, a share price you can explain without using the January high as the fair value. If you need $5,500 gold to make the shares look smart, you do not own a bargain. You own a rebound.
All-in sustaining cost is the number people use for the first test, and it is a constructed number. Each company defines it. Some state it per ounce produced. Some use gold-equivalent ounces and fold other metals in. Some give themselves by-product credits. A table that lines them up is a map, not a law. Use it to see direction and distance from the gold price. Do not use a one-spot difference between two companies as a verdict that one is the better stock.
Why growth is the dangerous half of the headline
People also ask why undervalued gold stocks could have high growth potential. The flattering answer is operating leverage. If the cost stays near $1,500 and the gold price rises, almost every extra dollar of price falls through to cash, because the workforce and the fleet were already paid for. That is real. It is how gold mining stocks outrun the metal in a rising market. It is also how they give it back faster. Leverage is not a growth rate. It is a gear. A gear does not care which way the price goes.
The less flattering answer is the unbuilt ounce. A company can grow production by building a pushback, a decline, or a second mine. That growth spends the cash the high gold price just created. If the new ounce arrives at a low cost, the growth is a business. If it arrives only because the study used $4,500 or $5,000 gold, the growth is a bet that the spike was normal. After 2011, aggressive building and deals erased on the order of $129 billion of shareholder value in the following years, a figure Sprott's October 2026 note attributed to McKinsey. The industry remembers. A reader who treats "high growth" as a gift is forgetting the bill.
So growth potential is high only in a narrow sense. The company can add ounces you can count, at a cost that still clears if gold is $3,500, without issuing so many shares that you own less of the mine than you thought. Anything else is a story about the gold price, wearing a company name. Stories about the gold price can be right. They are not fundamentals. Gold stocks with strong fundamentals are the ones where the cost, the cash, and the plan still agree when the price is boring.
The margin that makes every list look easy
Here is the sector fact that should slow the headline down. Against a spot price near $4,131 on October 8, a median reported all-in cost near $1,926 still leaves a gap above $2,200 an ounce. In the second quarter, when gold averaged about $4,512, the gap at many companies was wider still. Sprott's Kenny Zhu, writing on October 7, said forward earnings expectations for the big-miner index had risen nearly fourfold from the end of 2023, while the forward valuation multiple had compressed toward 5.7 times, from a peak near 9.9 times in 2014. Record earnings. A multiple that did not follow. That is the closest thing the public record has to a sector-wide "undervalued" claim.
It is a sector claim. It is not seven stocks. A low multiple can mean the market is wrong. It can also mean the market expects the cash to shrink. John Hathaway, at the Precious Metals Summit, said the miners are more profitable than at any time he has seen since 1998. He also said that if gold marks time in the mid-$4,000s for two or three years, margins get squeezed, because costs inflate. He put that pressure near 8 to 10 percent, from capital spending and labor, not from a central bank target. A multiple that looks cheap against last quarter's cash can look fair against that calendar. Cheap against a boom is the normal price of a boom, one year later.
Gold in early October is already off about a quarter from the late-January high. The 10-year Treasury yield has been near 5.3 percent. Gold pays no coupon. The shares are claims on future cash, and a higher yield makes that future cash worth less today. You can call the shares undervalued only after you have done that discount, not before. The list that skips the yield is a list of moods.
Seven costs, not seven bargains
The figures in this section are second-quarter 2026 all-in costs and ounces as the companies reported them, lined up in an early-October compilation. They are history. They are not a forecast, and they are not a score you should buy. Where a later guidance change exists, it is labeled as later. Canadian gold stocks, in this set, means the company is Canadian or lists in Canada. It does not mean the rocks are in Canada. The rocks are wherever the mine is. Jurisdiction is part of the cost even when it is not in the all-in number.
