The evolution of the modern automobile is arguably one of the most exciting and most important developments in the tech world today. In fact, it's probably one of the most important business and societal stories we've seen in some time.

The leadership at no less venerable a player than Ford Motor Co. obviously felt the same way and just replaced their CEO, despite his long-term tenure with the company, and the record-setting profits he helped drive during his 3-year leadership there. The reason? Not enough progress on advancing the company's cars forward in the technology domain, particularly with regard to electric vehicles, autonomous driving, and new types of transportation service-focused business models.

As has been noted by many, these three capabilities---electrification, autonomy, and cars as a service---are considered the key trends driving the auto market today and into the future, at least as far as Wall Street is concerned. In reality, the picture isn't nearly that simple, but it is clear that tech industry-driven initiatives are driving the agenda for today's carmakers. And it's pushing many of them into uncomfortable positions.

It turns out, however, that in spite of the importance of this critical evolution of automobiles, this is one of those issues that's a lot harder to overcome than it first appears.

Part of the problem is that as cars have advanced, and various technologies have been integrated into them, they've evolved into these enormously complex machines. Today's automobiles have as many as 150 programmable computing elements (often called ECUs or Electronic Control Units), surprisingly large (and heavy) amounts of wiring, numerous different types of electronic signaling and interconnect buses, and up to 100 millions of lines of software, in addition to the thousands of mechanical parts required to run a car. Frankly, it's somewhat of a miracle that modern cars run as well as they do, although reports of technical glitches and other problems in newer cars do seem to be on the rise.

In addition to the mechanical and computer architecture complexity of the cars themselves, the organizational and business model complexity of today's car companies and the entire auto supply chain also contribute to the problem. Having evolved over the 100+ year history of the automotive industry, the system of multiple Tier 1 suppliers, such as Harman, Delphi, Bosch and others, buying components from Tier 2 and 3 suppliers down the chain and car brand OEMs (such as Ford) piecing together multiple sub-systems from different combinations of Tier 1s to build their cars is notoriously complex.

If ever there was an industry ripe for disruption, and if ever there was an industry in need of a tech overhaul, the automotive industry is it.

But toss in the fact that there are often groups within the car maker that are specifically responsible for a given ECU (such as, say, heating, a/c and other "comfort" controls) and whose jobs may be at risk if someone suggests that the company changes to a simpler architecture in which they combined the functionality of multiple ECUs into a smaller, more manageable number and, well, you get the picture.

If ever there was an industry ripe for disruption, and if ever there was an industry in need of a tech overhaul, the automotive industry is it. That's why many traditional carmakers are concerned, and why many tech companies are salivating at a chance to get a piece of the multi-trillion (yes, with a "t") dollar global automotive industry.

It's also why companies like Tesla have made such a splash. Despite their very modest sales, they're seen as a credible attempt to drive the kind of technological and organizational disruption that many people believe is necessary to transform the automotive industry. In truth, however, because of the inherent and ingrained nature of the auto supply chain, even Tesla has to follow many of the conventions of multiple Tier 1 suppliers, etc., that its rivals use. The problem is that deeply embedded.

But even as those issues get addressed, they are really just a prelude to yet more innovations and opportunities for disruption. Like many modern computing devices---and, to be clear, that's what today's cars have become---the technological and business model for autos is slowly but surely moving towards a software and services-focused approach. In other words, we're moving towards the software-defined "digital car."

In order for that to happen, several key challenges need to be addressed. Most importantly, major enhancements in automotive security---both through architectural changes and software-driven advances---have to occur. The potential for life-threatening problems if either standard or autonomous cars get hacked should make this point painfully obvious.

Connectivity options, speed and reliability also have to be improved and that's where industry-wide efforts like 5G, and specific products from vendors like Qualcomm and Intel can make a difference.

Finally, car companies and critical suppliers need to figure out the kinds of services that consumers will be willing to pay for and deliver platforms and architectures that can enable them. Like many other types of hardware devices, profit margins on cars are not very large, and with the increasing amount of technology they're going to require, they could even start to shrink. As a result, car companies need to think through different ways of generating income.

Thankfully, a number of both tech startups and established vendors, such as Harman, are working on creating cloud-based platform delivery systems for automotive services that are expected to start bringing these capabilities to life over the next several years.

As with any major transition, the move to a digital car model won't be easy, fast, or bump-free, but it's bound to be an interesting ride.

Bob O'Donnell is the founder and chief analyst of TECHnalysis Research, LLC a technology consulting and market research firm. You can follow him on Twitter . This article was originally published on Tech.pinions.