Home 
Links
 
Words
 
Family
 
SF Archive
  
Internet Archive 
FAQ 
E-mail me
 

Copyright © 2002 John Zipperer unless indicated otherwise.

From and copyright by Internet World:

Major League Baseball Outsources Its E-Commerce Structure
By John Zipperer

(6/24/02) Major League Baseball had a challenge: unify its multiple team sites and increase revenue from the sale of services and products. The response it came up with was partly organizational and partly an outsourcing strategy; together, they helped drive solid increases in visitors and online sales, as well as position it for international growth.

New York City-based MLB Advanced Media was set up by Major League Baseball in June 2000, the result of a determination by the 30 teams in the league to operate all of their team sites jointly and get the most out of the ability to market their teams and related products. Today, MLB Advanced Media can boast more than 50 million visitors a month, with fans that look for information about games past, present, and future; want to buy jerseys or other products; subscribe to audio of games; or use other features. The point of MLB.com and its network of team sites is not necessarily to convert everyone into a buyer, but to be a must-visit for anyone interested in baseball, and be the place they visit when they do want to buy.

MLB.com uses Digital River to host and run the back end for its sites. The initial agreement with Digital River from more than a year ago covered hosting the site and its stores, which carried thousands of different MLB items and subscriptions to audiocasts. That agreement was then extended in April 2002, to include fulfillment, customer service call center, a site aimed at fans in the U.S. armed forces, and e-marketing.

The decision making that led to that outsourcing was a fairly traditional build-or-buy choice. "I think the decision is, like anything else in life, if you need to have something that is truly customizable, and you can do it more cheaply and better, than you ought to build it," says Bob Bowman, president and CEO of MLB Advanced Media. "But if you're trying to get functionality and scale and extensibility and great performance, then you ought to buy. We bought this service because Digital River's been doing it for so long. They can, quite frankly, provide the service at a better cost than we could do it ourselves."

What Bowman's team at MLB Advanced Media doesn't outsource is its marketing management, and as a result, Digital River works with that team in its marketing efforts. "They're a great, demanding client," says Joel Ronning, Digital River's CEO. "They know what they want, and we work closely with them about continuing to grow this thing."

Digital River, based in suburban Minneapolis, designs, builds, and manages e-commerce sites for 32,000 companies worldwide. Ronning says it works in 27 different languages and 27 currencies, and fulfills product orders from 54 locations around the globe.

MLB Advanced Media uses the capabilities of its outsourced provider to let MLB focus on the content and creation of new features. The game plan — not inappropriate for the game of baseball — is for steady expansion. Bowman says the depth and breadth of MLB's sites, as well as the editorial independence of its reporting have surprised many initially skeptical fans. International expansion is also in the plan, as the site builds Japanese- and Spanish-language subsites to provide information and sell licensed products to, for example, Japanese fans of Seattle Mariners pitcher Ichiro Suzuki. Bowman says product fulfillment in foreign markets may well bring in other outsourcing partners, such as its use of Amazon.com last year for a project in Japan.

(See the August issue of Internet World magazine for a feature interview with MLB Advanced Media's Bob Bowman.)


Loudcloud Sells Managed Hosting Business to EDS, Changes Name
By John Zipperer

(6/24/02) In what Loudcloud Inc. founder and chairman Marc Andreessen called a "market-changing event," the Sunnyvale, Calif.-based company sold its managed hosting business to EDS . Plano, Texas-based EDS will buy the managed hosting service and enter into a licensing agreement for Loudcloud's Opsware IT-automation software. EDS will first make use of Opsware in its internal network, then extend it to its hosting customers.

With the loss of its managed hosting business, Loudcloud also changed its name to Opsware Inc. In the agreement announced July 17, EDS will acquire Loudcloud's hosting customers, and will also get about 140 of its employees when the deal closes, which is expected in September.

What does this mean for enterprises? Those that are EDS customers can expect to see that company deploy Opsware to automate service delivery and management of applications in its 50,000 servers located in 14 major data centers, and 140 other data centers worldwide. EDS will also be looking to extend use of Opsware to new customers. As for the newly named Opsware company, it will focus solely on developing and selling its software, which it developed and used over the past few years in its Loudcloud business. Its main targets will be Global 2000 companies and government agencies.

