Data v. Opinion: The Ultimate Battle

Data v. Opinion: The Ultimate Battle

One of the challenges that we commonly run into as Product Managers is the battle between opinions and data.  And though it would be nice to pretend that data always wins, and that there’s always truth in Jim Barksdale’s famous quote, “If we have data, let’s look at data. If all we have are opinions, let’s go with mine,” we know from practical experience that this is simply not the case.  Let’s talk about some common situations where data bends to opinion…

HiPPOS Still Exist

The simple fact is that HiPPOs (Highest Paid Person’s Opinion) still exist, and they aren’t always persuaded by data that differs from their own opinions and beliefs about the market, the company, or the product.  They may have preconceptions about how things should look, how they should work, or how things should be done.  They may focus on micromanaging rather than allowing teams the freedom to create and innovate.  They may have a vision that’s not clearly communicated until it’s too late.  But when they’re the ones who are signing your paycheck at the end of the day, it can be a challenge to determine the best ways to convince them that a solution other than the one they envision is not only feasible, but preferable.  Sometimes, no matter what the data says, we have to bend to the opinion of a HiPPO — it’s an unfortunate fact of life in many organizations.

Collection Impacts Usefulness

“Data-driven decisions” has become a massive buzzword in the product industry over the past few years, and while in general it’s certainly preferable to make decisions based on data over opinion, there are times when data can actually lead us to make bad decisions, where our opinion might have led us down a better path.  This is because not everyone truly understands how to collect statistically significant, actionable, and reliable data.  I can’t count the number of times I’ve seen or taken a product survey that’s obviously intended to draw out actionable data, but which is written in such a way that the underlying biases are terribly obvious.  It’s actually hard to obtain data that’s truly unbiased, unaffected by selection of customers, and that can be easily collated, analyzed, and presented in a rational manner.  Half-assing your data collection means that your conclusions from that data are going to be just that — half-assed.

Presentation Makes the Difference

The last mile of data-driven decisions is one that is often overlooked and underappreciated — until you get in front of a room of people who look at you with a blank stare and bewilderment in their eyes.  I’ve often heard people claim that “data speaks for itself”…which is so far from the truth that it’s laughable.  Data speaks when it’s presented in a manner that’s compelling to your audience, just like anything else.  You need to take time to create a compelling argument from your data, and ensure that when you’re presenting it that the end goal of your journey is so obvious as to be the natural, inevitable conclusion.  All backed by solid, statistically significant, and compelling data.
Back To Top