Most brands that spend a lot on advertising have some kind of a tracking study in place. And more often than not, use of the findings are restricted to checking whether the marketing team performed better or worse in the last week / month / quarter.

But once in a while a team of good researchers (or, as it happened with me, great researchers!) come along and teach others to do better.

Here's what I learnt:
1. Find what is generic to your category, and what drives the niches
A tracking study has a lot of image parameters and not all of them are equally important in the market. So how do we know which ones are the critical ones?

Take a dump data of fairly long period (say six months for a weekly tracking study) and run a Jaccard analysis on the image parameters and brand preference as the two sets of variables. What you will get is a score for each image parameter. These scores indicate how strongly the image parameters drive brand preference. Sort them in order and look at both the ends of the spectrum.

At one end will be the ones that drive preference the strongest across brands. These are the category generics. Your brand can't score low on these. At the opposite end, are the ones that drive preference for one or few brands, but not for others. These are the image parameters driving the niches. Pretty good data to look at while making brand portfolio decisions.

2. Track how consumer preferences change over time
Even the most boring categories will see changes in consumer preferences over time. To check this, just take two dumps of data over, say, two years and compare scores for image parameters. You can look at average score across brands, or average brands chosen for each image statement, etc. and figure out if certain factors are becoming more important over time.
There's a catch here. In some categories, these changes are a result (and not cause) of communication. For example, if deodorant brands talk about no-gas deos, it may lead to consumers asking for no-gas deos over time. Hence, discretion is advised. :)

3. Measure the lead time of communication on brand scores
The tracking study already tells you how much impact your communication has on the brands scores for you and competition. It can also give you an idea how much time it takes for your communication to show impact on the scores.

Take 2 trend lines - your advertising GRPs and one of your brand scores. Normally, we look at these trend lines as they happen in real life. But creating an overlap can provide a completely different analysis. What if you see the brand score with a one-week or a two-week lag? Maybe it shows a higher correlation between the two trend lines. This can be checked for promotions too. Then, you will theoretically have a better idea on when to begin your Diwali promotion, or back-to-school sale.

If you have already tried these or other such ideas, do share the hits and misses.
Because, well, it helps.

--
With thanks to Ipsita Arora for helping me with this post.