2 Reasons I'd Buy Marvell Stock More Than 20% Below Its Record, and 1 Reason I'd Wait

On June 4, Marvell Technology (NASDAQ:MRVL) stock ended the day at $316.43, a record close. As of this writing, shares trade near $248, or about 21% below that record, even after bouncing more than 10% from their Oct. 22 close.

The business hasn't gone backward in the meantime. At an investor day on Oct. 6, the company raised its revenue forecast for fiscal 2028 (the year ending in January 2028) to around $20 billion. In August, it had called for about $18 billion. Reaching $20 billion would mean growth of about 67% from the roughly $12 billion Marvell expects this fiscal year.

I think the drop is a chance to buy, but not all at once. Here are two reasons I'd buy shares now and one reason I'd hold some money back.

Image source: Getty Images.

1. The data center target moved up a year

Data centers are where almost all of Marvell's growth comes from. They made up 79% of revenue in the fiscal second quarter of 2027 (ended Aug. 1).

Showing how fast that business is growing, Marvell's data center revenue rose 46% year over year to about $2.17 billion in fiscal Q2. What's more, management guided for data center growth to accelerate to about 75% year over year in the fiscal third quarter, and it now expects data center revenue to climb about 60% for all of fiscal 2027, up from its earlier forecast of about 50%. "AI-related bookings remain exceptionally robust," CEO Matt Murphy said in the fiscal second-quarter earnings release.

The longer-term targets have moved up even faster. In 2024, Marvell set a goal of $15 billion in data center revenue for fiscal 2029. Last year, it lifted the goal to about $18 billion. And at this month's investor day, it said it now expects to hit about $18 billion a year early, in fiscal 2028.

For reference, Marvell's data center business took in $6.1 billion in fiscal 2026, so the new target is almost triple that. Put another way, every update has pulled more growth into the same years.

2. The stock got cheaper as the forecast rose

Granted, Marvell stock isn't cheap on this year's numbers. At around $248, it trades at about 59 times the earnings expected for this fiscal year. Against expected earnings for fiscal 2028, though, the price-to-earnings ratio drops to about 34.

For comparison, that same fiscal 2028 earnings estimate would have put the stock at about 43 times earnings at its June record. And since June, management's revenue outlook for fiscal 2028 has climbed by about $3.5 billion, from roughly $16.5 billion in May.

Profits are keeping up with sales, too. Marvell's non-GAAP (adjusted) earnings per share grew 40% year over year to $0.94 in fiscal Q2, up from $0.80 in fiscal Q1. Management guided for $1.10 this quarter.

Sure, 34 times earnings is a premium price. But for a company expecting revenue growth of around 67% in fiscal 2028, I'd call that a fair price -- and arguably a better deal than buyers got in June.

The reason I'd wait: Too much rides on a few customers

Marvell's growth relies on a handful of buyers. Its 10 biggest customers, distributors included, accounted for 82% of revenue in fiscal 2026, according to its latest quarterly filing. One direct customer alone made up 16% of revenue in fiscal Q2.

This concentration matters most in custom chips, which Marvell designs for individual cloud companies and expects to more than double in fiscal 2028. In the same filing, Marvell warned that big cloud computing providers "have already begun investing in proprietary chip design capabilities," a trend it said "could reduce demand for our products."

A single lost program likely wouldn't stop Marvell's growth. But with so much revenue coming from so few customers, one cloud giant taking work in-house could eat noticeably into Marvell's revenue. After all, a growth stock at 34 times expected earnings can fall hard on that sort of news.

Is Marvell stock a buy?

I think so. Marvell's data center outlook is bigger than it was when the stock peaked in June, and shares cost less.

But customer risk like this often shows up with little warning, so I wouldn't put a full position to work right now. I'd consider buying some shares here and adding more as reported results show custom chip sales growing the way management expects.

Daniel Sparks and his clients do not have positions in any of the stocks mentioned. The Motley Fool has positions in and recommends Marvell Technology. The Motley Fool has a disclosure policy.