Targets are Slippery

It’s notoriously hard to come up with robust targets.

Many sites have monthly traffic targets based on Page Impressions and Unique Users. But is this the best measure? Does it really measure what you want? What i want to know is

  • How good is the product?
  • How loyal are the users?
  • How adaptable is the business in a rapidly changing environment?

Would the following be more a more useful measure?

  • Frequency of repeat visits. Combined with a minimum number of unique users per month.
  • Number of tweets, facebook likes, etc

Maybe these alternative measures could run in parallel to the traditional ones for comparison.

A real-life example

The following article highlights what i mean. “Andy Carroll to quit Liverpool for Manchester City in shock move“. A sensational headline. The article is about a van with the livery “Andy Carroll Tilers Ltd” parked near the Man City ground. What a let down. Not great for the product. Not great for user loyalty. But great for the monthly stats.

In fact, the target encourages such articles.

The comments for the article slated it. User comments criticising journalistic quality is fairly common (even on the BBC it seems). But it makes the stats for user comments look good!

An offline example

Economies target GDP (the size of the economy). It includes health services. More sick people means better stats! There are more excellent examples in the book “The Tiger That Isn’t“.

(I’m not saying that the government games the targets by deliberately making people sick, just pointing out how misleading targets can be).

Targets and rewards

When targets are linked to rewards (such as financial bonuses) things get even more difficult.

Research shows they can often backfire. The book “Irrationality” has two good chapters (8+9) on it.


Targets are a powerful tool but it’s easy for them to backfire. Tread carefully.

Leave a Reply

%d bloggers like this: