Success in the consumer electronics (CE) industry has always revolved around hardware innovations. But as products become more complex and feature rich, it is the design and function of software-driven interfaces that increasingly differentiate the players and determine the winners.
Software developers still face all the challenges you'd expect in a market where nearly every application is embedded and must be as fast, cheap, proven, and modular as possible. Development cost is also a major factor in profitability, especially in commodity categories such as VCRs and televisions, where margins are slim.
But perhaps the most significant trend in consumer electronics today, particularly from the standpoint of software engineering, is the accelerating need to integrate devices -- from cell phones to cameras -- with PCs, the Internet, and other technologies. Indeed, ease of connectivity can be the deciding factor in a product's acceptance. Bringing a well-designed, well-connected product to market at warp speed requires a shared vision and a well-coordinated process encompassing all the players on the team: hardware engineers, software developers, product designers, and market researchers.
To learn more about how changes in the consumer electronics business are impacting software development, reporter Scott Cronenweth interviewed two industry experts at Forrester Research: Josh Bernoff, Principal Analyst, and Jed Kolko, Senior Analyst. Bernoff currently specializes in the television marketplace and has broad expertise on the supply side of consumer electronics; Kolko focuses on the demand side, through his research on consumer devices, access, and services.
SC: What are the key factors that determine success in the consumer electronics marketplace? Getting an innovative product to market first? Pricing? Gee-whiz features? Usability? Interoperability with the PC?
JB: There are really only two strategies for success in consumer electronics. One is to get a new product out before anyone else and then defend your market leadership position. You can command a comparatively high price for your product by going after early adopters. The trouble is, however, that in this extremely competitive environment, any worthwhile advance generally gets copied pretty quickly.
So the second strategy is to be a fast follower. In the long run, most of the money ends up being made by the Toshibas and Sonys of the world coming in on the heels of the early leaders and producing products that have either some special design feature or a lower price tag. Those companies' success rests on having the cheapest manufacturing and prodigious distribution strength. To achieve economies of scale you need global reach and global manufacturing capability, and you have to get into the right stores. Unless you're in Circuit City and Best Buy you're unlikely to succeed in consumer electronics.
The upshot is that, by and large, products in the same category are very similar to each other. When this happens, buyers become very sensitive to price, and they pay less attention to what company label is on the product. Cost is especially important at the lower end of a given market. Design and look-and-feel -- the way the remote control looks; the way it feels in your hand if it's a portable device -- are important differentiators at all price points.
SC: Let's go back to what's required to succeed with that first strategy. How do you capture those early adopters if you have an innovative product?
JK: What we consistently find in our research is that the most successful product launches happen when consumers are presented with reasonable expectations and basic applications.
If a consumer is not comfortable using the keypad on her cell phone to send a quick message, then she will not have the experiential background to comfortably embrace SMS (short message services) or wireless instant messaging, or to get excited by an ad campaign around mobile commerce or video conferencing. This critical need to focus on what consumers really want makes it imperative that organizations establish clear, ongoing communication and a shared, clearly understood vision across their marketing and development functions in the design, development, and introduction of a new product. Otherwise, consumers may fail to adopt even well-designed offerings.
In introducing wireless technology for cars, for example, the telematics industry has wisely focused first on emergency assistance and other very basic safety and security features. As opposed to trying to hype real-time traffic reports and automatic trip calculations, which don't even really exist yet. What's currently going on in the home networking arena will be an interesting test case. The hype there now is "Get broadband, get a home network, and soon your PC will be talking to your washing machine." The right message is "If you get a home network, then multiple PCs in your house can share a broadband connection." This is a very straightforward application of the technology, but one that resonates with a need consumers have -- and are aware of -- now. Few people really care about the sexy-but-impossible applications.
SC: The logic of what you're saying seems almost beyond debate. So why are straightforward value propositions so hard to come by in the marketplace?
JK: I think it's simply that marketing and market research teams on one hand, and design and core product development teams on the other hand, represent two very different cultures and viewpoints. Engineers focus on creating a device that incorporates cutting edge technology and provides the best solution to a problem. Whereas the marketing team is focused on the more mundane factors that drive a consumer toward a new technology -- or keep them away from it. Because most product developers and designers are accustomed to pushing the limits of technology, it's naturally rather easy for them to lose sight of the pragmatic reasons why consumers either adopt or don't adopt the technology.
