A wafer of 3D XPoint memory chips. Image source: Intel. 

At Intel's (INTC 1.53%) investor meeting last year, the company highlighted its non-volatile memory solutions group, which the company says "includes NAND flash memory products primarily used in solid state drives," as a key growth segment. It was even kind enough to finally break out revenue and operating profit for this segment separately some time ago so that investors could keep tabs on this business' revenue and operating profit trends.

Unfortunately, this segment did quite poorly last quarter. Revenue plunged 20% year over year to $554 million, and the business swung from a $92 million operating profit in the second quarter of last year to a loss of $224 million.

What's driving these poor financial results? Let's dig a little bit deeper.

Competition driving pricing down; start-up costs hurting profitability

CEO Brian Krzanich said that the 20% year-over-year decline during the quarter represented a shortfall relative to the company's prior expectations. This revenue weakness was, per Krzanich, a result of "a more competitive pricing environment."

However, this "competitive pricing environment" wasn't the only contributor to the steep decline in the company's operating income in this segment. CFO Stacy Smith attributed it to pricing along with two other factors.

The first factor is that the company decided last year to convert its China-based logic manufacturing plant from a trailing-edge logic factory to a factory that builds non-volatile memory devices (both 3D NAND flash and 3D XPoint technology). The start-up costs associated with bringing this factory online contributed to the operating profit decline, according to Smith.

Additionally, Smith said that the company saw "increased 3D XPoint spending." I suspect this increased spending is research and development aimed at facilitating the development of future 3D XPoint memory technology generations.

Expect more losses during the second half of the year

In terms of financials, Smith told investors to expect operating losses to continue in the second half of 2016 and will be "consistent to what [Intel] saw in the first half [of the year]." He also indicated, though, that revenue in this segment should "see an improvement" in the second half of the year. "We'll see the first production costs play through on 3D XPoint, which as you know from watching us over the years, those production costs tend to be fairly high in any factory when you're starting it up," Smith explained.

Taking the longer-term view

Although this business is expected to continue to bleed money for the remainder of the year, management remains upbeat about the longer-term prospects of the company's non-volatile memory business.

Krzanich was quite positive on its 3D NAND technology, claiming both "large cost advantages" and a "good performance position" with this technology. He also took the opportunity to play up its 3D XPoint technology, which he expects to arrive in traditional storage drives by the end of the year and as memory modules next year.

Smith, unsurprisingly, said that over the longer term, due to both 3D NAND as well as 3D XPoint, the company's memory segment should have a "good value proposition and a very good overall profit position for the business."

In light of these comments, investors should pay close attention to what management has to say about what it expects for its non-volatile memory business next year. Can this business swing to profitability as 3D NAND and 3D XPoint are in full swing? Or will investors have to wait another year or more before it starts to contribute positively to the bottom line?