With the college football season in full swing, I’m having a harder time focusing. After all, the college basketball season is only about one month away. There’s nothing I love more than college basketball, and there are surprising correlation between measurement and the NCAA, specifically predicting the outcome of that tournament in March.

1. The past becomes context – Each season, the past year’s triumphs and losses become history. They become context. Your favorite team may have lost their best player, but based on other player’s performance, you’re still hopeful for success this year. And if your team barely won three games last year, then winning six this year will mean a winning season (and if the reverse? Not so much). Similarly, in measurement you want to use past campaigns and measurements as context to understand your success moving forward. If you have consistently driven 1,000 new leads or sales with each campaign, driving 5,000 with a new initiative will be a huge victory. Only driving 500 will be somewhat of a failure.
2. Nothing is set in stone – If you plan to reach out to a group of bloggers to drive conversations about a new product, then you’re measure of success is likely centered around the blogosphere. So what happens when a few active Facebook communities pick up the conversation and run with it? Your measures of success need to change to account for these unexpected changes. It’s the same way your predictions may completely change when your cross-town rival’s point guard takes a spill and is out for the rest of the season. Or when you suddenly lose your best defender and have some tough road games ahead. You have to roll with the punches and adapt for the situation at hand while keeping the ultimate goal in mind.
3. You need to be in it to win it – As much as I’d love to choose my Tar Heels to win the Big Dance every single year, there are (rare) years when it’s just not appropriate. If UNC doesn’t make it into the tournament, how can I realistically expect them to win? Along the same lines, if you don’t bring your measurement team or analyst to the table, how can you expect to properly measure? Unless you consider measurement and plan for it all along, you cannot hope to successfully show your value when all is said and done.
4. Predictions can’t be made in a vacuum – If you try to measure your success based solely on numbers directly related to your campaign, it will be hard to make your case. But if you bring in last year’s numbers, previous campaign records or even benchmark against your nearest competitors, it gives that needed context to prove your value. You can’t show success in a vacuum any more than you can predict your team will win the NCAA just because your star player scores 30 points per game. The trickiest part about predicting the winner of the Big Dance is taking into account all of the different variables, players and schedules. But if you only look at your own teams numbers, your bound to make shoddy calls.
5. Attribution is impossible – I said it before, and I’ll say it again: attribution is measurement’s biggest problem. Even if you find yourself in an ideal situation, it’s hard to be 100% sure that it was your PR campaign which caused that massive increase in sales. Similarly, even if you can predict which skill will be the most valuable or prevalent in the tourney, can you really attribute a win directly to that one skill? Think about it. Maybe Tyler Hansbrough does score a record number of points, if the defense also plays their best game of the season, can you give Psycho T all the credit?