Andreessen says that makes the software a proven benefit to customers. "We think this is a huge win for our managed services customers and our Opsware technology," he told journalists and analysts in a conference call. He says customers will like it because they will continue to get the power of Opsware in running their applications, but now they'll get the experience of EDS. "When we first started Loudcloud, we said we wanted it to be the EDS of the Internet," says Andreeseen, "and now EDS is the EDS of the Internet."
Jeff Kelly, president of EDS' global hosting services, says the Loudcloud deal "does reinforce our leadership position in hosting and Internet services." As for the Opsware software, "We think the tools give us advantages and automate many of the processes we do manually today."

Andreessen claims that this event changes the market dynamics in outsourcing. "One of the core things we're doing is marrying Silicon Valley innovation with one of the largest and most respected global services firms in the world," he says.


Strategic Outlook
True Grid
A Look at What Parts of Grid Computing Are Ready for Prime-Time Implementation and What Parts Are Definitely Not

By John Zipperer

(06/01/02) Make the work easier. That's the promise of grid computing, and that's what Cognigen Corp. wanted. Cognigen performs data analysis of large data collections for pharmaceutical companies as they go through the FDA approval process for new drugs. It had been using two Sun 450 servers, on which it could run four jobs. The scientists using the system could not queue their jobs, so they were constantly logging in to see if there were open CPUs-something that each of them was spending up to an hour a day doing.

As a system, it was slow, inefficient, and difficult to use. "Jobs could run anywhere from five minutes to 10 hours," says Duncan Ross, director of IT at Cognigen. "It was a pain." The solution was to download Sun Microsystems' Grid Engine, a distributed resource management tool, get the scientists up and running on it within eight hours, and replace the 450s with 12 Sun Netra X1 servers (which is where the Grid Engine runs). The new servers were cheaper than the ones they replaced, and the new system allowed scientists to queue their jobs and check them from anywhere in the company, as well as receive notifications when they were completed. "Now we can focus them on real work," says Ross.

Cognigen's experience is the perfect example of the state of grid computing today. Grid became popularized by schemes to use the unemployed processing power of connected machines around the world, but today it's mainly reprseneted by the cluster grid systems deployed within enterprises to better utilize available processing power.

The big players in the commercial grid computing field should not surprise anyone. After all, Sun executives have publicly declared it to be one of their priorities in 2002 to establish grid computing in targeted markets. But much of the real work in grid computing is still being done in the research labs and their academic projects [see "The Galactic Whirl-pool," February 2002, p. 14]. And it is there that potential users need to look to see how this technology will affect them.

Defining the Grid's Power
There are two faces to grid computing: distributed processing power, and distributed data storage and access. Mark Canepa, Sun's executive vice president for storage, says grid computing is just one part of a multi-faceted approach to storage and computing. "Grid computing has a role," Canepa told journalists earlier this year. "It is one of the ways our customers are going to optimize their application deployment, but it isn't the only one."

Hewlett-Packard Co. addresses the data storage-and-access problem with its Utility Data Center (UDC) offering, which is a mix of software, hardware, and services. Designed to make the data center more efficient and able to handle spikes in demand while not wasting the power of unused computers during slower times, the UDC can increase utilization of data center resources from 35 percent to 75 percent, according to HP. It is essentially a load-management platform for server clusters; resources from those servers are allocated as needed for incoming requests.

In April, HP announced that the UDC could be connected to a global grid, bringing it into contact with the greater-and potentially more revolutionary-aspect of grid computing, where the computing power is not just outsourced to an efficient data center but actually makes use of vastly distributed computing power from what is in effect a global computer.

"The grid computing market today-if you understand grid as the research community understands it, which is really grid computing over the Internet-is not yet a real market," says Wolfgang Gentzsch, Sun's director of grid computing. Gentzsch, speaking to Internet World from outside Munich, Germany, where Sun's core grid team is located, says his customers are asking what they can realistically expect grid to do for them today.

