Many of our clients have trouble striking the balance between the Art (their intuition and gut feels) and Science (analytical results and ROI) of their marketing plans.  As a rough starting point, here are the three most important times in which to ignore the analytics:

1) Damn the torpedos. We use this when there is a strategic imperative in the business that must be accomplished, regardless of cost of other impacts.  A recent and famous example is after BP’s oil spill- they bought tons of newspaper ROP with messaging about what they were doing to clean it up.  The near-term ROI on those ads, if they even bothered to measure it, must have been abysmal- though it was a reasonable thing to do.

2) Tomorrow won’t be like today. To paraphrase the mutual funds, you should ignore your analytics when “past performance is not an indicator of future results.”  If you are drastically changing your marketing, you should expect it to behave differently than it did when it was measured.  This can be groundbreaking creative, or even big shifts in execution.

3) Breaking New Ground. One key way to ensure an innovation’s failure is keep it secret.  When you’re launching a new product, or even a new campaign around an existing product, spend ahead of it.  The risk of overspending on a launch is often much smaller than the risk of underspending in this situation.