Lundin Gold reported an all-in cost near $1,176 and about 119,000 ounces in the quarter. On that compilation it sat near the bottom of the cost list, among the cheapest rocks being mined by a company of its type. At $4,130, a cost near $1,176 is a very wide gap. Wide is not the same as undervalued. A low-cost mine in Ecuador can be a wonderful business and a fully priced share, or a wonderful business and a share the market refuses because the country risk is the thing the cost table leaves out. The thing to watch is whether the cost stays near the bottom while the company decides what to do with the cash. A low cost that gets spent on a worse ounce is how a good mine funds a bad idea. Growth, here, is only growth if the next ounce looks like Fruta del Norte's cost, not like a slide.
K92 Mining reported about $1,376 and about 43,000 ounces. It is a smaller Canadian name with a mine in Papua New Guinea. The cost is low. The ounce count is not Agnico's. Small and low-cost is where people go looking for high-growth gold stocks, because a new ounce is a large percent of a small base. Percentages flatter small bases. A doubling of 43,000 ounces is still not a major, and it still has to be built, permitted, and kept safe in a place that is not Toronto. The growth potential is the gap between a good cost and a modest size. The risk is that the gap is also where a single operational miss becomes the whole quarter. Watch the cost. Watch the tonnes. Do not watch the percentage alone.
Agnico Eagle reported about $1,459 and about 856,000 ounces. This is what scale looks like when the cost is also low. The cost had ticked down from the prior quarter, near $1,483, while several peers paid more. At a $4,512 average gold price, the implied margin was about $3,053. At $4,130, if the cost holds, the gap is still near $2,670. That is the fundamental. It is not a coupon that says the shares are cheap. Agnico is large enough that "high growth" cannot mean a double in ounces without a decade of mines. Growth at this size is a few percent of a very large book, plus the gear if the gold price rises. People who want a penny stock's upside and a major's balance sheet are asking for two contracts. Agnico is the second contract. It is the one that still works if gold is boring, which is the only growth that deserves the word durable.
G Mining Ventures reported about $1,690. It is a younger producer, the kind of Canadian gold stock that sits between a build and a habit. A first mine that arrives near $1,700 all-in, in a world of $4,130 gold, is a business. It is also a company whose growth potential is the second act, not the first. The first act is proving the cost can be repeated. Junior gold stories often skip that proof and go straight to the next project. A producer that just learned how to pour gold should be watched for whether the cost stays put for four quarters, not for whether the map has more dots. Dots are not cash. Four quarters of a stable cost are closer to a fundamental than a resource estimate that has not met a mill.
Wesdome reported about $1,763 and about 44,000 ounces. The mines are in Canada, which is the thing the phrase "Canadian gold mining stocks" sometimes pretends is true of every ticker with a Toronto listing. Domestic rock does not make a cost low, and it does not make a share cheap. It does remove one kind of surprise. You are not underwriting a new mining code every election in quite the same way. You are underwriting grade, depth, and a small ounce base. High growth, at 44,000 ounces a quarter, is a loud percentage and a quiet number of ounces. Look at the ounces. A Canadian address is a fact about politics and infrastructure. It is not a valuation.
Kinross reported a second-quarter all-in cost near $1,821 on a gold-equivalent basis, and about 492,000 ounces. That quarter still left a wide gap against $4,500 gold. Then the story changed shape. A late-September account of the company's own revisions said 2026 and 2027 production guidance had been cut by about 7 to 8 percent, to roughly 1.84 to 1.86 million ounces a year, and that the cost band had been lifted by about 9 percent, toward $1,850 to $1,900. Net cash was put near $1.9 billion, with a target of returning about half of free cash flow. Read both halves. The balance sheet is what people point at when they say strong fundamentals. The cut is what people skip when they say undervalued, because the share fell and the fall looks like a sale. A sale of a smaller future is not the same as a sale of the same future. Kinross is the name that stops this list from being a cheer. Growth potential that just got revised down is a sentence you should be able to say out loud.
