Interview with John Koenig

Background

Taos: Tell us a bit about your most recent professional experience.

John: Most recently I was the CIO at Lucasfilm. I spent 5 years there. I focused on two areas: One was infrastructure stabilization and the other was organizational maturity. There are about 2400 employees at any particular time within Lucasfilm and these employees are mostly spread over four primary campuses. My IT staff to support the organization was over 100. Before Lucas I was with Silicon Graphics. At that time Silicon Graphics was trying to build a direct-to-consumer web ordering system, much like Dell has, to lie on top of the SGI infrastructure. I played a significant role in this effort.

Taos: From the CIO perspective, in what way has the challenge of managing and maintaining your IT infrastructure changed in this new economy?

John: I think that there’s a lot more attention paid to the business benefit of any IT initiative. I think during the dotcom boom anything that was online was seen to be intrinsically important and worth investing in. Now senior management has come to the realization that that’s not necessarily true and so you really need to quantify the business benefits of a project and then make sure you deliver those benefits at the budget you proposed.

Taos: When you look at where strategic dollars need to be spent, over the course of the next two years, what do you see as the most compelling areas?

John: I think there are two prongs to that. One is a real focus on cost reduction in the areas that don’t have business value. That’s usually around maintenance contracts, day-to-day fire-fighting and system administration. You get there by consolidating your server farm, standardizing, and stabilizing your infrastructure so that you don’t have as many servers and you don’t have to spend as much time maintaining them.

The other prong of the fork is in adding to the business effectiveness. This means that you’re looking at the business processes in the company and really being smart about where you apply new technologies to make the business more productive. An example of this is VoIP. Voice over IP (VoIP) telephony is in itself not that big of a deal but if you leverage it with pervasive wireless networks, then all at once you have a lot of people, within a campus, mobile yet still having their desk phone functionality hanging from their belt. You’re not giving everybody cell phones with a cost per minute but you’re really allowing them to move around and get their job done and not come back to their desk to find that they have 30 voicemails.

Delivering on Business Effectiveness – Planning for Project Success

Taos: How do you go about scoping a project like the one you just described?

John: First I work with the executive sponsors and users to make sure the business drivers are clearly understood and articulated. Then I keep the end goal in mind for the overall scope of everything that is desired and work to break the project down into manageable chunks. There are several reasons to do this. First, you need the full picture in order to make sure that the underlying architecture/foundation on which you are building is capable of scaling to the big picture of what you are trying to achieve. That said you need to break out what’s most important and most urgent into the early phase so that you can clearly scope those aspects and gain funding to support them. In addition, you need to be able to show frequent deliverables along the way, so I tend not to have phases which last longer than 3 to 6 months. I approach chunking phases by asking “how little of a scope can I acquire in this round of the project and still have a meaningful deliverable at the end?”

Taos: Technologists often struggle with Project Management because it is difficult to estimate how long it will take them to come up with the right idea or solution to a difficult problem in the project. How do you allow for creativity or that stroke of genius which couldn’t be anticipated in the timeline for the deliverable, but which in the end created a much better outcome?

John: In most cases, I craft the project with a lot of discussion around user experience or business goal up front so that the genius element can be captured early. Where there is an element of “art” involved, I try to give time to early exploration with a loose timeframe. Then, I want to lock down the prototype or concept so we can start building a more tight scope around that. Once this has happened, any change must be carefully considered before it’s allowed. Scope creep is such a slippery slope that it can just blow up your project; it would have to take a really, really compelled flash of genius to go back and start that over again.

Leveraging an Outside Partner

Taos: When you look at a project like the VoIP project you mentioned earlier, is this the type of thing that is done once and therefore you’re going to engage a professional services group to leverage existing skill rather than build your own?

John: In the business enhancement area I think you almost always have to go outside to get that specialized knowledge about the leading edge technology. It’s rare that you’ll find that your staff had a chance to deal with both the day-to-day issues as well as research the newest technologies. Even so, researching the newest technologies isn’t the same as hands-on experience; they won’t get enough knowledge so that they aren’t going to fall into the traps that someone else has already discovered.

Taos: Do you sometimes choose to outsource only the scoping or the architecture, but pursue the execution with your internal team?

