Thanks to the personal computer, the ink-jet printer and the digital camera, Kodak's 120-year head start in the photofinishing industry has been all but erased in the past decade. In fact, analysts such as IDC's Chris Chute say that Kodak is playing catch-up to companies like Sony, the world's leading digital camera manufacturer, and Fuji, which launched a FujiNet site to enable authorized users worlwide to massage digital images online, in figuring out how to shift from a paper-based world to a digital one.
On this new digital playing field, Kodak also must find a way to compete with companies such as Hewlett-Packard. The industry's leading printer manufacturer now makes more than two dozen digital cameras that are compliant with its printers and image-uploading software. Recently, HP began offering a printer-camera-memory card package for $299.
Carp's conundrum: How to not only make selling digital prints as painless as processing film for retailers, but also figure out a way to ensure Kodak products and services are needed in every step of the process.
The answer so far: Universal standards for digital printing that can be employed by anyone with an Internet connection.
Enter the International Imaging Industry Association (I3A), a not-for-profit group comprised of more than 90 members. Two years ago, the I3A announced that a dozen companies, including Kodak, had formed the Common Picture Exchange Environment Initiative Group to define the standards that would make it routine for digital images to be sent to retail photofinishers via the Internet, and for those photofinishers to do business electronically with vendors of supplies and services, such as Kodak.
The "environment" would link digital cameras, online photo service providers, retail photofinishers, and home PCs and printers through such Web technologies as Extensible Markup Language (XML), Simple Object Access Protocol (SOAP), Web Services Description Language (WSDL) and Universal Description, Discovery and Integration (UDDI).
Nuts, Bolts and Bits
A single set of technical standards holds out the promise of making it easier for the average person to order prints of digital photos over the Internet from either a Web site, which would mail them back, or a local retail outlet, where the customer would pick them up.
Today, this is typically a disjointed process. The consumer downloads photos from the camera using software provided by the manufacturer, in one series of steps. Then, the same prospective customer uses a Web browser to locate or relocate a printing service, in a second series of steps. In a third series of steps, the customer must upload the photos. Finally, in a fourth set of steps, the buyer specifies the number and size of prints for each image, and how to pay for the order.
Kodak wants to streamline this, using the EasyShare software that comes with its digital cameras.
When the digital photo standards are set, a Kodak customer might never leave Kodak's EasyShare software when ordering prints. After the camera is put in its cradle, new shots would automatically be uploaded and an EasyShare page would appear on screen. The user would highlight the images he likes and order prints immediately. The default supplier of prints might be Kodak, but the software would store settings of a local retailer or other provider, if desired.
The standards would make it possible for other camera manufacturers and vendors of photo editing software to embed similarly easy "order prints online" features into their products. And, hopefully, use Kodak retailers or Kodak itself as the default supplier of prints.
Camera phones are already on their way to having such features. In a promotion with Cingular Wireless this spring, callers could snap a shot on their phones, punch five digits and automatically have their results posted on the Web. It's not far-fetched to expect a user to punch in a different five numbers to have prints made and automatically mailed home, or ready for pickup at a store like Mortensen's. In-store kiosks would be another alternative to the home printer.
In all of these cases, the common standards would allow images to be shipped from any device—from phone to laptop computer to desktop hard drive—or from a Web site, to any other photofinishing location.
Version 1.0 of the CPXe gambit lays out the framework for how photofinishing services will be listed and looked up in a directory; and how to manage the uploading of images, placement of orders and confirmations of purchases.
The CPXe standards provide a vocabulary for scaling photos and for defining available print sizes. Tags indicating sizes and other features of the desired prints would be attached using XML, which regards data as a series of precisely formatted text documents. Instructions on what to do with the actual text or images in a file are set off in brackets that can be read by many different programs on different operating systems. That is in contrast with documents written in a binary language, which uses series of ones and zeroes. Such binary files are often intelligible only to a single program on a single platform.
XML tags define context. For example, the instruction Self portrait indicates how to regard the words "Self portrait."
But much more than contextual tags will be needed to effectively and efficiently transfer photos and tell each participant what to do with them.
SOAP, for instance, makes it possible to convey a request for a service to any remote computer and get back a response. For example, a message might include a "get price" command accompanied by parameters specifying how the price information given in response should be formatted. The application that receives a SOAP message is responsible for decoding it, executing the appropriate software instructions and sending back an answer or an error message if there's a problem, such as the service being unavailable.