OceanaGold reported about $2,151 in the quarter, on about 139,000 ounces, and its 2026 guidance has been a lower band, near $1,750 to $1,900. A company that guides below the year-to-date cost is telling you the second half has to be cheaper. That may happen. It has not happened yet in the print you can hold. The cash, as summarized from the company's mid-year figures, was about $655 million at June 30, with no debt. First-half free cash flow was about $385 million. The capital budget for the year has been described near $645 million, with growth spending set to rise, including work at Waihi North and at Haile. This is the cleanest picture in the seven of what "high growth" actually costs. The mine throws off cash. The company spends a large share of it on ounces you do not have yet. If those ounces arrive near the guided cost, the spending was the fundamental. If they arrive late, or at $2,151 forever, the cash pile was the growth story's fuel and also its risk. A chief executive's planned retirement in 2027 is a side note, not a thesis. The thesis is the gap between the cash and the budget.
What this list refuses to be
None of these seven is a TSXV gold stock. That absence is on purpose. The venture exchange is where the phrase "cheap gold stocks" goes to get smaller share prices and larger dreams. A junior with a drill hole and a treasury that lasts two quarters is not undervalued. It is unfunded, or it is funded only until the next raise. High growth potential, on that market, usually means the hole was good and the mill does not exist. The last owner of a real deposit often spent more proving it than the junior's entire market value. Sometimes that is an opportunity. More often it is a reminder that paper ounces are cheap because they are paper.
Canadian gold stocks that do belong in a fundamentals conversation are the ones that already sell metal, or that hold enough cash to finish a mill without a desperate raise. A developer with cash in the bank and a dated pour is a different object from a TSXV ticker at thirty cents. GoGold, to take a name that has been public about its own numbers, has described cash near $284 million against a Los Ricos South build whose living budget is above the old study, with a first pour aimed at 2028. That is a funded wait. It is not on this list of seven because it is not yet a producer with an all-in cost you can audit. It is the right comparison. Growth you can date and pay for is a plan. Growth you can only hope a buyer will fund is a pitch.
Best gold stocks, best Canadian gold stocks, undervalued Canadian gold stocks, gold stocks to watch. The search phrases all want the list to end in a winner. The costs above do not pick a winner. Lundin and K92 are low-cost and smaller. Agnico is low-cost and large. G Mining and Wesdome are in the middle of the cost list and early in the habit of being producers, or small in ounces. Kinross has cash and a worse guide. OceanaGold has cash, no debt, and a cost that has to fall to meet its own year. If you need one of them to be the undervalued one, you need a share price, a share count, and a cash-flow statement. This article does not pretend a cost rank is that work.
The growth that is just the gold price
Why could these shares have high growth potential. Because the gap between $1,200 or $1,800 of cost and $4,130 of price is so wide that a modest rise in the gold price, or a modest rise in ounces, produces a large rise in cash. That is true, and it is also true of a producer you did not put on a list. The median miner has the gap. The seven are not special for having it. They are useful because their costs are public and different.
The growth that is not special is the one that requires gold back at $5,500. From about $4,130 to the January closing high near $5,405 is roughly a 30 percent move in the metal. Miners would move more, if the old gear holds. That is upside in the shares. It is not a fundamental of the company. It is a forecast about the metal, and the metal is down because yields are up and the January spike was a spike. China's central bank bought about 21 tonnes of gold in September, which is a reason the metal has a customer. It is not a reason a miner's multiple must expand. Official buyers do not buy Kinross. They buy bars.
There is a version of growth that is inside the company and still easy to fake. Resource additions. A drill campaign. A study that uses a gold price above the spot price and then shows a net present value that looks like a gift. The gift is the price assumption. Change the assumption to $3,500, and many "high growth" projects become ordinary or worse. Gold stocks 2026 should be read at $4,130 and at a number below it. If the project only works at the high, it is not growth. It is an option on a rally, and options expire.
Strong fundamentals, said in one page
Gold stocks with strong fundamentals, if the phrase is going to be used at all, look like this. The mine is producing. The cost is falling, stable, or at least not walking up faster than the price. The company is net cash, or the debt is small next to the cash the mine throws off at a price under today's. The growth spending is a number you can find, and the ounces it buys are in a plan with a date. The share count is not the silent funding source. Management returns some cash and does not treat the treasury as a monument to the last scare.