He says there are similarities between the Internet grid and the smaller cluster grids currently spreading through enterprises: distributed clusters, lots of data and processing, and a need for rapid remote access that is secure. Sun's grid solution is a distributed resource engine; users submit jobs to a master, without knowing the detailed components of the network, just specifying the job requirements and letting the engine execute it.

Gentzsch's central-European location isn't an accident. The grid-engine project began in 1993 with funding from the German government and the European Commission. After a "significant rewrite" in 1996-97, the project had produced what Gentzsch calls a mature product. In 2000, Sun acquired it. Sun's enterprise edition of the product allows users to plan the utilization of resources, instead of simply submitting to the engine.

Rival IBM has been building out its grid strategy based on a view of server-based applications that link together a network of servers, which work together as a supercomputer. "We're convinced at IBM that grid computing and distributed computing are the next phase of Internet communications," says Michael R. Nelson, IBM's director of Internet technology and strategy. The company is investing more than $1 billion over the next three years on grid technology, and it plans to build a services infrastructure on top of the grid revolution, positioning itself as a provider of outsourced computing power.

There's the ritual elbowing between the commercial vendors. Sun's Gentzsch suggests customers start small with an industry-proven software, which is what Sun got with its grid engine acquisition. Nelson plays the open-standards card, saying that Sun and HP talk about grid computing, "but if you look closely, you see that a lot of the things they're doing are not really based on open standards."

One enterprise that took the IBM route is the University of Pennsylvania's National Scalable Cluster Lab, which has been working for a couple years on a project for the rapid storage and retrieval of medical data, particularly mammograms. Currently, these images are stored on film, which is then warehoused. The project will allow doctors to take advantage of digital technology that will give them deeper analysis from the images, as well as better patient histories and comparative images. "Digital-mammogram machines are now available and approved and will probably come into fairly widespread use, if not for the difficulty of storing the information that comes out," says Robert Hollebeek, director of the lab.

Hospitals aren't set up to handle the huge storage required, so the lab entered into a partnership with IBM to develop a system that could handle petabytes of data and make it available to hundreds of hospitals. IBM is providing the initial hardware as well as experts on databases, storage, Linux, and other areas. "The vision here is that you can have computers as they are now-with a little bit of engineering and system integration, we think there are no big showstoppers to building this sort of thing-with an ability to provide almost plug-and-play really high-quality digital storage analysis and retrieval," says Hollebeek.

Security Under Construction
But what about the global grid? "We find this somewhat irritating," says Gentzsch. "There is so much hype, and no company can make use of this type of global grid computing. There is lots of research still going on."

One important aspect of that research is happening at the University of Wisconsin-Madison, where Barton Miller, a professor of computer science, is looking into the security vulnerabilities of such a mass-distributed system. He says there are three security problems if a user sends files and data out to a grid system that's not between dedicated, known users. First, there's the possibility of someone grabbing hold of the program while it's out running on another person's machine on the grid; second, someone might be able to modify the program subtly, changing the calculations it produces; third, the data is vulnerable to sniffing or stealing while it's out running on machines that are not under the user's control. The first two threats can be handled through redundant versions of the job running on the grid or other methods. But "that third one we don't have solutions for," says Miller.

Miller also suggests that the firewall will have to adapt to an environment in which data and whole programs are flowing back and forth over the grid. There are thus plenty of security issues involved with grid computing to keep security executives insecure.


Internet Whirl
On the Record
Sometimes Technology Business People Say the Darndest Things

By John Zipperer

(06/01/02) Journalists get to talk to a lot of people, which means, first, that people tell us lots of stuff and, second, that we hear all kinds of things that don't end up in the published articles. Sometimes they don't tell us anything that isn't in the well-prepared PowerPoint slides they're showing us. But other people are more willing to share some opinions or wisecracks with us. On the roughly one-year anniversary of this column, I have selected some quotes culled from my notes from interviews, press events, and trade shows over the last year. Some are amusing, some thought-provoking, some just odd.