The protocol doesn't guarantee, though, that the recipient of a message will actually know what to do with it. WSDL, another form of XML coding, makes it possible for an organization that publishes a Web service to define the capabilities of an application and the messages it understands. A particular photofinisher might choose to publish WSDL documents for a subset of the standard CPXe services, leaving out those the business doesn't support. WSDL documentation could be provided for additional functions, such as the ability to display the points a buyer has earned in a customer loyalty program.
Next Page: The final piece of the puzzle.
The final piece of the puzzle is UDDI, a Web services standard. This protocol allows companies to register their identities and services in an easily searchable vendor directory. Once a match is made, the two businesses can quickly conduct a transaction in the applications of their choice. In the digital photo business, the directory would maintain a master list of storage, processing and printing services.
This is not the same as an address book, however. Consulting a UDDI directory helps answer basic questions, such as the countries in which a particular service operates and where it will ship orders. A service-locator application can be built on top of the directory, but is not inherently part of it.
Given that such applications have to be custom-built, Kodak had the option to install in EasyShare a lookup application that only directs images to services operated by or allied with Kodak, which is what the company has today. However, other software could allow users to choose among multiple lookup services, which might make EasyShare less attractive to consumers.
So far, the only public directory of digital photo services is operated by the not-for-profit Picture Services Network (PSN), established by the I3A, and it remains nearly empty. Kodak and Kodak China maintain listings there, as does 30 Minute Photos Etc., a small photofinishing chain based in Irvine, Calif.
Jack DeMarti, a Kodak executive who chaired the CPXe 1.0 committee, says that using the PSN directory is not required for CPXe compliance. In fact, 30 Minute Photos is listed on both PSN and Kodak's proprietary directory, but does not yet provide services that use the CPXe standards. Instead, consumers bookmark a conventional Web page at www.30minphotos.com that includes instructions for uploading photos and placing an order from a browser.
Most of the services listed in Kodak's directory remain at this preliminary level of integration. But if the standards make digital exchange of photos slick and fast, it will justify 30 Minute Photos' enthusiasm for being among the first services listed. "It's essential to make sure we have a really seamless network," says 30 Minute Photos president Mitch Goldstone.
And if Kodak happens to benefit? Fine. "As Kodak goes, so goes the industry," Goldstone says.
30 Minute Photos is a customer of Graphx, which produces software for retailers to smooth out ordering photos over the Web. Graphx founder and chief technology officer Pete Traversy says the Woburn, Mass., company will incorporate CPXe standards into the next release of its Photogize software to speed the ordering of services over the Web from a UDDI vendor directory.
It's somewhat of a chicken-and-egg problem. Because Graphx services more than 50 photofinishing chains in the U.S., that release is being counted on to help jump-start the Picture Services Network. According to Patrick Ganglione, the network's chief operating officer, a directory listing will be bundled with the upgrade. That prospect in turn will help get the directory stocked with listings.
How important the standards will be remains to be seen, given that key suppliers of software for the exchange of documents over the Internet, such as Microsoft and Adobe Systems, "are doing their own thing" with online photo ordering, Traversy notes. "No one wants to give up the revenue, so no one wants to go to an open system,'' he says.
When the CPXe initiative was in its infancy, it had support from all the major players in the photofinishing world. Kodak was joined by Hewlett-Packard, Fuji, Canon, Agfa-Gevaert, Olympus, Pixology and a handful of other imaging services providers.
Omissions could spell confusion for consumers—and the end to CPXe's promise of a single worldwide set of standards.
"It was a great idea in theory," says Hans Hartman, an analyst at Future Image, a San Mateo, Calif.-based digital imaging research firm. "The remarkable thing about it all when it started was that all these companies said they were going to stop making proprietary systems. But that hasn't happened."
Back in early June 2002 when the initiative was gaining momentum, even Kodak's archrival, Fuji, seemed ready to sign on in a big way. "The CPXe initiative will create an entirely new and convenient option for digital camera owners to send print orders to their favorite local retailers from their home PCs," said Yukihiro Shibakawa, divisional manager of Fuji's digital imaging software development division, when the initiative was announced. "Fuji Film strongly believes in providing a choice of output options for all camera owners."
Fuji wasn't alone.
"In supporting this innovative CPXe standard, HP is pushing to create greater choice, control and convenience for consumers as they engage in the digital conversation," said Mary Peery, senior vice president for HP's digital imaging and publishing unit.
Hartman called it a galvanizing moment in the digital photofinishing industry. But two years later, most of the original participants have either abandoned the initiative or are re-evaluating their roles, according to Hartman.
Representatives from Fuji and Canon refused to comment. Those who do speak about CPXe are far from enthusiastic.
