There are potential earnings catalysts for the company beyond revenue guidance, a J.P. Morgan analyst says

Expectations for Nvidia's earnings reports have soared ever since the artificial intelligence boom got going. But as the AI frenzy has continued, the power of strong revenue guidance in those Nvidia reports to move its shares higher has diminished.

On Wednesday, after the stock market closes, Nvidia will release its latest earnings report. One Wall Street analyst is pointing to important catalysts for the stock that will come on the investor call immediately following the earnings release.

J.P. Morgan analyst Harlan Sur noted that while the chip maker's (NVDA) revenue outlook has come ahead of Wall Street's consensus over the past four quarters by an average of 4%, its stock has fallen between 3% and 5% on average in the week and month following those reports.

Sur said in a Monday note to clients that what Nvidia says to address concerns over its competitiveness and some of its business decisions when it reports July-quarter earnings results could be more key for the stock.

Questions over Nvidia's dominance in the AI-chip market have weighed on the stock as its customers roll out custom silicon and companies like Cerebras $(CBRS)$ emerge with inference-focused products, he noted.

That narrative "will be challenging to dispel entirely" as several chip programs are expected to ramp up in the next few years, Sur said.

Still, Nvidia could emphasize its advantage in having a flexible platform that can meet the needs of enterprise and sovereign AI efforts while offering better performance costs, Sur said. He said those factors will help Nvidia keep its lead, even as its GPUs eventually lose some share of the total addressable market for AI compute to custom chips.

Nvidia's commentary on the long-term benefits to its business from its recent infrastructure funding agreements could also be a driver for the stock, Sur said, adding he expects the company to frame the deals "as infrastructure enablement endeavors" using third-party capital to address bottlenecks in power and other data-center needs.

Earlier this month, Nvidia said it is partnering with a crop of financial firms, including Apollo Global Management $(APO)$, Blackstone (BX) and Goldman Sachs $(GS)$, to deploy more than $500 billion in third-party capital toward the AI data-center build-out. The chip maker also announced a partnership with SB Energy to deploy its AI infrastructure.

UBS analyst Timothy Arcuri noted earlier this month that the infrastructure agreements help address worries about the durability of AI demand because they ensure infrastructure funding.

That supports "both the financeability of AI compute and the strength of [Nvidia's] ecosystem," he said.

Any updates Nvidia provides about its business in China could also be important given a report that some customers in the country have received initial deliveries of approved H200 chips, Sur said, citing the Financial Times. Those shipments "represent a material upside lever" to Nvidia's guidance for the October quarter, he said.

Meanwhile, AI-driven demand for memory chips has led to tightening supply and surging prices. Sur said Nvidia has adjusted the high-bandwidth memory content in its upcoming Rubin and Rubin Ultra GPUs, and the memory content in its Vera central processing units, to work around supply shortages, citing research from another J.P. Morgan team.

While there's uncertainty over how that will work out, how the chip specification change "impacts top-line growth and margins has been front-of-mind for investors," Sur said.

UBS' Arcuri said that inflation in memory-chip prices has been a major driver of hyperscaler capital expenditure increases in recent quarters. The dynamic of compute supply lagging far behind demand "is ultimately bullish for [Nvidia] and as a result, we may hear about another significant stair-step in backlog on this earnings call," Arcuri said in a note to clients.

Don't miss: Nvidia is the beating heart of the AI boom and the stock market - which sets up a big test

Nvidia is expected to report revenue of $92.3 billion for the July quarter, according to estimates compiled by FactSet. Analysts are looking for its data-center segment revenue to grow 110% from a year ago to $86.3 billion. The FactSet consensus calls for adjusted earnings of $2.09 per share.

For the October quarter, analysts are eyeing adjusted earnings of $2.37 per share, and overall revenue of $104.2 billion, according to the FactSet consensus.

Arcuri said in a note earlier this month that he could see Nvidia's October-quarter revenue reaching more than $110 billion as its Blackwell shipments are "holding stable" and the upcoming Rubin platform is entering the mix. Rubin "should be a more substantial revenue step-up" in the fourth quarter as shipments accelerate, he said.

-Britney Nguyen