"The Internet is a full-employment act for trademark lawyers."
-Dale Cendali, partner, O'Melveny & Myers LLP

"You do not build a business by reading Sun Tzu's The Art of War."
-Steve Mills, senior vice president and group executive, IBM

"You know how when you're giving a presentation, you're supposed to imagine your audience in their underwear? When you present to a Web seminar, half of them are."
-Overheard in the halls of the RSA Conference 2002

"This has been like a Hundred Year's War we've been working on in databases."
-Janet Perna, general manager, data management solutions, IBM

"People loved it, but the U.K. press went all over it. We almost had to shut the office down for three days because we couldn't get any work done with all of the press interest."
-Manuel Funk, managing partner, Fork Unstable Media, on his Berlin-based company's poor-taste Princess Diana driving game based on trying to avoid hitting objects while driving through a tunnel

"If you swim with the sharks, do not act like food."
-Michio Kaku, physics professor, City University of New York, on Internet business strategy

"This is where we release the poison gas and replace you with an android."
-Jay Adelson, CTO and cofounder, Equinix, after leading me into a corridor with interlocking doors on each side

"You know we're not all living by the same ethics. This is a very fragile infrastructure that we're building a lot of interesting things on. And there are a lot of crazies out there."
-Christopher Klaus, founder and CTO, Internet Security Systems, speaking at Microsoft's Trusted Computing Forum 2001

"I think it's so stupid. No artist would ever lose money on [Napster]. They [the artists] behave like record companies."
-Marian Gold, lead singer, Alphaville, on other musicians attacking the file-sharing site.

"I don't see us making a lot of money on this. I'm trying to figure out how I as a business take advantage of these paradigm shifts and how I [Sun] am going to make money."
-John Shoemaker, former executive vice president and general manager, Sun Microsystems, on Sun cofounder Bill Joy's JXTA P2P project

"The attitude is that if I can do things critical to my day-to-day operation using other products, what the hell is the problem with security?"
-Becky Base, technical advisor, IntruVert, discussing users' disappointment with security products

"I like to say that our customers don't have a sense of humor."
-Kay Hammer, founder and CEO, Evolutionary Technologies International

"You wouldn't solve the cholera problem by treating each person individually, you'd do it by fixing the water. When you're trying to deal with corporate-wide security problems, you [need to] treat your corporation like a community."
-Peter S. Tippett, vice chairman and chief technologist, TruSecure

"This was a bitter, resentful screed. Obviously Mr. Zipperer has an axe to grind when it comes to white people, men, the successful, the wealthy, the well-educated, etc."
-Reader responding to my critical review of a video celebrating the party-and-networking ethos of the dot-com companies


Profile
The R&D Imperative
Agilent's Tom Saponas Knows the Importance of Aggressive R&D Programs

By John Zipperer

(06/01/02) His boss likes to hang out in his lab. He gets to direct far-thinking research strategies that make plans years and even decades into the future. So envy Tom Saponas, director of Agilent Laboratories and CTO for parent company Agilent Technologies, who has a job that is both critical to his company's success and immensely satisfying.
Agilent, spun off from Hewlett-Packard in 1999, serves the communications and life sciences industries, both of which have voracious technology appetites. To be in charge of an R&D effort charged with feeding that hunger is a heady position, but Saponas welcomes it.

He illustrates the importance of R&D to business success by pointing to a seemingly small change with huge consequences. "When I joined the company, things were solved primarily with analog circuits. Now, almost everything has a transducer that gets it into digital form, and then we operate on that with some form of embedded computer," he says. "At the time I joined HP, some very small percentage of employees were doing software, where now probably half of our people are writing software. That has probably more than any other factor increased the rate of change of products, because software is easier to change than hardware. The rate that we have to turn over our product is much shorter than it used to be. We're now at the point where about two-thirds of revenue comes from products introduced in the last two years. If you're that dependent on the flow of new products, you can imagine how essential R&D is. You either do it or you go out of business."

His company works hard to keep a steady flow of ideas and research through its brain trust. It does this by holding working groups, hosting a series of guest speakers, and encouraging cross-pollination of ideas between its researchers in different areas. It even has a Halloween event called Tech or Treat, in which researchers visit-sometimes in costume-demos put together by each of the laboratories.

