Oracle's Fall
Oracle's stock ran from 240 to 340 on its September earnings report. Since then it has been in freefall, cut by about two-thirds. It now trades near one-year lows around 116 to 118.
Web demand data tracks interest from enterprise customers on Oracle's business website, and this measure lines up fairly well with revenue. That enterprise web visit number is down about 11% compared to a year ago. The stock has dropped far more than 11% over the past year, so the sell-off runs well ahead of the demand drop.
Where Oracle Makes Its Money
Oracle's business breaks into three parts. Software as a service is about 21% of revenue. This includes HR, supply chain, CRM, planning, financials, and project management. All of it looks very exposed to AI, and to AI-driven software development that is coming soon.
Oracle Cloud Infrastructure is about 30% of revenue, growing quickly. This is the part everyone is betting on.
Legacy business is about half of revenue and is fairly defendable. It is less open to being replaced by AI development than the software-as-a-service lines are.
So only about 20% of what Oracle does is highly at risk. That is why the sell-off is puzzling. Oracle is not like Salesforce, whose whole business model could be flipped over by AI software development.
The All-In Bet
Oracle has purchase orders backed up in the $600 billion range, with about half of that coming from OpenAI. Oracle and everyone else are waiting on an OpenAI IPO. The backlog is very healthy, with a lot of future revenue lined up.
Against that, Oracle is spending money, taking on debt, and issuing new stock to build physical assets. Cash is getting tight, debt is climbing, and credit ratings are being cut. The credit rating sits just one notch above junk. If it drops further, forced selling could follow. Free cash flow is negative, and capex is heavy.
The market dislikes all of it: heavy capex, negative free cash flow, borrowing, and new share issuance. Investors want asset-light, easy models. If everyone thought that way, no one would ever build a home. Oracle is doing the opposite, and that is unpopular right now.
Everything piles onto one large bet on the future. I hesitate to call it a House of Cards, but that is the shape of it, and that is what makes investors nervous.
The Case For Buying
Oracle is best at critical data, which is its core strength, so few firms are better fit to serve this need. The customers are a wide spread, not just AI startups and OpenAI. Amazon, the Pentagon, and the Navy are all spending with Oracle because it has the hardware and infrastructure these operations need. Big established buyers, including the government, are betting on this hardware and these solutions.
Two years ago, or even a year ago, this was a different conversation. Now many investors ask first what could go wrong, and for Oracle that list is long, with cascading domino effects that could hurt the company. If you have the risk tolerance, you also have to ask what if everything goes right.
The stock is cut by two-thirds from September and sits at one-year lows. If AI works out and everyone moves as far forward as expected, 118 could look unbelievably cheap in hindsight. If they clear this rough patch, this could be the genius move of the year, and people may wonder how anyone could pass on Oracle at 118. The main risk to that story is an AI implosion.
This is almost a binary bet, which is rare for a company this size. Binary bets usually come from much smaller companies going all in one direction. Oracle either wins big or does not.
The Competitive Picture
Oracle's enterprise web visits are down 11% while competitors are up. Salesforce is up 8%, even though I am pretty bearish on Salesforce. IBM and ServiceNow are also gainers. Being last in a group of companies all exposed to future change is not a good spot.
The Roller Coaster
At the end of 2024 the stock was $99, under 100 dollars. It climbed all the way to 345, then fell back to 118. Shareholders have been on a rough ride.
For me it comes down to leverage: are they borrowing too much, is free cash flow burning too fast, is capex too high. The -1 demand reading is a negative, but the backlog is strong and the company is building for success. It is a tough call, weighing risk against reward.
