Think like a Pilot to Build an Effective IT Dashboard

IT Editorial

Thursday, 21 August 2008 13:58

Written by Dustin McNabb

Once you’re off the ground and the wheels are up, you monitor systems, collect more data, and make necessary course corrections to keep things on track – all the while, handling constant interruptions from others in the plane and on the ground that have their own ideas about where and how you fly. 

Equate the concept of flying an aircraft to piloting a business and the similarities still apply – lots of data to analyze, lots of things to plan, lots of systems to monitor along the way to ensure a successful outcome. 

Interestingly, when flying an aircraft, you can strip away much of the fancy navigation, communication, weather monitoring systems and what you end up with are a few critical instruments that, even if everything else goes “belly-up”, will guide you to a safe landing.   This is especially true for the instrument-rated pilot who – usually because of bad weather -- depends upon instruments as the sole reference to arriving at a destination airport.  The same can be said in business – when dark clouds loom on the horizon the business executive has a core set of information that will guide him through the impending storm.  This is especially true for IT service dashboards which provide essential real-time performance and availability information regarding mission-critical applications that are essential to the business.  The trick is if you’re going to build effective dashboards to more accurately and effectively monitor your business or IT services, then you need to start thinking like a pilot.

When a pilot flies, there are essentially six basic instruments that provide critical information to keep the plane upright and traveling in the right direction – basically, the same thing you want to do with your business.  There’s an Attitude Indicator (AI) tells whether the nose of the plane is pointed up or down, left or right.  There’s an Altimeter to tell how high the aircraft is above the ground.  There’s a “vertical speed indicator” (VSI) to tell how fast the plane is going up or down – and there’s an Airspeed Indicator to tell how fast the plane is going through the air.  Finally, there’s the Directional Gyro (DG), which is essentially a fancy compass to tell the direction the aircraft is headed in.

For those that are still paying attention and noticed that I’ve mentioned only five instruments – there really is a sixth – it’s a poor man’s backup to the attitude indicator, but it only gives information about whether the plane is going left or right – there’s no up or down information in this humble holdover from flying days of long ago.  It’s located in the bottom left of the figure above.  You’ll notice that some pilots have replaced this instrument with a second AI – as shown here – redundancy in an airplane is a very good thing! 

For the past 100 years that we’ve been drilling holes in clouds with airplanes, the basic information that a pilot uses to keep the “blue side up”, so to speak, still comes from the same six instruments.  Sure, times have changed and “steam gauges” have been replaced with fancy glass screens with color displays driven by solid state electronics – but much of information needed to fly a plane still remains the same.

More importantly, while each flight instrument provides a single piece of information, piloting an aircraft from one place to another requires constant, accurate interpretation of information from all the instruments – only in this way is the true aircraft and flight status known. 

The same can be said for running a business – especially from an IT perspective.  Today’s IT executive dashboards have a similar set of “gauges” that provide a busy IT executive with an instant snapshot of business health – something akin to “am I open for business”?  And yet, no single piece of information will provide the whole story – what’s needed is constant, accurate interpretation of a real-time set of information to effectively understand the health and status of a mission-critical application or service.

Having implemented hundreds of IT dashboard projects for customers around the world, we’ve learned that there’s an equivalent “six pack” of information that lets an IT executive know at any given time whether he’s “open for business”.  For each mission-critical service or application, these include:

- Application availability
- IT component availability
- Application response time
- Application transaction volume
- IT alarm summary
- Trouble ticket summary

Application availability is the Attitude Indicator of the business world.  It’s an instantaneous measure of am I “up” or “down”?  Without it, it’s impossible to determine whether the all-important service itself is working for the business.  More importantly, when IT issues do occur, they need to be associated with a specific application – this allows IT to better prioritize which IT issues get worked on first. 

IT component availability is next on the primary instrument list.  You can think of this one like humble “turn and bank indicator”.  Its information you need to create trouble tickets, isolate and resolve IT component failures – but, just as in the cockpit, its information that must be used in conjunction with another instrument – in this case application availability. 

Application response time, like the airspeed indicator, is a great indication of how fast an application is responding to users.  However, speed is only a partial indicator of aircraft or business status.  Sure, if the airspeed slows, it’s probably safe to assume the plane is going up.  Alternatively, if the airspeed increases, then the plane is probably headed down.  Of course, a similar analogy holds for the business as well – if performance slows, then usage is probably up – and vice versa.  But, it’s important to remember that this isn’t always the case – because performance (and aircraft airspeed) is dependent upon a number of different factors.  You’ll need to reference one or more of the other “IT instruments” to get an accurate determination of response time issues. 

And then there’s transaction volume.  A little like the altitude indicator, transaction volume provides instant information on “how high am I?”, but it’s not going to give you an instant indication of application status.  Again, it’s one more piece of information necessary to help confirm the rest of the information you’re seeing.  In the cockpit, if the altitude indicator reads 10,000 feet and I really want to be at 9,000 feet, then I need to change my direction of flight downward using the airspeed and VSI indicator to help me descend at a safe speed.  Likewise, in business, it transaction volumes are too high, then I need to consult IT component availability and response time indicators to understand the effects of transaction volume – and whether I need to make course corrections to balance workload or risk exceeding system capacity.  Effective  business pilotage requires not only accurate application status understanding, it also requires effective in-flight course corrections – through load balancing or other means – to smooth out the rough spots and ensure safe flight within specific boundaries – not too fast, not too slow, not too high, and not too low. 

Alarm and trouble ticket summaries are the last of the primary IT flight instruments.  Each provides essential information to determine whether real or potential problems lurk in the IT infrastructure.  As well, they work together to provide essential information on what’s potentially broken as well as what’s currently being worked on.  Ensuring alarms and IT issues are effectively prioritized and remediated based on business impact is essential to keeping the business on “wings level” course. 

In addition to the primary business “flight instruments” there are four essential “gauges” that any IT executive should have handy to provide the essential trend information necessary to foresee impeding issues before they arise.  These instruments not only help to determine overall service or application health, they also provide key supporting information to primary business flight data.  For each mission-critical application, these include:

- SLA compliance trend
- Customer satisfaction trend
- Open ticket aging report
- IT outages trend

So, while aircraft cockpits and business dashboards have evolved over the years, there remains specific “primary” information that is necessary to each – ensuring the successful operation of both the aircraft and the business.  For companies looking to start a new dashboard project – or enhance an existing project underway – the data elements defined above should serve as a good starting point. 

Finally, it’s most important to remember, that even the most sophisticated aircraft or IT instrument panel won’t steer you clear of the bumps ahead unless you pay attention and carefully interpret provided information.  Proof of this occurred on July 31st when a single engine Lancair Legacy II crashed upside down a few hundred feet short of the runway at Oshkosh, Wisconsin – home of one of the world’s largest airplane gatherings during the last week of July.  In what appears to be a case of a pilot ignoring his airspeed the results were ultimate and final.  And while business decisions may not have the same “life or death” results as this example, the consequences of ignoring the key information that drives your business can be catastrophic indeed. 

So, let’s all be careful out there… gather, interpret, and effectively use the information that’s needed to fly your airplane and drive your business – and most of all… keep your wings level and the blue side up.

About the Author: Dustin McNabb has more than 20 years of experience as a pilot and in  IT management including holding executive management positions in marketing, development, and professional services for such leading companies as IBM, Fujitsu, Computer Associates and Merant. He is currently the vice president of marketing for Business Service Management vendor, Managed Objects.

TrackBack URI for this entry

Comments (0)

Subscribe to this comment's feed