Saponas has experience of his own in a broader experiment in cross-pollination. In 1986, he served as a White House Fellow, a program in which experts from the private sector serve in advisory roles to government leaders. As a special assistant to the secretary of the Navy, his work earned him a distinguished civil servant award. It also convinced him of the importance of the sharing of ideas between government and business leaders, something he says he hopes the Enron scandal hasn't made impossible.


Internet World Interview
CRM Consulting
Going Global
PWC Consulting's Global CRM Leader Talks About What Consulting Firms Bring to the Table

By John Zipperer

(06/01/02) Adam klaber sees customer relationship management (CRM) in global terms. He has to; Klaber, a PwC Consulting partner, is the firm's global leader of its CRM group. Therefore, he has been in a good position to see what his customers want in the CRM space-what they are trying to achieve, how they are going about doing it, and what kinds of services and relationships they want from their consultancy.

PwC Consulting, a part of the PricewaterhouseCoopers family, has more than 35,000 consultants on the ground in 70 countries. Those consultants work with more than half of the Fortune Global 500. As PwC's global CRM leader, Klaber has responsibility for $1 billion in revenue. Previously, he had served as the firm's CRM Americas leader, where he increased its North American CRM staff from 35 people to more than 2,000. He clearly sees it as a growth opportunity both for his company and for his clients.

Klaber earned his MBA from Columbia University and joined PwC in 1986. But Klaber's professional experience isn't confined to the boardroom. From 1983 to 1986, he was a systems engineer at Electronic Data Systems, which we discussed with the New York-based executive.

Internet World: You started out as a systems engineer with EDS.
Adam Klaber: My background started out in core technology, where I was doing programming and analysis work-which I think is a great background. If you work in the business-integration area, which is using technology to help transform how companies operate, if you've never been in the guts of making systems work-staying up late at night and running conversions and supporting production environments-it makes it hard to effectively implement solutions. I joined Pricewaterhouse to get more of a business focus around what we did.

And I can tell you our CRM story. About six years ago, I was working with a client on a large ERP project, where we were really putting in the infrastructure to help order management, distribution, logistics-a lot of things that hopefully were going to enable better sales. What we found out is we were ignoring the sales and marketing professionals as to how do we get information to them, how do we support them, how do we make them more customer-facing? And then from that point, we started to get into some of the CRM process and technology areas. At that time, we had about 20 or 30 people working on it around the world. And now I think we've got about three thousand. So it's been a pretty wild growth area for us.

iw: You and [PwC consulting partner] Dan Hirschbuehler wrote an article discussing the goal companies have for CRM and the fact that so many of them are admitting they're not there yet. Why not?
ak: Well, at one end I don't think companies will ever be really, truly done with being as focused on their customers as they could be. It's hard to think that anyone ever gets it done perfectly. I think what's really happened is that by the advent of some new technologies, you could create the environment of a small store where everybody knows your name, if you will, and apply that to a series of technologies across multiple channels for large organizations and across B2B environments as well as B2C environments. So then this name came up: CRM. And I think the challenge has been that everyone viewed it as the Holy Grail, a silver bullet; if I go do CRM, I become more customer-focused. A positive was that you can do pieces of CRM relatively quickly-very different than ERP, for example, where it takes a long time to get the ERP components together. You can do rapid implementations of pieces of CRM. You don't have CRM, but you can at least say you've implemented it.