For instance, many consumers hesitate going to broadband because they don't want to give up the e-mail address that they have with their dial-up service provider. All the technological improvements in the world won't get consumers to make the shift if their main concern is the hassle of telling all their friends they have a new e-mail address. Now that's way below the development radar screen! But factors like these are often on the minds of the marketers whose job it is to put themselves in consumers' shoes. The key is to communicate and maintain a common goal and a coherent business model for the product as it goes to market. That's a step that product teams sometimes gloss over, and that can spell disaster.
Figuring out the requirements for a product should begin with figuring out what's important to your potential customers. The issue isn't a lack of features; it's a lack of usability. A lot of developers rightly focus on the most extreme possibilities for a new technology. But marketing has to counterbalance that bias; its focus should on the less glamorous capabilities that will actually get consumers to buy the product.
Take a look at the Palm Pilot. It succeeded in a category where all the previous products, like the Newton, had flopped completely. Why? First, the original concept of what would appear on the screen, and where the buttons would be, was extremely well thought out. Second, it was tested and refined for ideal usability, so people liked it immediately. If you follow a new product vision completely in your own head, then you're probably not going to understand everything that users want. But if you have a brilliant idea and you combine it with the testing that's required to really refine it, then you'll succeed.
JB: At Forrester we see examples all the time of new products that do things nobody's ever seen before, that were engineered very creatively, and that fall flat on their faces. And in many cases it's because they do something that people don't want. Like the WebTV viewer, 1 which doesn't particularly appeal to that many people. But it's more frequently the case that a product flops because its design is so clunky that it's a major challenge for people to figure out how to use it. The ZapStation 2 comes to mind in that category.
One of the insights that we found really instructive came from a report I co-authored at the end of last year called The Secret to Device Success. We looked at MP3 players, digital cameras, and PDAs (personal digital assistants), comparing their interfaces and their overall usability. What became abundantly clear was that the ability of these kinds of devices to connect in a rational way to the PC was hugely important in their success. And you can readily see that all three of those product types succeed or fail in many ways based on how easy it is to connect them to the PC and copy or move information back and forth.
In that comparative context, one can also cite examples of products that seemed to have been introduced too early, before the company really understood the usability issues they needed to address. And a product that's inferior from a usability standpoint in any of those spaces is simply not competitive. I think the lesson here is that the design and features have to add up to making usability effortless -- which is what it really takes to be successful. And as hardware designers hand things off to the software designers, you have to ensure that the focus on usability is not lost.
SC: So how does this need for effortless usability, along with all the market pressures you mentioned --- for rapid introduction, rapid innovation, low cost -- impact the people who develop software for consumer electronic devices?
JB: Consumer electronics today is a challenging environment for developers, obviously. The software that runs in these products is almost always embedded, and the bottom-line goal is to make it as small and cheap as possible. Modularity is also increasingly important: If you can't drop in and reuse components, then you won't make it to market quickly, and your costs will go up. A component-based architecture also makes it easier for you to refine the design or the interface of a device in response to usability testing or other market research.
Then add new levels of complexity to the equation. Software in CE devices used to consist of little more than the enabling layer that drives things like the display that appears when you hit the Volume Up key on your television remote. But that's changing rapidly as we see more and more sophisticated products like a TiVo or a cable set-top box. There's some relatively advanced stuff going on in there. Inside the TiVo, for example, is Linux.
By far the most important driver for increased software complexity, however, is the value consumers perceive in being able to connect a device to your PC or to something on the Internet. ReplayTV, for example, has a system whereby you can go to a Web site and tell that site what programs you want the device to record. Then, when it does its daily call to download information, the device will recognize and act on your commands; there's actually a Web component to its command interface. Likewise, products like PDAs and cell phones are designed to communicate with PCs and in some cases with the Internet, in order to update the information they're presenting. Now you even have PDAs beaming software at each other.
Integration across CE devices hasn't traditionally been much of an issue, but now that's changing, too. By and large people have simply bought components. They bought speakers or a television or a DVD player, and these products were relatively easy to hook up together. Now that television setups in particular are getting more and more complicated, the ability of these products to communicate with each other is becoming a whole lot more important.
What all that means for developers and manufacturers is that you can't just shove the software into the device any more. A more connected kind of world is opening up, a world in which these new devices need to work well across a much broader spectrum of interfaces and inputs. There's more to test, more code to control, more people involved, and so forth. There's also much more emphasis on interfaces. Cell phones, MP3 players, the TiVo, and even some DVD players have interfaces. These are not the hardware products of the past, when the interface essentially consisted of pushing the Play button on the remote control. As more devices get screens, more products require the capability to connect to one another, and more and more of them demand interaction, this sort of interface competence becomes increasingly important.
