Driving Results Through EA: A Racer’s Perspective

driving successful enterprise architecture

Racing involves the bringing together of driver, car, technique, and support to achieve desired results. Enterprise architecture brings together people, processes, methods, and tooling to achieve desired results for your stakeholder. Good drivers, like all good athletes, have a natural skill, and yet they are also smart and/ or humble enough to know that there are known techniques they must practice if they are to be proficient. There’s nothing like a day at the track to validate (or invalidate) your design decisions and driving techniques. Likewise enterprise architects, to remain relevant and competitive, must also continually hone their skills and keep abreast of emerging trends and regulations that impact our profession. I love racing because it brings together automotive engineering, driver skill, and a support team with measurable metrics (e.g., lap times). I’m also passionate about enterprise architecture as it brings together business and information technology with demonstrable impact on mission effectiveness. Like racing a car, enterprise architecture is not something you can do “by the book.” While you can learn many of the basics of EA by reading, you need hands-on experience to be an effective enterprise architect.

YOU’RE RACING AGAINST YOURSELF

In racing, we tell drivers to “forget the competition and focus on your own performance.” Novice drivers tend to focus too much on what their competitors are doing. Each EA program is different with respect to funding, scope, and purpose. Concentrate on getting the most out of your particular EA program. It’s important to establish your own metrics for success. Having easy-tomeasure and meaningful metrics is key to process improvements. Unlike racing, EA does not lend itself to simplistic performance measures (e.g., lap times). EA programs differ with respect to purpose and hence differ with respect to meaningful metrics. One of the most important discussions to have with your stakeholders is to clearly establish and quantify the desired outcome for their EA program.

Modern race drivers are highly instrumented; even low-budget racers have in-car video, multichannel data logging, and GPS. Like racers, enterprise architects need instrumentation to check the impact of changes or improvements to the EA program. With the advent of modern EA repositories and modeling tools, EAs can be highly instrumented. EA repositories can be configured to generate reports and dashboards of relevant metrics. As in racing, EAs should be willing and able to make course adjustments based on sound feedback. Each track (business scenario or architecture segment) is different, and you’ll need to adjust your approach for a given scenario.

RACING IS A TEAM SPORT

Whether you are a race car driver or working as an enterprise architect, you have to see what you want, and believe that you can do it. You must have confidence in your team, and if you are working alone, you must have confidence in yourself. If you have this mentality, you can get what you want in any area. With enterprise architecture, there needs to be recognition of each team member’s contribution to the success of the program. Small, focused EA teams working toward a common goal are more successful than a loosely knit collection of independent consultants.

driving successful enterprise architecture

SUCCESS RELIES ON MULTIPLE FACTORS

Optimization of a single component doesn’t translate into winning (e.g., brakes, engine, suspension). Too often novices focus on simply dropping a huge engine into a car. The result is a car that they can’t handle, overheats, burns up brakes, and spins out into the dirt in corners. Every change you do must result in tangible and measurable improvements. Successful EA programs focus holistically across all EA domains (strategy, architecture, capital planning, and governance). Successful EA programs have defined metrics that can be measured. Narrowly focusing on tools (EA repositories or modeling) or making changes without understanding their impact to EA success metrics is a recipe for failure. Novice racers may also fall into the trap of attempting to optimize across all domains. We often tell them speed, reliability, cost—pick two. Likewise in EA. It often makes sense to pick a single architecture segment or business area rather than attempt to address the entire enterprise.

REACH OUT TO SEASONED EXPERTS

There’s nothing like seat time next to a national champion to hone your skills and validate your approach. At a recent track event, I was able to ride along with Jeff Winters (National Nissan Champion) and learn some of his techniques. It’s important to bring in recognized subject matter experts into your EA program to independently validate your approach.

A few minutes of an expert’s time will often save you weeks of rework. Examine other EA programs and seek out best practices and techniques your team can “borrow” for your stakeholder. Like drivers, you’ll find most enterprise architects enjoy talking about their EA program and are more than happy to share lessons learned.

IF YOU CAN’T AFFORD THE PROPER SAFETY EQUIPMENT, YOU CAN’T AFFORD TO RACE

No one would argue that racing is a dangerous sport and that using proper safety equipment and adherence to regulations is important. Prior to going on the track, every car must pass a safety inspection, and all drivers must attend a basic overview of track rules and regulations. EA governance is akin to the role of regulations in racing, and security is akin to safety equipment. Security, like safety, must be seamlessly integrated into your EA. EA governance processes should be established early and agreed upon by, and clearly communicated to, all EA stakeholders.

MAKE INCREMENTAL ADJUSTMENTS

Small improvements can equal big victories. Avoid making drastic changes in an EA program. Strive for continual improvement. EA dashboards and trend analyses are excellent tools within the EA domain to track incremental improvements. Try not to make multiple adjustments simultaneously—it becomes unclear what is actually contributing to performance improvement. As in racing, EA teams should encourage experimentation and reasonable risk taking with an understanding that some initiatives may fail. That requires a safe environment from a cultural perspective but also requires regular feedback from stakeholders.

ACKNOWLEDGE YOUR SPONSORS

Racing is an expensive sport, and having sponsorship (tires, suspension components, track-side support) is critical. Stakeholder involvement and recognition is the most critical success factor for any EA engagement. Actively work with your stakeholders and showcase your accomplishments. Consider staffing an EA communications role. Watch any experienced racer and you’ll notice that his car and uniforms are covered with sponsor logos. During any press event, the racer will thank his sponsors. Take advantage of opportunities to showcase your stakeholders’ success. All EA programs should have some sort of EA website that communicates and promotes the stakeholders’ EA agenda. Understand that your EA program’s success is ultimately about successfully advancing the agenda of your stakeholders. Understand your stakeholders and to whom they are beholden. You need to ensure they are viewed as successful in the eyes of their superiors; otherwise, your EA program will be scrapped.

BRAKE LATE, ACCELERATE EARLY, CUT CORNERS

Being aggressive wins races. If your car feels like it’s “cornering on rails,” you’re likely driving too slowly. To be aggressive, you need confidence in your car, experience in the seat, and a competent support team. But winning racers are not reckless. It’s often said that “the slower you move, the faster the car moves.” This means that if you’re constantly correcting your driving angle and speed, it’s going to affect your lap times. Likewise with EA, you need to get your stakeholders to buy in to a strategic objective and avoid being consumed by day-to-day minutiae. We tell drivers to “look and think as far ahead as possible.” To win at EA, you need to focus on the road ahead and where your stakeholders want their EA program to go. Stepping up to challenges on an EA program and occasionally being the “hero” is all part of a successful EA program. Build in excess capacity into your team to handle ad-hoc requests from your stakeholders. Like racing, EA engagements can have the thrill of victory and the agony of defeat. As with racing, your EA program may not go exactly as planned. You need to plan for the occasional missed shift and that inevitable slide off the track into the grass.

About Dr. Fred Collins 1 Article
Dr. Fred Collins is an IBM chief executive architect within Global Business Services. He's had a lifelong affair with automotive racing and restoration and fields two 40-year-old Datsun 240Z cars in Colorado. He leads IBM's Federal EA Practice and is one of IBM's global EA instructors. Collins supports both the US Department of Agriculture and Department of Interior as chief architect. He has led or supported EA engagements throughout the public sector including engagements at the Department of Energy, Farm Services Agency, Forest Service, Bureau of Land Management, and the Department of Defense.