Intel guidance crash: why honest forecasts still destroy value

November 4, 2026 · 9 min · Free, with full transcript

Intel beat its third quarter, then CFO Dave Zinsner guided to lower fourth-quarter revenue, roughly halved earnings per share and a gross margin about six points lower, and the stock fell nearly thirty percent. The hosts debate whether this was a communication failure, with consensus left sitting above Intel's range, or a market that priced the stock for perfection at more than fifty times estimated 2027 earnings. You come away able to explain how guidance gaps, selective disclosure rules, the AI-driven memory squeeze on PCs, the Kiryat Gat shutdown, supply constraints on wafers and substrates, the foundry strategy and the government stake shape how markets read a CFO's forecast, and what signals to watch next.

Listen to the podcast

8 min

Chapters

Key takeaways

  • The market punishes the gap between your forecast and its expectations, not the honesty of the forecast itself.
  • Use legal channels such as conference appearances or a more cautious tone on earlier calls to manage expectations, never private hints to selected analysts, which is selective disclosure.
  • Quantify known one-off hits, as Zinsner did with the Israel plant shutdown, so investors can separate them from the underlying business.
  • Distinguish supply-limited revenue from weak demand, but remember that if you cannot make the product a rival may take the order.
  • Watch whether a very low guidance bar is beaten easily next quarter, since that invites accusations of sandbagging.
Read the full transcript

Host:This is MBA Training. On the table: Intel's guidance crash and the price of being honest too late. Nearly thirty percent in one day. CNBC says it's Intel's worst day since the seventies.

Expert:And they beat the quarter. That's what makes every CFO put the sandwich down.

Host:Wait, what? They beat, so what exactly did people sell?

Expert:Next quarter. Guidance, the company's own forecast for the coming three months, came in below the quarter they'd just reported. Revenue down, profit per share roughly halved.

Host:So they won the match and then used the post-match interview to say they'll lose the next one.

Expert:Pretty much, and they were very precise about it.

Host:Who actually said it out loud?

Expert:Dave Zinsner, the CFO. Lip-Bu Tan, the CEO, told the story, but the numbers came from the finance chief.

Host:That's the bit of the job they leave off the recruitment poster.

Expert:The underlying story is odd. Demand for data center chips is so strong Intel can't make enough. The PC side is in trouble, because AI is eating the world's memory chips, memory prices jump, PCs get more expensive, and corporate IT budgets give out.

Host:So AI built the rally and AI is also mugging the PC business.

Expert:Same storm, two towns. Tan said on the call, according to TechCrunch, that customers started signalling a slowdown because of memory prices and availability.

Host:Funny how nobody remembered the PC business. Nobody has any memory.

Expert:They literally don't. It all went to the data centers.

Host:Fine. What else is in the forecast?

Expert:The grim part. In October there was an explosion at Intel's plant in Kiryat Gat, Israel. Reuters reported that people were killed. Zinsner told Reuters the shutdown takes a clear bite out of fourth-quarter revenue and margin, and he expects a similar hit next quarter.

Host:People died at that plant and an hour later we're all arguing about margin. That doesn't sit well with me.

Expert:Me neither. Tan called it a tragic accident on the call. Then the questions went straight to supply, because that's how these calls go. It's not pretty.

Host:Okay. Back to work. Margin.

Expert:Gross margin, meaning what's left of each sales dollar after the cost of making the chips, falls about six points from the third quarter to the fourth, according to Investing.com's read of the slides.

Host:Six points on a business that size is...

Expert:...basically a different company. Yes.

Host:Here's my view, and you'll hate it. This was a communication failure. Tan himself says customers were flagging the slowdown. Consensus, the average analyst forecast, sat well above Intel's range, and someone in investor relations watched that gap for weeks and let it be.

Expert:Oh, come on. What should they have done, ring every analyst and whisper "lower"? That's selective disclosure, giving material news to some investors before everyone else. It gets you a letter from the SEC, the US markets regulator.

