How come it’s like driving a car?
Consider….
You have a target to reach : destination You have a benchmark : the starting point and the starting time You usually have more than one way to reach destination The fuel is your main resource You have incentives (arrive in time, not get arrested for over speeding, be fuel efficient)
How do you achieve all these goals?
Evaluating the allowed speed on the road and knowing the trip distance will give you an estimate when to start the trip. Keeping an eye on the speedometer will make sure your speed is within authorised limits. Monitoring the fuel gauge will tell you when a stop at the gas station is wise In other words, your success will depend in 2 families of factors :
Strategical decisions : correctly planning the departure time, choosing the right roads
Good execution : drive to the limit and not run out gas by monitoring Key Performance Indicators (KPIs)
Wait a minute, what are these “KPIs”?
A good KPI is an indicator that helps you understand how your plan is executing and which should give you indications where and how to act if things go off track.
Example: You are on a highway limited to 120 km/h and your KPI (the speedometer) shows that you are going 140 km/h. You act by releasing a bit the gas pedal and the KPI comes nicely down to 120 km/h – confirming that the action got the desired effect
Example: you see on your fuel gauge that the tank is reaching the red zone and you see a sign on the road telling you that the next station is 2 kms away. Most likely you will stop, buy some gas, start again with fuel gauge showing fuel tank : you wisely monitored the KPI (the fuel gauge) to allocate more resources before the shortage occured.
What does web analytics have to do with all of this?
In most cases, a part of the companies efforts to reach their goals uses a website. In that respect, the web presence is the car. In order to drive it successfully, you’ll need to monitor the KPIs provided by the Web analytics.
Building KPIs
Keeping with the car analogy : the car is, in fact, ignorant of the speed it is moving. Instead, most modern cars rely on the ABS sensors to provide the speed at which the wheels are turning. The sensor reports a number of impulses per second. The number of impulses per second is converted by the car’s computer into number of wheel turns, which are later converted into kilometres per hour.
You still following me?
Now why do we need these conversions? Why not report the number of impulsions via a digital display to the driver?
Simply because the driver will not be able to make sense of the figure. The speed limits are displayed in km/h, not in Impulses per second.
Back to the web analytics : most (if not all) web analytics tools provides the raw data, the impulses per second. Unless this data is “digested” into KPIs, it is mostly useless.
Let’s see how is really calculated the speed of the vehicle :
The analytic system (the sensor) provides a number of 500 impulses per second Which gives 500 * 3600 = 1 800 000 impulses per hour. 25 impulses equals one complete wheel revolution, so the turns per hour is 1 800 000/25 = 72000 wheel turns per hour If the wheel developed size is 1.5m, the speed of the vehicle is : (72 000*1.5)/1000 = 108 km/h
Looking back at this example :
Turning the raw value of 500 impulses/second into the KPI : 108 km/h required few things:
a) the car manufacturer knew that the driver is interested in km/h, not in impulses/s b) the number of impulses needed to make one wheel turn and the size of the developed wheel were known but are not reported
1) KPIs can not come from the bottom up. Management should always participate to the KPI creation. If you communicate in impulses per second (or bounce rate, or conversion rate, or page views) you’re not communicating in the management language which is usually following a widely understood such as profit and loss, revenue, sales or costs.
2) KPIs are usually a combination of on-line and off line data. It is the combination that make them meaningful
3) It is easier to manage a few KPIs than be flooded by them
Data are just summaries of thousands of stories – tell a few of those stories to help make the data meaningful.