Hathaway called large idle cash piles lazy. Zhu called the same piles proof the industry grew up. Both are tests you can apply here. Lundin's low cost will produce cash. What it does with the cash is the fundamental, not the cost itself. OceanaGold is already spending. That answers lazy, and it raises the other risk, which is spending on the wrong ounce. Kinross is promising to return about half of free cash flow while cutting the ounce guide. That is a mix, not a slogan. Agnico's falling cost is the rare print where the boom did not automatically raise the bill. One quarter is not a habit. A habit would be the same direction again.
Weak fundamentals are easier to spot once you stop calling them cheap. A cost that is above the company's own guidance. A production cut explained as temporary for the third time. A growth budget that uses a gold price the market has left. A treasury that covers less than a year of spending. A jurisdiction story that is doing more work in the presentation than the grade. None of that becomes a buy because the share is down in October. October was a few percent in the metal, inside a much larger retreat from January. The retreat is the context. The context is not a coupon.
How a share becomes undervalued, if it ever does
A share becomes undervalued when the cash you can defend is large next to the price of the company, and the defendable cash does not need a higher gold price to exist. Defendable means you have cut the January spike out of your model. You have used a cost that includes the inflation Hathaway described, not the cost from the best quarter. You have subtracted the growth budget, because cash that is already spoken for is not yours. You have noticed the yield, because a dollar in 2028 is worth less at a 5.3 percent 10-year than it was a year ago.
Do that, and some of these companies may still look inexpensive. This article will not tell you which. Doing it badly, with last quarter's peak margin and this week's fear, is how lists of seven get written. The list feels like work. The work was the sort. Lundin is not Kinross. Kinross is not OceanaGold. A table of costs is the start of the sort. A headline that calls all seven undervalued is the end of it, and it is the wrong end.
You can be early and still be wrong about the word. If gold falls toward $3,500 and costs rise 8 percent, the wide gap narrows and the shares will not wait for your model to agree. If gold returns to $5,000, the shares of the high-cost names can rise more than the shares of the low-cost names, because the gear is larger where the margin was thinner. In that world the "cheap" stock was the worse business, and it won the year. Winning the year is not the same as having been undervalued. It is having been a torque. Know which one you wanted.
A way to use the seven without a shopping cart
If you already own Canadian gold stocks, match each line to a question. Is the cost in the last print the cost in the guide. Is the cash net of the build. Did the ounce guide rise or fall. Is the growth a dated project or a district on a map. If the answers are worse than the share price implies, the October dip did not create a bargain. It created a reason to read. If the answers are better, you still have to decide whether the price of the shares already knows. Better is not cheap. Better is the input.
If you own nothing, the absence is not a hole this headline is obliged to fill. Gold stocks to watch means watch the next cost print and the next reserve report. It does not mean accumulate seven tickers before the week ends. The median miner is profitable at $4,131. Profit is the reason you do not have to hurry. Distress is what creates hurry, and distress is not this tape. The hurry in the headline is a search engine's hurry. You can decline it.
If you want a single rule, use this. Do not call a gold stock undervalued because the cost is low. Call it a low-cost mine, and then do the share-price work separately. Do not call it high growth because the presentation has a larger number next year. Call it a plan, and then see whether the plan spends cash you were counting as yours. The seven names survive as a curriculum. They do not survive as a basket. A basket would average Lundin's cost with OceanaGold's guidance gap and Kinross's cut, and the average would look like a view. It would be a blur.
The share count is the quiet cost
One more line belongs on the card, and it does not appear in an all-in cost. It is the share count. A company can grow ounces and still shrink what you own, if the growth is paid for with new shares. That is common on the venture exchange. It is less common at the seven names above, which is another reason they are producers and not penny stories. It is not impossible. A large acquisition paid in paper can do in a month what a year of operations spent a decade building. Hathaway said deals are close, because cash is burning a hole. Cash deals are one thing. Share deals are a tax on the holder who thought the stock was undervalued on yesterday's count.
