"Only an 8? Not a 10 today?" Gamed HealthCare Surveys

What’s going on with Providence’s healthcare surveys?

Does Providence’s CEO Rod Hochman want to execute a strategy based on gamed survey data? Seems unlikely.

Or is Providence Health & Services unaware that it collects deeply flawed survey data? Seems more likely.

At Interaction Metrics, our mission is to collect science-based facts, so I can’t understand why any organization would seek to game its surveys. Still, Providence may have its reasons.

At any rate, here’s what happened:

Recently, my friend, John, had a call with Providence.

At the end of the 25-minute interaction, the employee asked, “So, how would you rate your satisfaction with the Providence healthcare system today? The scale is one to 10, with 10 being the highest.”

John pondered for a moment and then said, “Hmm…let’s say eight.”

The employee on the line probed. “Only an eight? Not a 10 today?”

John stuck to his guns. “No, let’s stick to eight,” he replied.

“Uh…ok,” she replied before concluding the call.

Best Practices for HealthCare Surveys

I think it’s safe to say this exchange was uncomfortable for both parties, so let’s rewind. How did this routine interaction turn into a ratings negotiation?

And what is the quality of Providence’s satisfaction data if its employees are “survey begging” or encouraging patients to >increase their satisfaction scores?

Again, this may be part of a business strategy that I don’t have insight into. And I think Providence is a great healthcare system overall. So this is emphatically not a knock on Providence, just a big question mark about its patient surveys and data collection methods.

This Phone Call: 3 Main Problems

First, the employee administered the survey while on the line with John. Given social pressure, most of us will feel obligated to provide a high rating. Succumbing to social pressure is a well-documented phenomenon.
Second, the question is leading: “How would you rate your satisfaction?” implies that the respondent must be somewhat satisfied. And a scale of 1 to 10 doesn’t give them the option to select 0 if they aren’t satisfied at all.
Third, the employee actively tried to increase John’s score. Do employees at Providence regularly try to convince patients to change their satisfaction scores? If this is how satisfaction data is generated at Providence, it bears no resemblance to reality. What’s the point of that?

Data quality aside, negotiations like these are just downright uncomfortable and paint healthcare surveys in a bad light.

John stuck to his guns and didn’t increase his rating. But most of us wouldn’t be so impervious. When companies ask for feedback, they must take measures to ensure the data isn’t manipulated.

The Takeaway Applies to ALL Companies

When employees realize their compensation, bonuses, or shifts are tied to ratings, they’re highly motivated to manipulate survey data. The term survey-begging aptly captures what’s going on.

Unfortunately, survey gaming like this is prevalent. I’ve had car dealers, appliance installers, retailers, and all manner of companies game their surveys with me. I bet you have too.

This feels particularly egregious with healthcare surveys—because healthcare is supposed to be evidence-based and science-driven.

But let’s face it: survey-begging is bad for all of us who want pleasant interactions. Moreover, it’s bad for organizations that use this data to build their strategy.

Drilling into the experience just a bit more… copious research shows that how interactions end is what we remember, which is why you should do everything possible to avoid ending on awkward or unpleasant notes. Certainly, finishing on a forced survey qualifies as an unpleasant note.

How to Stop the Survey Game

So, what’s the solution? Assuming Providence wants to collect meaningful data, how can it do better? And what are some takeaways for all healthcare surveys?

Start by double-checking your survey for any leading constructs. If someone is unhappy with their experience, you want to hear about it.
Next, let individuals complete surveys after the interaction has closed and the employee is absent. A follow-up email or text several hours after the exchange has been completed is ideal.
Finally, if you want your satisfaction data to tell an honest story, don’t allow employees to coach respondents on their answers.

Again, perhaps Providence has a strategy and a reason for asking their front-line representatives to manipulate data.

Still, it’s much more likely they’re making the same mistakes with their healthcare surveys that countless other organizations make.

Besides, even if Providence does have a strategy up its sleeve, their method makes for awfully off-putting interactions.

Wondering how to game-proof your surveys? Get in touch today.

Categories: Customer Satisfaction Surveys