Innovation can be of several types, Geoffrey A. Moore writes a taxonomy in his article Darwin and the Demon: Innovating Within Established Enterprises (see below).

In the article, he then matches the different kinds of innovations with the classical product life cycle diagram. In the initial phases, according to Moore, we have disruptive innovation, application innovation, product innovation. As the product mature, process innovation, experential innovation and marketing innovation are important. In the declining phase, business model innovation and structural innovation can be used.

Disruptive Innovation. Gets a great deal of attention, particularly in the press, because markets appear as if from nowhere, creating massive new sources of wealth. It tends to have its roots in technological discontinuities, such as the one that enabled Motorola’s rise to prominence with the first generation of cell phones, or in fast-spreading fads like the collector card game Pokémon.

Application Innovation. Takes existing technologies into new markets to serve new purposes, as when Tandem applied its fault-tolerant computers to the banking market to create ATMs and when OnStar took Global Positioning Systems into the automobile market for roadside assistance.

Product Innovation. Takes established offers in established markets to the next level, as when Intel releases a new processor or Toyota a new car. The focus can be on performance increase (Titleist Pro V1 golf balls), cost reduction (HP inkjet printers), usability improvement (Palm handhelds), or any other product enhancement.

Process Innovation. Makes processes for established offers in established markets more effective or efficient. Examples include Dell’s streamlining of its PC supply chain and order fulfillment systems, Charles Schwab’s migration to online trading, and Wal-Mart’s refinement of vendor-managed inventory processes.

Experiential Innovation. Makes surface modifications that improve customers’ experience of established products or processes. These can take the form of delighters (“You’ve got mail!”), satisfiers (superior line management at Disneyland), or reassurers (package tracking from FedEx).

Marketing Innovation. Improves customer-touching processes, be they marketing communications (use of the Web and trailers for viral marketing of The Lord of the Rings movie trilogy) or consumer transactions (Amazon’s e-commerce mechanisms and eBay’s online auctions).

Business Model Innovation. Reframes an established value proposition to the customer or a company’s established role in the value chain or both. Examples include chestnuts like Gillette’s move from razors to razor blades, IBM’s shift to on-demand computing, and Apple’s expansion into consumer retailing.

Structural Innovation. Capitalizes on disruption to restructure industry relationships. Innovators like Fidelity and Citigroup, for example, have used the deregulation of financial services to offer broader arrays of products and services to consumers under one umbrella. Nearly overnight, those companies became sophisticated competitors to old-guard banks and insurance companies.

Peter F. Drucker writes in The Discipline of Innovation: 

There are, of course, innovations that spring from a flash of genius. Most innovations, however, especially the successful ones, result from a conscious, purposeful search for innovation opportunities, which are found only in a few situations. Four such areas of opportunity exist within a company or industry: unexpected occurrences, incongruities, process needs, and industry and market changes.

Three additional sources of opportunity exist outside a company in its social and intellectual environment: demographic changes, changes in perception, and new knowledge.

True, these sources overlap, different as they may be in the nature of their risk, difficulty, and complexity, and the potential for innovation may well lie in more than one area at a time. But together, they account for the great majority of all innovation opportunities.

when everything is free, differentiation must come from something else. As Seth Godin puts it, Free by itself is no longer enough to guarantee much of anything. 

Now, the same can be said for smartphone, when every phone becomes smart, how do you differentiate? A nice recent report is The “Smartphone” Is Dead: Long Live Smart Phones And Smart Gadgets.

The executive summary:

Apple’s and Google’s arrival in the mobile market is causing knock-on effects throughout the market and is opening up opportunities. All mobile handsets are becoming smarter and Internet-capable. Yesterday’s smart high-end phone is today’s midrange phone and tomorrow’s entry-level phone. The “smartphone” category is no longer useful as all phones become smart. Instead, we propose three new frameworks to segment the smart mobile device market: openness and extensibility; consumption and creation; utility and entertainment. All mobile strategies must adapt now: Consumer electronics makers must decide on their response to widely available smarter phones and the mobile Internet; handset makers must leverage software to play the mobile Internet game and differentiate long term; media, finance, retail, and other Internet companies’ strategies must exploit mobile opportunities now or lose ground to faster rivals. But the mobile market will remain fragmented with no single platform — no Windows PC equivalent — anytime soon on mobile devices. Therefore, mobile strategists must analyze their target consumers carefully before embarking on large mobile investments.
 

Software differentiation and Internet applications that complement the device seem to be the next battleground. Read NokiaWorld: Distracted by the N97 Flagship, Mobile App Stores Represent the new Battleground and Nokia: “Irresistible” solutions combining devices and services will be a key to future success.