Ramon Garrido, HP's director for consumer imaging and printing, says his company is no longer active in the initiative.
"Kodak has been the champion of this effort," he says. "That's just the way it is. When [CPXe] was first discussed, we were pursuing photofinishing at the retail level. But now our focus is more on home printing."
Kodak is not yet ready to concede defeat. "CPXe will make things better for consumers and better for us," says Mark Cook, director of imaging systems for Kodak's digital and film imaging systems division. "It will create standard interfaces for all hardware and services to connect together. Ultimately, it will make digital printing easier for everyone and increase the total number of prints made."
Cook says that early on, getting involvement was a "collegial activity." Now, hard business decisions are being made. That doesn't happen overnight.
"With any industry standard, there's a significant period of evaluation,'' he says. This might take a year or two for some companies, or maybe five years for others."
Adobe Systems was originally viewed as a critical partner because its Photoshop software is the market leader for editing digital images. But Steve Saylor, Adobe's vice president of digital imaging, said in a statement prepared for Baseline that the company's strategy on online photo services includes "initiatives similar to CPXe [that] are built on partnerships with the leading photofinishing and photo merchandising providers." In effect, a single standard is not likely.
Other photofinishing service providers say Kodak's drive to be so thoroughly entrenched in all aspects of the digital imaging market is precisely why so few companies are willing to participate.
"Kodak is the 800-pound gorilla in CPXe," says Bala Parthasarathy, chief technology officer and co-founder of Snapfish.com, one of Ofoto's largest competitors. "They have the most complex of all relationships because they have a retail presence along with their online site and their line of home printing products."
According to Parthasarathy, there are larger business issues preventing Snapfish and others from signing on to the CPXe standard. "This means that Wal-Mart has to be OK with us taking their customers, and vice versa,'' he says. "This isn't going to happen anytime soon."
Basically, Future Image's Hartman says, that means the infant industry is "back to square one.''
"Manufacturers and photofinishers are making partnerships with each other based on proprietary software and hardware,'' he adds. "That means that about 90% of all the digital images printed are still being printed at home on ink-jet printers."
Printing Profits
Meanwhile, a second version of the CPXe standard is in the works, says Lisa Walker, executive director of the I3A. But underneath the issue of standards lurks the ugly question: How is Kodak going to make money off prints made from digital files?
Take, first, a look at CVS Corp., which owns and operates more than 5,000 drugstores throughout the U.S. Its photofinishing operations use Kodak technology and services.
Kodak officials won't say how much they've spent to connect CVS stores to the Ofoto Web site, nor will they say how many of the stores have been converted to digital mini-labs to process prints ordered either online or in person.
Such digital mini-labs cost about $10,000 apiece, says Ulysses Yannas, an analyst at consultancy Buckman, Buckman & Reid in Shrewsbury, N.J. Setting up a Web site is relatively inexpensive, costing perhaps a couple of thousand dollars, depending on the number of features the store wants.
Who picks up these costs? Not CVS. Kodak is footing the bill to make digital printing just as ubiquitous as film, according to Walker.
"Kodak isn't saying how much this costs," she says. "But let's just say that the retailers aren't paying for any of it."
Then there's Ofoto itself. When Kodak shelled out $58 million for the online photo site three years ago, many digital imaging experts viewed it as an act of desperation.
Because the company could not bring itself to embrace digital images as the future of its business in the '90s, Kodak ended up throwing a lot of money at an online operation that has yet to spew its first dime of profit.
What Kodak got for its $58 million was, at the time, 1.2 million registered users who didn't pay a cent for having their images stored on Ofoto's servers. They could store the images for free, and even show them to friends and family at no cost on the Web. They only had to pay to make prints. The economic incentive, then, was to use the images only online.
Today, Ofoto has more than 13 million registered users, but only 1 million of them have ever ordered a print, according to Kodak. The company adds that of those 1 million, 75% also printed digital images on their home PCs and printers.
Exactly how much Ofoto lost last year is anyone's guess. Kodak will only invoke the "investment mode" defense.
"It's funny because the Ofoto.com story hasn't changed," IDC's Chute points out. "The problems that Ofoto.com and other online photo sites face are simple but devastating."
Chute maintains that online sites will never be able to compete with home printing or retail printing services because consumers become accustomed to getting their prints in an hour or, at most, a few hours after dropping off their images. At this time, there's no way Ofoto or any other online site can deliver the finished product in such a short period.
"Second, consumers are using the sites for sharing the photos, but they're rarely buying prints," he says. "The average digital camera today is 3 megapixels, and the average image is about a megabyte in size. That's a ton of storage dedicated to a non-revenue-generating liability. We know that 90% to 95% of the images end up sitting on the server."