Before you treat any of the seven as high growth, write down the number of shares. Then write down the ounces per share, not the ounces alone. A rise in ounces that loses to a rise in shares is not growth. It is a larger mine you own less of. Lundin's low cost does not protect you from a bad deal. Agnico's scale does not protect you from paying too much for someone else's reserve. Kinross's buyback, if it happens at a sensible price, does the opposite. It raises what each remaining share owns. Buybacks can also be a waste, if they are done at the top of a spike to satisfy a slogan about returning capital. The price of the buyback is part of the fundamental. The slogan is not.
This is the check the headline cannot run for you. Undervalued Canadian gold stocks, if any exist this week, are the ones where ounces per share are rising for a reason other than a higher gold price, and where the cost of those ounces still clears at a number you would not be embarrassed to use in a quiet year. That sentence is longer than a list. It is also the only version of the headline that survives contact with a filing. Seven tickers are a way to practice it. They are not a way to skip it.
The close
Seven undervalued gold stocks with high growth potential is a sentence that stacks three claims. The mine is better than the share price. The ounces will grow. You should care about both at once. At about $4,130 gold, the first claim is not proven by a low all-in cost. The cost proves the mine is alive, and for Lundin, K92, Agnico, G Mining, Wesdome, Kinross, and OceanaGold, alive is a fair description of the last reported quarter. Alive is not cheap. The second claim is not proven by a growth budget. OceanaGold's cash and its spending show the budget is real. Kinross's cut shows growth can be revised the other way. Agnico's size shows that durable growth is often small in percent and large in cash.
What should you look for. A cost that clears below today's price, cash that is not already promised to a mill, ounces in a plan, and a reason for the share price that does not require January to return. Why might any of them grow. Because a wide gap between cost and price turns a few more ounces, or a slightly higher gold price, into a lot of cash, and because some of these companies are spending to add those ounces. The cash is not yours until it survives the spend. The spend is not growth until the ounce arrives at a cost you can defend. Until then the headline is a wish. The filing is the potential.
A note on sources and limits
Spot gold near $4,131 on October 8, 2026, and a median company-reported all-in sustaining cost near $1,926, with an implied sector margin above $2,200, follow a same-day compilation of producer costs against the metal price. The company figures used here are from that compilation of second-quarter 2026 reports: Lundin Gold about $1,176 and 118,994 ounces, K92 about $1,376 and 42,931 ounces, Agnico Eagle about $1,459 and 855,816 ounces, G Mining about $1,690, Wesdome about $1,763 and 43,824 ounces, Kinross about $1,821 on a gold-equivalent basis and 492,326 ounces, OceanaGold about $2,151 and 138,800 ounces. All-in sustaining cost is a non-GAAP measure. Companies do not all define it the same way. Implied margin is price minus that cost. It is not free cash flow, not earnings, and not a valuation.
Kinross's later production cut of about 7 to 8 percent, toward 1.84 to 1.86 million ounces, the revised cost band near $1,850 to $1,900, net cash near $1.9 billion, and the free-cash-flow return target are from a late-September account of company guidance, not from the second-quarter cost print. OceanaGold's June 30 cash near $655 million, the absence of debt, first-half free cash flow near $385 million, a 2026 capital budget near $645 million, and 2026 all-in guidance near $1,750 to $1,900 are drawn from a September summary of the company's disclosures. Check the filing. Those figures move.
The January gold highs, a closing print near $5,405 and an intraday print near $5,589, and the 10-year yield near 5.3 percent, are the early-October 2026 market record used in prior notes. Sprott's October 7, 2026 piece by Kenny Zhu supplied the index earnings path and the 5.7 times forward multiple, and it cited McKinsey for the roughly $129 billion of value lost after the last boom. John Hathaway's comments on profitability since 1998 and on 8 to 10 percent cost pressure if gold marks time are from his September 2026 summit remarks, as published by Sprott. Kitco reported China's roughly 21-tonne September gold purchase. GoGold's cash and pour timing are the company's 2026 disclosures, used only as a contrast with producers.
This is not investment advice and not a solicitation to buy or sell any security. Costs rise. Guides get cut. Countries change. Gold can fall further or retrace the January loss. A low cost is not a cheap share. Readers should read the quarterly reports and speak with a licensed adviser before any decision.