John: Yes, you need to find a firm that’s very experienced with the technology you are about to deploy and at least take advantage of that experience in the scoping and design phases, but often throughout the entire development and deployment. Then you can focus on the knowledge transfer at the conclusion of the project so that your staff is very comfortable with what they are maintaining.

Another important use of a professional services team is for benchmarking; assessing the current state of my infrastructure or a particular technology area against current best practices in the industry. This can only really be delivered from a professional services group.

Taos: When you determine to outsource a project, can you talk about the qualities you look for in a professional services company?

John: First, I look for people that are willing to understand the business. Every business has its own idiosyncrasies and special needs and before a professional service group can add value, they have to understand where the inflection points are in the business. Secondly, I look for a group that delivers on their commitments and has regular and high quality updates to me so that I can make sure the project is proceeding on target and that my management and I don’t get surprised.

Taos: Has your overall experience with services firms been positive?

John: It varies. There are a few firms that I stick with because we work quite well together. And then there are some others, especially the larger firms, where it’s pretty hit or miss. You might have a good experience and based on that good experience pretty much the entire team you are using gets promoted. Then you don’t see them on the next engagement and experience is not as positive.

Taos: So it depends on the team that you receive.

John: I think that there’s a sweet spot for a professional services firm as far as size is concerned. You get to establish a relationship with the individuals and it’s not that they’re serving an 18 month hitch in the bay area and then they’re off to Texas and then in New England.

Fixed Fee or Estimated Work Effort

Taos: When you are looking to engage with an outside partner to deliver a project, what are some of the elements that you use to determine whether or not the project can be done on a time and materials, SOW basis versus a fixed fee proposal basis?

John: It’s all dependent on how clearly you can describe the scope. If there is likely to be any honest disagreement on the definition of what done looks like, then you can’t fix fee it. At best, you can fix fee a very small phase of a project. To start it would be reasonable to expect someone to fix fee the initial generation of the specification document or the generation of the test plan or something along those lines. At the end of that phase they then look forward at the next phase to generate a fix fee bid on that and so on. Cambridge Technology Partners pioneered that five or six step project where every phase was a fix fee, but the phase ahead couldn’t be fixed until the current phase was completed.

Taos: Where it is not fixed fee, instead you are working from a time and materials SOW with estimated work effort, how do you budget for that? And then, how do you manage changes to that budget?

John: I think you always have to a have a good idea of what the project as a whole is going to cost. I drive to an expectation that it’s going to be approximately “x” and that the adjustments to that estimate happen every time we have a project status meeting. I can then set expectations with my management, including internal resources and external resources will not exceed a number and then they manage that expectation as things go along.

In fact, I find that with time and materials you have a little more flexibility rather than less because it will be easier to scale back if you need to, or accelerate development if you need to.

Managing your Partner

Taos: When you outsource a project to a partner, what do you need to do and what resources do you need to dedicate from your internal team to make sure that it’s successful?

John: My team absolutely needs to be a part of frequent project status meetings to make sure that the project plan is still accurate. Further, I need to be the interface between the development partner and the end users, making sure that the end users are stepping up to their commitment to provide review time and test time and things like that. You really don’t get to walk away from very much except the actually work of the project. In some cases, you’re project time and your communication time actually increase because you don’t have all the various back door channels with the developers as you would with in-house staff that normally alert you to a problem on the project.

Taos: Is there any rule of thumb that you use for the frequency and format of review meetings, milestones, and reporting?

John: I like to see project status meetings happen no less than every other week and as you get closer to the end, every week. In these, you need representation from, not only the partner but end users and in-house staff that are supporting the project. You need these multiple points of view about how a given facet is coming along. It can’t be me sitting across the table from the head of the partner because we just don’t have enough prospective from the other corners.

Taos: What else is important to you when you consider working with outside partners to deliver a project?

John: The final thing which, in fact, is the one area that sometimes gets overlooked is the knowledge transfer at the end of the project. The knowledge around what’s been done and how to support it has to be transferred from the development to the operation staff so they can continue to carry the project along seamlessly. This is critical to the project and can be the deciding factor in determining the experience you have working with a professional services provider.

Taos: Well, thank you, John. It’s always great talking with you.