Kodak executives are figuring, long-term, that most people will only take the time to upload images that they plan to either print or share with acquaintances who will want to order prints. But most consumers share images as attachments to e-mail messages and print copies off home printers, Chute says. The PC and digital camera are their main sources of images.
"Very few people are using the online sites as their exclusive source for digital prints,'' he says. "It's just a free place to organize and save your photos for your convenience. It's a wacky business model."
With more than 450 million images stored on Ofoto, Kodak adds roughly 5 terabytes of new storage each week to accommodate the 5 million images added to the site weekly. That sounds like a lot, but it's relatively small potatoes for Kodak. It costs roughly a penny to store each 1-megabyte image for its lifetime, according to Stuart Fox, director of new business at Fotango.com, the online photo printing and sharing site from Canon Europe. That means Kodak only spends about $50,000 a week on additional storage to accommodate Ofoto's needs.
But if Chute is right, Kodak stores 10 to 20 images for each image that actually generates a print. That's 10 to 20 cents of storage cost when Ofoto only garners 29 cents, or less, for a 4x6 print.
"Kodak's digital printing business will be profitable when the volume of orders increases," says Yannas of Buckman, Buckman & Reid. "Especially if they can get people to order larger prints, 5x7s and 8x10s, which have greater margins."
To grab those greater margins, Ofoto offers items such as personalized photo books ranging from $29.99 to $39.99 each, and photo greeting cards at $19.99 for 20 cards.
Gone are the glory days of the Brownie camera when Kodak could count on keeping 70% of each sales dollar, after the direct costs of making a print from film.
"Margin compression is something Kodak and other film companies are going to have to accept in this digital transition," says Joe Kowalik, CEO of Graphx, which is providing the San Juan Capistrano store with the software it needs to process digital photo orders. "You're going from a film-based model that took decades to reach the economies of scale, to digital. It's going to take years [for digital] to reach that same level of proficiency and profitability."
Backing Into the Future
As recently as 1998, Kodak still enjoyed a gross profit margin of 45.6%, meaning that when the direct costs of producing its products and services were subtracted from each dollar of revenue, the company had nearly 50 cents left to cover corporate overhead and other expenses of operating a worldwide photography business.
Today, its gross profit margin had dropped to 32.2% of sales, and Carp says that the margin will fall to 31% when digital photography becomes the bulk of its business.
Digital cameras earn razor-thin margins. Ofoto earns no margin at all. Kodak has yet to show how it will make money in the digital era from its 13 processing labs scattered throughout the U.S., and more than 55,000 digital printing kiosks around the world. Which is why Carp is counting on other digital businesses to begin pouring in profit, while the company waits to see if it can repeat its "pole position" in snapshots.
One of those businesses is capturing and processing images for hospitals and other health-care providers. In 2003, that business had sales of $2.43 billion, up 7% from 2002. More telling, the business unit returned a profit of $481 million last year.
The other digital hope is commercial printing. In January, Kodak completed a $250 million acquisition of Scitex Digital Printing of Dayton, Ohio. Scitex customers use digital processes to print utility bills, banking and credit card statements, invoices, financial statements and marketing materials.
Kodak's overall printing sales also rose 7% in 2003, to $1.6 billion, and returned a profit of $166 million. Scitex, now known as Kodak Versamark, had sales of $157 million in 2002.
"Health imaging and commercial printing are more important for Kodak in the long run than its consumer business, digital or analog," says IDC's Chute. "Those are two areas that aren't seeing the margin compression that its consumer business is experiencing. That's the real future for Kodak."
That future still includes pain. In January, after it cut its shareholders' dividends, Kodak announced it would take up to $1.7 billion in one-time restructuring charges over the next three years. That included about $900 million in severance payments to cover the elimination of up to 15,000 jobs, and the shuttering of about one-third of its manufacturing facilities.
The company promises shareholders that by 2006, it will be solidly profitable in an era when its customers hold the means of production almost as much as it does.
"All together, the digital portfolio grew at a collective 48%,'' Carp said in July, after Kodak released its second-quarter results. That, he said, is "considerably ahead of our expectations.''
If it can make a success of its common picture standards effort, Kodak could find itself at the center of a profitable model for big volumes of prints from digital photography.
If not, Kodak's model may well be Polaroid. The pioneer of instant photography fell into bankruptcy in late 2001, after having failed to figure out that digital photography was a faster, easier and more reliable means of staying in the business that once counted it as a major player.
Projects: Enterprise Planning 
