It's clear that Intel (INTC 0.61%) is finally making inroads in the tablet application processor space, but its efforts in smartphones have yet to be seen. While the company appears to be on track to ship its very first 22-nanometer system-on-chip, codenamed "Merrifield," by the end of the year for an early 2014 product launch, a single product isn't enough. Fortunately, Intel has a golden opportunity to present investors with a compelling roadmap for future products at its upcoming investor meeting. If Intel is able to do so, it is likely that sentiment will turn around heading into 2014.

What do investors need to see?
Although this Fool loves any and all things technical, most tech investors care about the "broad strokes" and try to use this to get a sense of how a particular company stacks up against its competition. That being said, there are a number of key buzzwords that investors will be looking for when the head of the mobile and communications group, Hermann Eul, makes his presentation at the investor meeting.

First and foremost, the big question on everybody's minds is integration -- in particular, cellular baseband and connectivity (Bluetooth, Wi-Fi, GPS, and NFC) integration onto the same piece of silicon as the applications processor. While MediaTek, Spreadtrum, and Qualcomm (QCOM 1.62%) have been able to participate in the high-volume/low-cost segments of the handset markets thanks to their ability to integrate most, if not all, of these functions, Intel has been largely on the sidelines here.

Interestingly, Intel had actually promised a "highly integrated" part at last year's investor meeting, but the part has not been mentioned for the entirety of 2013 (and whispers suggest that the project has been canceled). Naturally, it takes time to get the individual IP blocks right and then to actually integrate them on the same piece of silicon, but investors are going to start asking questions when everybody from the low-cost Asian chip vendors to the likes of Broadcom (NASDAQ: BRCM), Qualcomm, and others prepare to roll out integrated parts for deployment in designs next year.

Not just cellular integration
Intel's management has, on several occasions, alluded to the fact that its portfolio of connectivity solutions (Wi-Fi, Bluetooth, GPS, etc.) is complete. This is true in the notebook/desktop PC space, as Intel's combo solutions are quite popular. However, there seems to be a curious lack of design wins for these solutions in the recent crop of Intel-based tablets. Almost all of the devices that this Fool has examined have had either a Broadcom or a Qualcomm connectivity chip.

For the lower end of the market, having strong connectivity IP, and being able to integrate it onto the same die as the apps processor, is also key since this reduces the bill of materials and saves on both board space and power. An update from management on how this level of integration is progressing would be helpful in trying to predict a reasonable timeframe in which Intel will begin to participate in higher-volume portions of the smartphone market.

Don't forget performance
While the volume growth in handsets will clearly be at the low end, Intel also needs to participate at the high end, as this is where Intel's ability to design high-performance/low-power silicon will be a competitive advantage. With the firm's first 22-nanometer chip slated to be available in Q1 2014, investors will be able to verify via independent reviews just how competitive Intel can be in this market. More importantly, though, Intel needs to present a roadmap that is convincing and suggests a rapid product introduction cadence as well as leadership.

Foolish bottom line
Intel has made it clear that the competitive dynamics of the smartphone business are completely different from those of the tablet and notebook PC markets. It's going to take time for the company to advance beyond its market share of roughly 0.2%, but the question that should be on investors' minds is "when," not "if." This Fool is hopeful that on Nov. 21, Intel will give a compelling roadmap that finally answers the "when" question.