Canaan's Bitcoin stash is half collateral as revenue misses and losses hit $97.6M

Here's the thing about holding Bitcoin on your balance sheet: it only matters if you can actually spend it. Canaan just learned that lesson the hard way.

On Sept. 8, the Bitcoin mining equipment maker reported Q2 revenue of $31.9 million. That's well short of the $35 million to $45 million range management promised back in May. And the next quarter looks even worse.

The timeline of a rough quarter

This story starts in May. Canaan told investors to expect big things. The $35 million to $45 million Q2 revenue forecast was optimistic. It turned out to be way off.

The actual number came in at $31.9 million. That's a miss of at least $3 million, and as much as $13 million if you anchored at the top of the range.

Product revenue did the heavy damage. It collapsed to $13.6 million in Q2, down from $42.9 million in Q1. That's a 68% plunge. Canaan blamed weaker demand, less computing power sold, and lower prices. Basically everything that could go wrong, went wrong.

Mining provided some cushion. The company produced 243 BTC and brought in $17.7 million from mining during the quarter. Management says those operations made a positive cash contribution before depreciation. That's a narrower measure than profitability for a reason. Because the overall picture isn't pretty.

The company's total net loss for Q2 hit $97.6 million. Brutal. That number includes $25.3 million in inventory and prepayment write-downs plus purchase-commitment provisions. Another $9.2 million came from property and equipment impairment.

Here's where it gets interesting. Despite that massive loss, cash actually went up. Canaan reported $66 million in cash at June 30, up from $43.5 million at the end of March. It's still below the $80.8 million held at the end of 2025, but the direction was positive.

How does a company lose $97.6 million and gain cash? Noncash charges. That's the difference between accounting losses and actual cash burn. Worth remembering the next time someone screams about a giant loss number without reading the footnotes.

The Bitcoin stash is smaller than it looks

Canaan held 1,915.5 BTC at June 30. That sounds like a solid treasury. It's not that simple.

More than half of those coins, 1,117 BTC to be exact, were pledged as collateral for secured term loans. Another 100 BTC sat in a fixed-term product. So at June 30, only 698.5 BTC was actually sitting in the company's liquid cryptocurrency assets.

In dollar terms: the pledged and fixed-term Bitcoin showed up as cryptocurrency receivables worth $70.9 million. The liquid crypto assets were valued at $47 million. Those are very different pools of money, and mixing them up would be a mistake.

And just like that, the whole "Bitcoin treasury" narrative gets more complicated. A headline BTC balance isn't a war chest when half of it's serving as loan collateral. Traders watching Canaan as a Bitcoin proxy should pay attention to that distinction.

This matters because Canaan is clearly under pressure. Q3 guidance came in at just $11 million to $15 million in revenue. Let that sink in. The company generated $17.7 million from mining alone in Q2. If Q3 revenue lands at the midpoint, the core business is barely running.

So what does management do? It sells crypto to buy back its own shares.

In late August, Canaan sold 3,952 ETH and 54 BTC for roughly $13.9 million. Part of those proceeds went to share repurchases. By Sept. 8, the buyback program had scooped up about 16.4 million American depositary shares for $7.4 million, including $5.4 million spent in late August.

Think about that move for a second. Product revenue is collapsing. Equipment demand is weak. And management's response is to buy back stock with crypto proceeds.

Is that confidence or desperation? Maybe both.

On one hand, buying back shares when the stock is beaten down can be smart capital allocation. The market's verdict on this company has been harsh, and repurchases at these levels could reward patient shareholders if the core business recovers.

On the other hand, buybacks don't fix a broken product pipeline. They don't manufacture demand for mining rigs that aren't selling. Canaan's core revenue engine is sputtering, and crypto sales are keeping the buyback machine running. That's not a sustainable long-term strategy. That's converting assets to support a stock price.

What comes next

The Q3 outlook is the number that matters most. $11 million to $15 million in revenue means the slide isn't over. If product revenue keeps shrinking, Canaan becomes more dependent on its mining operations and its crypto holdings.

Here's the risk nobody wants to talk about. Those 1,117 pledged BTC are someone else's take advantage of. If Bitcoin's price drops hard, Canaan could face margin calls or forced liquidation. That would turn a treasury story into a distress story fast.

The counterargument: Canaan has shown it can convert crypto into cash when needed. The late August sales prove that. The $66 million cash buffer also gives some runway even if Q3 comes in at the low end of guidance.

But runway only matters if there's a destination. Canaan needs its core business to stabilize. Otherwise it's just a mining operation with a side of share buybacks, slowly burning through its own balance sheet to keep the stock afloat.

And one more thing worth watching: Canaan's management is clearly willing to touch its crypto holdings now. That's a shift. For months, the narrative was about accumulating Bitcoin as a treasury asset. Now the narrative is about liquidating crypto to buy back shares. Those are opposite strategies, and the market has noticed.

My take? Spending $5.4 million on buybacks in late August while your revenue guidance is collapsing is a bold move. It signals management thinks the stock is undervalued. Or it signals they're trying to support the share price while they figure out a real plan. The market's verdict will come with the Q3 report, expected later this year.

And if that report looks anything like Q2, the difference between Canaan's pledged Bitcoin and its spendable cash will matter more than ever. Because at this rate, the company might need every liquid coin it can get.