On a scale of 1 to 10, how would you rate this article?

We've recently moved house. And along with a move comes a flurry of services, like broadband installation, washing machine delivery, and address changes for magazines and newspapers.

We relied more on takeout and delivery, too, thanks to glitches in getting all-things-kitchen set up properly (yep, there is a such a thing as too-minimalistic design: I'm not sure what that induction hob is saying through Beep Language, but je ne comprends pas.)

Which means my emails and texts look less like correspondence with my nears and dears and more like this:

• How would you rate your delivery?
• How was your experience at our restaurant?
• Help us improve!
• Rate your installation experience.
• We would like your feedback!
• We'd love to hear from you!
• Your feedback helps us improve.

I used to fill out all of those surveys. I am conscientious by nature, so appeals to "help improve the experience" resonate with me, especially when I've had a particularly good or particularly bad experience.

But it's too much. I get a request for feedback not just for every transaction I have with a vendor, but for every touchpoint. I mean, literally, "Here's our newsletter. What do you think of our newsletter?".

We are in full feedback fatigue.

I am all for more data to help companies do the right thing, to hone the right offering, to introduce the right products, to improve the customer experience. But, as with most things, to quote Mark Taylor, the way an experience is delivered is the experience. Two gentle suggestions:

One: Ration customer feedback requests to "why" questions

There's a lot you can learn from a customer by looking at their behaviour over time. That means there are some questions you can infer the answer to pretty accurately from behaviour, like whether they "enjoyed" the transaction (that choice of wording always makes me chuckle).

For example, do I come back often? Does my basket size and spend increase or decrease across time? If I'm a seasonal purchaser, say for school uniforms, my purchase data gives a pretty good indication of the ages of my children. So, if I've got at least five or six years of schooling left, am I coming back year after year?

Similarly, mine and combine ecosystem data. For example, was the order delivered on time? If drop-shipped from a third-party, did they send it on time? Right now, I can get asked for feedback at least three times for a single transaction: from the vendor itself on my satisfaction with the ordering process, by the parcel delivery service on my satisfaction with the delivery itself, and from the vendor (again) on my satisfaction with the product or service itself, often asking for a review, to boot.

When this behaviour profile is assembled (and yes it's harder, takes longer, and is more expensive to execute), you'll definitely be able to see where problems and opportunities are. Then it's time to ask "why" questions.

Two: When you do ask, keep it human.

Don't bait and switch me with a "how many stars would you rate xyz?" and then click me through to a 10-minute survey. Not cool.

Your completion rate will be dismally low, and the minor twinge of annoyance at being baited and switched will bias me toward a more negative rating (which, of course you can mitigate by factoring in response biases, but why deliberately annoy your customers in the first place)?

Nor should you use every channel at your disposal to ask my opinion: I've gotten the same request for feedback over email and via text. Just because you can be omni-channel doesn't mean you should. It becomes more noise. (And for those of you saying, "just set your communications setting appropriately," you are missing the point .)

--

All of this is to say, be thoughtful when asking for feedback. A few questions to start:

1. What are you trying to solve for?
2. What are the right metrics that help you know whether you're making progress solving that or not?
3. What are the ways you can get those metrics?
4. If getting those metrics involves creating (yet) another customer touchpoint, how are you designing that touchpoint's experience?

Customer feedback data is just a start: it has to be paired with human insight and a thorough understanding of the process—one that's not always entirely within the bounds of your own organisation—it's trying to measure.

Now, if you have a moment to spare, on a scale of 1-10, how would you rate this article?