Host:There are legal ways to do it. A conference appearance, a more cautious tone on the previous call. Companies talk expectations down all the time.

Expert:And sometimes the facts show up late. The Israel shutdown happened in October, and the memory squeeze got worse through the quarter. You forecast on what you know.

Host:So the market gets the blame.

Expert:Some of it, yes. This is my opinion, not a fact: the stock was priced for perfection. Yahoo Finance pointed out Intel had recently traded at more than fifty times estimated 2027 earnings, way above its peers.

Host:For Intel. Whose foundry still hasn't landed a big outside customer.

Expert:Foundry meaning making chips for other companies, the way TSMC does. Morgan Stanley's analysts talked about how expensive the stock was on next year's numbers. D.A. Davidson downgraded it in a note CNBC says was titled "Not fixed yet."

Host:Brutal title. Fair, but brutal.

Expert:An analyst gets maybe one good headline a year. They used theirs.

Host:And the shareholder left holding the bag is...

Expert:The American taxpayer. The Trump administration took a government stake, and CNBC estimates the crash wiped about twelve billion dollars off it in one day.

Host:So the state is a shareholder that can't really sell, can't easily push management without a political fight, and just watches.

Expert:That's your opinion, and it's fair. Mine is that the stake makes Intel harder to read. Investors can't tell whether the floor under the price is the business or Washington.

Host:Fine, give me the strongest bull case. The real one, not the press release.

Expert:The demand problem is really a supply problem. When supply limits the quarter and orders don't, revenue comes back once the bottleneck clears. Investing.com reports that management expects shortages of wafers, the silicon discs chips are cut from, and substrates, the base the chip sits on, to keep limiting output in the first half of next year.

Host:"We can't make enough" is the humblebrag of earnings season.

Expert:It is. Still, it's a different problem from "nobody wants it."

Host:Is it, though? If you can't make it, your rival can.

Expert:Good hit. Fabless rivals, companies that design chips and outsource the manufacturing, can move orders to a partner's factory. Intel owns its factories. That's the whole bet, and right now it's also the whole cost.

Host:And they're already calling next year an investment year.

Expert:Investing.com flagged that. Growth for data center, and also higher capital spending.

Host:"Investment year." Translation: please don't open the cash flow statement.

Expert:Every CFO has used that line once. I have opinions about people who use it two years running.

Host:Here's what bugs me as a finance person. Zinsner did it by the book. He gave a range, he put a number on the Israel hit, he didn't hide anything. And he still got the worst day since the seventies.

Expert:That's my point exactly. The market doesn't punish the honesty, it punishes the gap. It's pricing the distance between your forecast and the story it had already told itself.

Host:And whose job is it to manage that story? The CFO's.

Expert:Partly. I'm not going to accept that a finance chief is responsible for whatever multiple a crowd decides to pay.

Host:I think you're letting him off.

Expert:And I think you want CFOs to be mood managers. We're not going to settle this one.

Host:No, we're not. So what decides who's right?

Expert:First, foundry customers. Tan told analysts he expects announcements. If a big outside name signs, the whole "Not fixed yet" narrative changes.

Host:And if nobody signs, the crash looks cheap in hindsight.

Expert:Second, memory prices. While AI keeps soaking up memory, PCs stay expensive and the PC business stays under pressure. Intel doesn't control that at all.

Host:Third, I'd say, is the next forecast. Intel just set itself a very low bar.

Expert:That's the awkward one. If they clear it easily in three months, people will ask whether this was honesty or sandbagging, which means deliberately setting a low bar so you can beat it.

Host:Picture that call. "Congratulations on beating the guidance that crashed your stock."

Expert:And the stock goes up, and nobody mentions the crash.

Host:Because nobody has any memory.

Expert:It's all in the data centers.

Host:Then here's the question I can't shake. If a forecast can be honest, specific and right, and still cost this much, what is a CFO's credibility actually worth to the market? Drawn from CNBC, Investing.com and TechCrunch, links in the show notes. We'll stop there. The whole CFO track, in order, is waiting at mba-training.com.

Read and go deeper