Particularly the software and technology vendors focusing on CRM would sell the broad customer-centricity as a simplistic little thing you can do, and there's been a gap between expectations of the business buyers within the marketplace today. And that's where we-PwC Consulting-focus our CRM efforts: helping our clients understand what it is, what can be done realistically, setting correct expectations, and creating implementation and change programs that have rapid deliverables that move you toward an overall CRM vision that isn't just purely laissez faire, but they're certainly not the two- to three-year mega-projects because they don't get done. They don't deliver the value.

iw: They don't deliver the value, or the businesses just are not willing to spend that much time waiting for the project?
ak: No, I think it's easier today to say, "Hey, why should we; it's just too much money." I think the reality is, even when it was a healthier market, we were not recommending those programs because they get too complicated. And people get weary of, "When are we going to start seeing value?" Again, the good thing is that you can implement pieces relatively quickly.

iw: Is there a particular chunk of that they should start with first?
ak: There are certainly multiple places to start. When you deal with anything across process channels, process business units, process geographies, what we look to do through our CRM visioning work is identify what's some of the low-hanging fruit that will add some business value that you can deploy relatively rapidly. And it differs by industry; it differs by company. Some companies have some contact center that is very sophisticated in capability that you can quickly do some optimization on; with others, it's around a Web site and a Web interaction. And some of the more straightforward ones are very quick to implement, and some of the marketing campaign management over the Web, e-mail-based marketing, things of that nature we can do pretty rapidly.

iw: The benefit to a company of doing this is improved customer relations, hopefully building up loyalty. You talk about moving some of the less valuable customers into more valuable positions over time through good service to them.
ak: Or lower-cost channels, right?

iw: Yes. Is there really a danger of moving customers, because where are they going to go? Are they going to go to another company that also doesn't have good CRM, or are there really folks out there who —
ak: — Sure. I was dealing with that issue with a client today. I guess there are a couple points. One is, if you think of the vision of CRM, which is that "I want to have excellent lifetime value customer service across channels, I want to up-sell and cross-sell at each touch point, I want a personalized experience, and to do that for all of my customer base," there's a high probability that the cost of doing that could put you out of business. What you need to do is understand who your customer segments are and how to treat different customers in different segments. So as a telecommunications company, I may create this very elaborate customer service model. I know who you are, I may try to up-sell you, I may try to cross-sell you. But at the end of the day, if all that customer is doing is shopping on price and is not appreciating the value of any of these value-added components, it's better for that customer to leave. And in the telecommunications industry, or the credit card industry, there are some very good examples where you have a high churn rate with customers.

And the real trick to CRM is not treating everyone well; it's treating the right people well based on their profit profile to you. I can really fall in love with a vision, but you need to realize that that vision may deliver great service, but it may not be at the right price point. Today it's important to understand that the e-economy boom, which was just get customers, acquire and maintain them at any cost, is gone. And so if people haven't adjusted their CRM visioning to really leverage that cost point by different segments, I think they're going to be in trouble.

iw: That attitude of pulling in customers en masse-was that really a widespread thing in traditional companies, or is that just more reflective of the dot-coms?
ak: It's always hard to generalize. But think about the telecommunications market. They just need as many customers as they can get. I think it very much was a B2C phenomenon.

iw: At what point do clients tend to come to you as a consultancy? And at what point do you think they should be coming to you?
ak: It's a good question. The best time for them to come to a business integrator like ourselves is when they realize they want to do something to make some improvements and set a vision and a plan. And then we can, in a very independent and objective manner, help define the plan, the components, the time frames, set the expectations correctly. About half our business is driven through that area, through that mechanism.
The other half of our business is when a company wants to do something — whether it be a product, new contact center, a new Web site — they have some view of what the technologies are going to be, and they want someone to help implement it. And the challenge there is that expectations have already been set, and it's harder sometimes for us to make sure the programs are successful because the key to a successful program is meeting or beating expectations.

iw: Do you find yourself in the position where you have to change expectations? Can you change them?
ak: You know, I think that in today's business environment, I think in anything but certainly accentuated within CRM, change is an inevitable event-spinning off divisions, re-estimating of revenue growth, profit growth; there's always change in an enterprise. I think our role as the integrator is to help define what is constant, what may be variable, and implement action plans as well as contingency plans around what may happen. I think we're getting better. I think that the market is understanding and appreciating what can be done within CRM a little more today than it was, certainly, a year or two ago. It was a lot more confused; it was a lot more confused when everyone was talking about e-business.

