Old Habits Die Hard: Status Quo Bias in the NewMR
- June 25, 2018
They say, “Old habits die hard.” My dad gave me a real-life understanding of this the other day that has implications for how we operate in the NewMR.
After a nice lunch together, my dad mentioned he needed to go to the bank to withdraw some cash.
As we walked up to the bank and I headed to the ATM, but he kept moving in a different direction – into the bank’s main entrance. I asked my dad if he was taking out more than the daily limit; he looked at me like I was speaking Aramaic and responded with a simple, “I need to get cash.” He headed to the teller, pulled out his checkbook and wrote the amount to “cash.”
I let this go on in amazement. It was as though I stepped out of my time-traveling Delorean to watch an ancient ritual that no longer exists. Once he was done, I asked why on Earth he wouldn’t just go to an ATM. He replied, “Because I’m used to doing it this way.” I tried to convince him how outdated and inefficient that was. To my surprise, nothing I said seemed to have any impact on him.
Then it dawned on me – I remembered a lesson from my cognitive psych classes at USC. My dad, like many of us, was falling victim to a common bias that we all face across a wide range of scenarios – the status quo bias.
Sticking with what we know
Status quo bias is a bias towards the current state. When facing a decision, people tend to be biased toward doing nothing or maintaining their current or previous decision. Furthermore, we compare what’s new with what is familiar and what we feel comfortable with.
In research, this often manifests in a resistance to adopting new approaches.
Sound like a stretch? These arguments may sound familiar:
- Social data/online conversations aren’t representative.
- We’re only sampling from/talking to the same group of people.
- We only have behavioral data for a portion of our sample – it’s not representative.
In other words, people often stick with what they are already doing. They compare new ways of conducting research to the most easy-to-envision example of what might have been and react accordingly. This is why, for example, many struggle with deep listening social analytics, passively metered data or online community research. They can’t help but compare it to what they already know and are most comfortable with – which often includes representative, or quota based online surveys – and marginalize the insights.
In order to drive innovation and explore new ways of thinking, we have to focus on, and help our organizations focus on, what social or other new methods are (rather than what they aren’t or might otherwise be) and most importantly, their individual benefits.
So for example:
- Do we think positive online conversations have a positive impact on our brand?
- Do we think it’s good to have a longitudinal understanding of our core audience? Then why is it a problem if we talk to engaged category/brand users over a long period of time?
- Do we think understanding online behavior or purchase behavior will help us better understand consumer journeys and path to purchase? Then why does it matter if it’s only a subset of our sample?
This bias isn’t isolated to how people talk within their companies about these new methods/data sources; conversations between client and vendor are filled with bias. They set the stage for status quo bias. When we develop RFPs or discuss them, we are often asked to incorporate new methods as “optional add-ons” or separate things to consider. Don’t get me wrong, budgets are budgets, but RFPs should not be created around methods (“I need a focus group”; “I need a 20 minute online survey.”) The most important thing is to find the right tool, or combination of tools, to address the business issue.
So… if you are trying to drive change within your organization, it will be important that you be aware of any incoming bias and focus on the benefits. Avoid comparing new methods to the easiest-to-imagine alternative. Easiest isn’t always best. Status quo bias can be an implicit barrier to driving positive change.