Bad Math and You

Activity Based Accounting, Part Two

The key to Activity-Based Accounting is never to separate the numbers from the activities of the people represented by the numbers. When numbers in a business lose their connection to people and their actions, the numbers are no longer trustworthy.

Never forget these 3 things:

1. Not all facts are helpful.

2. You lose sight of the big picture when you get too close.

3. It’s easy to tell lies in the language of numbers because most people believe numbers never lie.

Here are a couple of facts taken from the 2000 Census:

The average American family size is 3.14 persons.

The average number of children per household is .90 children.

Problem: Let’s say you want to move to a “family” town, a place where lots of people are married with children still living at home. You’ll need to find a city with more than nine-tenths of a child per household, right?

Using the logic of traditional Cost-Based Accounting, you’ve narrowed your search to 3 towns with 50% more children than the national average of .90 per household. Riverview, Prairieville and Mountaintop each have 1.35 children per household. On paper, the 3 towns look equal.

But Activity-Based Accounting would reject the 1.35 children per household average and look at the raw data behind the numbers. Here’s what Activity-Based Accounting would discover:

The people of Riverview dislike children. That’s why

90 percent of all Riverview households have no kids.

But Riverview has a Polygamous religious group, so

5 percent of Riverview families have 13 children each and

5 percent have 14 children each. Welcome to Riverview, where tension hangs thick in the air.

Prairieville is composed largely of immigrants from an overpopulated nation. Consequently, the people of Prairieville believe it’s immoral to have more than one child.

8 percent of Prairieville households have no children.

74 percent of Prairieville households have 1 child.

5 percent have 2 kids.

5 percent have 3 kids.

4 percent have 4 kids.

4 percent have 5 kids.

In Prairieville, 82 percent of the population looks down on the 18 percent with more than one child. “Breeder” families like yours are social outcasts. You and your 2 kids are just going to love it here.

The people of Mountaintop are happy to be alive. The town motto, “Live and Let Live,” is painted on the water towers and the police cars.

33 percent of all households in Mountaintop have no children.

22 percent have 1 child.

33 percent have 2 children.

7 percent have 3 children.

2 percent have 4 children.

1 percent has 5 children,

1 percent has 6 children,

1 percent has 7 children.

Did you notice in the raw data that none of the children were fractioned? In Activity-Based Accounting, any step that creates a fractional person is a false step.

Although it’s true that each of these 3 towns has 1.35 children per household, it’s a completely irrelevant fact and

(1.) Not all facts are helpful.

We took one step too many when we calculated the number of children in the average household. This illustrates the fact that

(2.) You lose sight of the big picture when you get too close.

I’m sure you would agree that we learned more about Prairieville, Riverview and Mountaintop from the raw data than from their identical averages of 1.35 children per household. Now that you've taken a step back from the misleading “average” and looked at the raw data, Mountaintop is obviously your best choice.

Advertising professionals, you realize I’m talking about Gross Rating Points, don’t you?

Reach (the number of different people being reached)

times Frequency (repetition)

equals Gross Impressions.

Gross Impressions expressed as a percentage of population equals Gross Rating Points. (One million Gross Impressions in a city of one million people equals 100 Gross Rating Points.)

The Cost-Based Accounting logic that created Gross Rating Points isn’t just plain stupid; it’s fancy stupid. (Stupid with raisins on it.)

An advertiser is considering 5 different plans. All he knows is that each plan delivers 100 Gross Rating Points. An Activity-Based study of the raw data tells us Plan One will reach 100 percent of the population once. Plan Two will reach 50 percent of the population twice. Plan Three reaches 10 percent of the population 10 times. Plan Four reaches 5 percent 20 times. Plan Five reaches 1 percent of the population 100 times. But with “100 Gross Rating Points” hovering before his eyes, there’s no way for the advertiser to see the varying effectiveness of these schedules because when he multiplied reach times frequency he took one step too many.

Gross Rating Points are valid only if you accept the false premise that reach and frequency are interchangeable.

Do you believe reach and frequency are interchangeable? If so, you’d be just as happy in Prairieville or Riverview as in Mountaintop. After all, they each have 50 percent more children than the average American town.

And real estate is so much cheaper in the first two towns than in Mountaintop! I wonder why? Oh well, it doesn’t matter. Because the numbers are the same.

And numbers never lie.

Roy H. Williams