I'll give you a good example. I was working with a client, and they asked me to do a CRM vision. We were within the first week of the six-week project, and we found out there was another consulting firm doing an e-business vision at the same time. All of e-business is not CRM, but certainly a big chunk of CRM is e-business. So how do you do these divisions concurrently? So that's what we recommended to the client: You've got to shut one of these things down.

iw: And please make it the other one.
ak: Yeah.

iw: Where does the initiative in a company tend to come from to go to a consultant?
ak: I think it comes from a couple places. It comes from the people responsible for the sales, marketing, and service areas that are looking for change. Sometimes they're re-engineering heads or change agents that are trying to do something. Sometimes it's CIOs who are seeing that there are many CRM proposals being presented to them, but there isn't an enterprise-wide plan. So they're going to step forward and help bring the different people together. Because in most companies, it's kind of hard to have a chief customer officer. They have many different functional areas that focus on customer issues, so it often comes out of someone who's in charge of those change programs or the process areas.

iw: Is there any way of suggesting a certain base of knowledge that a company should have and beyond which they would need the expert folks?
ak: One of the keys to a successful CRM project is during the implementation portion of the project, the transfer of knowledge from the business integrators, the business consultants, the technology consultants, to the internal teams. It's absolutely critical that that happen so that they can be self-sufficient. Now, there is a push to outsource all of the ongoing operations, and that's a discussion. But for the companies that want to support this, it's critical that they be part of the program, be part of the implementation plan as they move forward. And so if we see a project team, and 90 percent of the people are from PwC Consulting and only a few from the client, you're going to have a very hard time transitioning post-implementation. It's a much longer term relationship, including helping assist in the operations versus just helping in the change to go live.
You know, a situation where the client is just handing it to us to do on our own is very risky, because the thing you can't outsource is the knowledge of your own internal organization.

iw: Do you find your government clients have different needs than a commercial company in the way they want to work with you?
ak: In all the government agencies, the government area has a very formalized set of procurement vehicles and time frames, and certainly from our perspective, never does anything particularly fast. There are many legal requirements around how things are bid, how technology is selected, and that takes longer than within a private organization.

iw: Government now is very concerned about the security aspect. Is it generally talking about security of the privacy of information, or the security of its systems from being used for malicious purposes?
ak: There are two pieces. One, from a homeland-security perspective, is how do I really understand who our citizens are, whether it be when they travel, whether it be immigration components? There are some great examples of some of the Sept. 11 terrorists still getting direct mail, still being contacted. How do you close the loop on all those things? And then the privacy piece relates more on the commercial side of, say, if I don't want people sending me e-mail messages, how do I get off their list? I don't want people calling me up at home. How do I get off the list? I don't want to share information about myself with a company because I want it to be private.

iw: Are enterprises going to be using the same technologies and approaches with their internal customers as they do with their external customers?
ak: Well, that's the concept of these employee portals, where in many ways you can break out CRM technologies today. Some are used by employees to better sell or serve companies. So there's a call-center application or a sales-force automation application. Others are actually used by end customers so when you go on a Web site to acquire or you go on a call center you use an IVR for self-service, right? Some of the applications are used by customers; some are used by employees. What people are now thinking is, well, gee, if my sales force has all this information, maybe it would be good to expand that knowledge so my executives and people from other departments have that information. The whole enterprise becomes more customer-focused. So we're starting to see that as a benefit to having these applications.

iw: With packaged CRM solutions, does that put you in a position where you'll be working in conjunction, or you'll be bringing in the services staff of the
particular packaged vendor?
ak: Sure, all the time. We feel strongly that if you're implementing a packaged solution, it's critical to engage that software provider through the duration of the implementation. That may be at a very small level of quality reviews or expert advice during the project. It may be having a couple members of their own services organizations taking part in the implementation. In some cases it may be subcontracting, billing of any product functionality gaps directly to them. It really depends upon how complete the products are of that software vendor that you're implementin and how well the business integrator knows its operation.

iw: And when you're working with them, they are, in essence, reporting to you?
ak: Multiple scenarios work. Some clients like us to be in charge, and they're subbing to us. Other times they like to have the software vendor in charge who subs to us. I don't think there's any definitive answer there.