Since Intel announced its latest quarterly earnings, which were not stellar, and recently announced it would be dropping work on certain of its mobile chips aimed at the mobile phone market, much has been speculated about Intel’s future. Earnings more or less met expectations, but expectations were already lowered. And Intel’s prediction for the next quarter wasn’t inspiring. The company also announced a 12,000 employee reduction over the next year as it moves to a more efficient operation. And dropping chips aimed at a market where it was struggling to gain traction made it look as if it was giving up.

[Full disclosure: Intel is a client of mine.]

So what is Intel’s problem? A large piece of its revenues (>50%) is driven by the Personal Computing Group, which includes PC, tablets, phones, wireless, etc. By its own estimation, Intel expects the PC market to decline this year in the 7-10% range. That clearly hurts. On top of that, this quarter saw Intel’s sales into the tablet space drop by 44%! Tablets, especially high end tablets where Intel plays most effectively, is also a market in retreat. But in my estimation, a large part of Intel’s reduction was due to backing away from its high level of subsidies to vendors who used Intel chips. Indeed, Intel stated its margins increased dramatically in the tablet segment, so clearly it is no longer willing to sell parts at any price to establish market position. Finally, its pullback from its SoFIA product line — aimed squarely at the smartphone market, a market it has failed to penetrate in any meaningful way, and with chips that were well behind the competition in wireless connectivity capability — looked like a capitulation to its fate. But not all is bad news in Santa Clara.

The highlights in Intel’s earnings were in the Data Center Group (servers) and IoT markets. While only about half the size of the PCG, the DCG generated close to as much in profits ($1.885 billion vs. $1.764 billion). Clearly, cloud is becoming an increasingly important market as more severs go into data centers whose purpose is to enable cloud-based apps and services. Intel has played a key role in supplying the processors that power these cloud “factories”. And while the rollout of enterprise based data centers has done well for Intel in the past, the larger market is now being driven by other service centers geared towards consumer and business services.

But in this space, where Intel has traditionally sold well, it is now being challenged by small, lightweight but large-volume servers powered by ARM-based chips. This will remain a competitive threat to Intel going forward, but I expect Intel to continue to react appropriately and x86 architecture to remain the dominant engines powering the quickly growing cloud ecosystem for the foreseeable future. In this space, it’s not only about chips, but also about the total software ecosystems.

In the IoT segment, Intel is clearly coming from behind. ARM-based technologies, so dominant in the mobile phone space, have captured a large portion of the market for consumer-based “things”. But clearly there is a very large and lucrative industrial IoT space that is much less cost sensitive and more interested in high performance and compatibility in software. Although not as big a potential market as consumer products in volume, it represents a much more profitable place to play and can generate substantial revenues (and profits). It’s also a market that Intel knows well and has played in for some time with its large embedded business.

That’s not to say that Intel won’t look to move downstream into the higher end consumer-focused “things” space, but it will pick and choose its markets. It has been bulking up in key areas, with chips (Atom, Edison, Quark), software development (Wind River), security (McAfee), sensors (Basis, Recon), virtual reality (Intel RealSense, Replay Technologies), drones (Ascending Technologies), etc., which will put it in good position to play in health care and personal appliances as well as smart cities and AR/VR application areas. I expect to see more strategic asset purchases in the coming months and years.

And we shouldn’t assume Intel is done in the mobile/wireless chip segment. Even though the current generation of relatively non-competitive SoFIA chips has been cancelled, Intel has a good deal of momentum and R&D in place for the next big leap — 5G. Despite some high profile demonstrations, commercial large-scale deployments of 5G are still 3-4 years away (there isn’t even an agreed upon 5G spec yet). So Intel has plenty of time to create a major presence in the 5G space, although it faces stiff competition, primarily from Qualcomm.

Going forward in the cloud and IoT markets, Intel’s traditional PC and server market competitors will change. Indeed, Intel’s primary competitors will be ARM-based, and its single biggest competitor will no longer be AMD but Qualcomm, which itself is trying to build out a broader base of chips targeted at the “things” space. Certainly there will be many other players nipping at Intel’s heels, especially those in the Far East, but the best-positioned competitor will remain Qualcomm. AMD will make a play for the server space with its ARM-based products, but its relatively small footprint won’t be a big threat.

Intel does have some interesting leverage points that could give it a competitive advantage in the IoT and small server markets. It will continue to make PC and server chips. After all, this is still a huge market generating $7.5 billion in PCG and $4 billion in DCG revenues this quarter. Its large investments in CPU and GPU architectures for the high end, together with its work in new wireless technology, can be readily repurposed to lower end systems, and for relatively low cost. This means Intel can produce cost competitive chips and sell them at relatively high margins as the R&D to develop the chips has already been accounted for.

Another advantage that Intel has is its ability to create a complete infrastructure for the new cloud-based and “things” environment. Intel hasn’t just developed expertise in chips but has also invested heavily in the analytics capabilities required to make any “things” environment work. It also has a significant play in security (McAfee), which is a must-have in the new cloud-based world of “things”. And it has a significant embedded OS operation (Wind River) to help it with developers. All of these are competitive advantages that other players will have difficulty matching on their own.

Overall, the realignment of Intel’s business towards growth areas and away from more mature and stagnant markets will help it, even if there is some pain involved (the reduction of 12,000 workers by next year). It would be a mistake to think that Intel can’t pivot to the new markets and generate significant revenues at compelling margins (it has consistently had the best margins in the business). The transition will take a couple of years, but Intel will remain a major force in the emerging market areas, and in some cases, will be dominant.

[A version of this story originally appeared on the author’s website.]

Jack Gold is founder and principal analyst at J.Gold Associates, an information technology analyst firm based in Northborough, MA. Follow him on Twitter: @jckgld.

VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Learn more about membership.