Software design and development practices therefore need to be as good as they can be, because the competitive playing field in consumer electronics is so even and so fiercely contested that something like your embedded software interfaces can easily make or break you.
SC: A couple of your previous remarks imply that many consumer electronics products are designed and built using what software developers term a "waterfall" approach, which can make it difficult to make changes as requirements are refined. Is that a valid assumption?
JB: Yes, I think you're right on target. In many instances we've seen of substandard usability, the manufacturer could easily have done exactly the tests that we did to detect flaws. But often by the time that level of testing takes place, those sorts of problems are already "baked in." What's interesting to me is that computer companies are increasingly beating traditional consumer electronics companies at their own game -- like HP with digital cameras and Apple with MP3 players, for instance. And I think a big reason is that the PC vendors have a better software development methodology. They're accustomed to the idea that your first take on how things might work doesn't necessarily give you the best result. You have to test prototypes with users and redesign the system iteratively until you get it there. That means validating components at each step in the development cycle. The more complex the software, the more crucial that approach becomes, both to reliability and to usability. It also means implementing automated testing and other controls to ensure quality. A lot of high-tech companies have adopted a process that provides these best practices and guides the activities you need to accomplish all of this. Traditional consumer electronics manufacturers could profit from adopting this approach.
Another issue is that sometimes there's a real disconnect between process demands for different products. Say you're manufacturing a personal stereo. You want to have the best possible design, you want to design the features well, and then you want to be able to make 100 million of them so they're cheap. But, suppose you decide to branch out into making digital cameras. For those, you need to ensure that the device works well in conjunction with a PC and has a seamlessly usable interface. That requires a different process and a different set of competencies than you need for stamping out personal stereos, and companies that don't recognize this aren't going to make it when they venture into new markets.
SC: So are companies that mass produce cheap goods like personal stereos outsourcing to get the skills and process they need to make more sophisticated consumer electronics?
JB: What we see is a growing trend among the big consumer electronics companies toward marketing and distributing products designed by startups, and in many cases the products are actually assembled and manufactured by yet another company. Again, the TiVo is a good example: It's designed by TiVo, Inc. but manufactured and distributed by Philips, Sony, and others. So you do, increasingly, see a sort of division of labor in this business.
And even within those startups, you may see development teams turning to third-party software to speed things up and control costs. If you're making a VCR, there's not a whole lot of value in writing the embedded operating environment yourself. The problems have already been solved. So you might start with a commercial application and perhaps customize it somewhat. But if you're making a cell phone, the interface capabilities of that phone in many ways represent your differentiation in the marketplace. So you are more likely to expend some serious effort in either writing the operating system yourself, or customizing an existing platform extensively. If you're making a personal video recorder, then once again the software is really your point of differentiation. It is everything.
In short, the decision about whether or not to develop software in-house really looks very different, depending on whether you're creating a commodity product, or one that you plan to differentiate based on software-driven features. Today, commodity products account for perhaps 90 percent of the consumer electronics industry. But if you go into Circuit City and look closely at what they're selling, for many of the higher margin products, it's the software that distinguishes that product from everything else in the category. You could even say that those new, software-driven interfaces and capabilities are a primary driver for growth and innovation across the whole CE industry.
1 A Microsoft offering that simulates the television browser on a personal computer, to help verify that the Web content displayed is appropriate for the form factor and resolution of the receiving device.
2 A digital media product from ZapMedia, designed to enable consumers to access digital content stored on a PC via a television or stereo.
Scott Cronenweth is a freelance writer specializing in technology and business strategy, including e-business, software development, and knowledge management (KM). His fifteen-year career includes diverse experiences as a documentation manager, technical writer, marketing copywriter, and reporter. Currently, he writes magazine articles, whitepapers, and marketing collateral for technology companies, publishers, and PR agencies nationwide. He holds an M.S. in information science from the University of Pittsburgh.
Comments (Undergoing maintenance)
Table of contents
Local resources
My developerWorks community
Tags
Use the slider bar to see more or fewer tags.
Popular tags shows the top tags for this particular content zone (for example, Java technology, Linux, WebSphere).
My tags shows your tags for this particular content zone (for example, Java technology, Linux, WebSphere).
Popular article tags |
My article tagsSkip to tags list
Popular article tags |
My article tags
