1. Quick indicator about whether the company is doing as expected
2. Gives people something to aim for
Let's take a look at those 2 things, quickly.
1. The problem with assuming that your KPI (target vs. achievement) indicates how your business is doing, is that it excuses you from understanding how your business is doing. If the KPI is what you think it should be, then you can walk away from it, without understanding the underlying reasons.
2. Almost nobody in the company has direct influence on your numbers. Sales people can only sell to people that want to buy. Some items stop being attractive during a recession. Some markets suffer (or grow) when the sun comes out. Since nobody can actually influence those things, it's silly to put the target there as something to aim for. People will do the best job they can, and if externalities remain relatively stable, you might hit those targets. But externalities are never predictably stable.
For example, let's take a comparison site which allows people to compare prices in a market which is controlled by a few vendors, many of whom are simply distributors for a single company. At the beginning of the fiscal year (we'll use a year for simplicity's sake), we set our annual targets. We don't know what's coming up over the next 12 months, but we make an educated guess and set our targets based on our assumptions.
What happens when several months into the year, the single company raises prices (due to their production lines being flooded, for example), vendors raise prices to maintain margins, and our site suddenly looks more attractive to users? We do better. Our business improves. We almost certainly meet our targets, and there's a good chance we exceed them. But how much of that is due to things we have control over? Did we improve the site? Perhaps. Did we make things more effective? Perhaps. But the single biggest change that affected us was something completely outside our control.
If you're only looking at the targets, it looks like we've done an awesome job. But since we can't separate the impact of the change in prices from the work we were doing anyway, it's impossible to say where credit is due (other than the majority being due to increasing prices).
Now, let's set targets for the next year. We review the previous year, and we do our best to keep in mind that the last year saw price increases. Prices have now stabilized (the flood damage has been taken care of, and new factories, in drier regions, have been opened). However, as the manager of a company, I don't think I can go back to my board and tell them that last year was a fluke. This year will be much worse. All expectations indicate that. We may improve our processes or site in such a way that we can increase business by 30%, if we're extremely lucky, but the odds of making significantly more than 2 years ago is really hard to see. So, if I'm a typical manager, I set targets where I think the board wants to see them, knowing that they'll be almost impossible to hit, without external intervention.A month into the year, it's clear the targets are set too high. 2 months in, and it's really clear.
Now we're in the position of knowing that our KPI is useless, but we've agreed to be held to our own numbers, so we feel stuck. This puts in the position of thinking about the targets, when we should be thinking about growing the business. And if you've ever read Drive, you'll understand that putting numbers in front of people on a regular basis increases their focus (on those numbers), but decreases creativity, productivity, and motivation. It also tends to make us risk averse, as we see bonuses, kudos, and promotions going down the drain. And lastly, it tends to kill our long term view. After all, if we don't meet short term goals, we won't be here to meet long term ones. The end result is that during a time where creativity and risk are most needed, they are least likely to be utilized.
In the end, by setting targets, we set ourselves up for failure.
What's the alternative?
If you can't bring yourself to get rid of targets, start by not holding people to them. They're an indicator of our ideas of success at the time of setting them, not a definition of it.
Even better, however, is removing them completely.
When running a price comparison site, there are lots of other ways you can indicate success:
1. Conversion. When people click out is it to buy, or to escape?
2. As you roll out new features, do your numbers improve? Do you get new traffic? Does a higher percentage of your existing traffic convert?
3. If you use SEO, are you ranking better? Is it driving up traffic? If it's driving up traffic, do those users tend to commit?
4. If you use PPC, are you getting value for money?
5. Are people finding it easy to get through your site? Is the average time per page going down, or up? Which direction should it be going (if you want commitment to your site, it should be going up. If you want people to come to your site and leave to buy something, it should be going down).
6. Are you capturing more and more of the people who left your site? Why did they leave? Have you resolved their problems?
7. How do you compare to the rest of the market? Are you leading? Following? Are you increasing your slice of the pie?
8. Can you increase the size of the market? Are you improving market penetration? Finding new markets? Finding new products? Experimenting?
Understanding your business cannot be done with a single number. It takes time, and effort. And managing by targets doesn't allow for either.