Customers aren't lying. Nobody asks again.
A restaurant hears “delicious”. An app hears “easy to use”. But those customers may never come back. There are three reasons, and none is the customer's fault. You can't fix the first two. You can fix the third.
They don't want to hurt you
The owner is right there. Who wants to say the food was bad, or the app is hard?
Can't fixTelling you gets them nothing
There's no reason to spend time explaining. A short answer is easier.
Can't fixNobody asks again
They say “fine”, and that's the end. The real reason never gets a chance to come out.
You can fix thisQuestions that reach the real reason differ in three ways
Use them at the counter, on the phone or in chat. The conversations below are made up.
Ask what already happened, not opinions
“Do you like it?” “Would you buy it?” Most people say yes, because saying it costs nothing. And people are bad at guessing their own future. Things that already happened are harder to make up.
If we had this dessert, would you order it?
Sure, sounds good.
The last time you wanted something sweet after lunch, where did you buy it?
The shop by my office. Yours closes before I finish work.
Don't put the answer in the question
If a question can be answered yes or no, customers pick the polite answer. Ask so they have to tell you, not choose.
Was the service OK?
It was OK.
Compared with the place you usually go, what's different?
My usual place remembers I don't want coriander. Here I have to say it every time.
Follow up every time you get a short answer
This is the hardest one, because the person asking feels shy too. But the real reason, and things you never thought to ask, usually show up here. If you still get “nothing really”, ask again from another angle.
How was everything?
Fine.
Fine in what way? How was it last time you came, from when you walked in?
I waited a bit long. I was in a hurry that day, so I took it to go.
The milkshake and the question that changed
- Clayton Christensen's team once helped a fast-food chain that wanted to sell more milkshakes.
- First they asked customers what to improve. They heard: thicker, more chocolatey. The chain did all of it, and sales didn't move.
- So the team watched in the store, and changed the question to something that had just happened.
- The answer: a long, boring drive to work. People wanted something to hold in one hand, not messy, that lasts the whole trip. If not a milkshake, then a banana or a donut.
- So the milkshake's competitor was not another shop. It was bananas, donuts and boredom in the car. “What should we improve?” could never find that.
As told by Clayton Christensen, Competing Against Luck (2016)
What should we improve?
“What did you buy it for this morning?”
Five things to watch for
The ways above work anywhere. Conversations in Thai have a few things of their own.
- Don't ask “So you mean …, right?”
Most people will say yes, even when it isn't right. Ask “What happened next?” and listen for whether the answer fits what you understood. You can repeat three or four of their words, but don't ask them to confirm.
- “ค่ะ” and “ครับ” don't mean yes
They mean “I'm listening”. Don't count them as an answer.
- On a vague answer, don't repeat the question
“เฉย ๆ”, “ก็โอเค”, “แล้วแต่”, “ไม่เป็นไร”. Asking the same question again feels like pressure. Change the angle, make it smaller, or ask about a real time it happened. After two angles, move on.
- If answers don't match, it's your own mishearing
“Sorry, I missed that. Did you buy it at the shop or order it?” Not “But you said something else before.” Making someone lose face can end the conversation.
- Wait a moment after they stop
Count three seconds in your head before the next question. Often the important part comes when they keep going on their own.
Check your questions once more
- One question at a time, never two in one sentence
- No questions that can be answered with just yes or no
- Ask about the last real time, not “usually” or “what if”
- Don't explain or defend your product while the customer talks
- Don't correct the customer or say they're wrong
- Know what each question is trying to learn
Questions
What's wrong with asking customers “Do you like it?”
Nothing, but you usually get a polite answer that doesn't say what they will really do. Ask about the last time they used or bought it, then ask why. You'll get something you can decide with.
The customer only says “fine”. What should I do?
Follow up on their own word: “Fine in what way?” or “How was it last time?” If you still get “nothing really”, change the angle. Don't repeat the same question.
How many customers should I talk to?
More important than the number: talk to the right people, ask the same questions, and follow up the same way with everyone. One customer who complains loudly is one story. Listen to the end, but don't change the whole shop for one voice. And if nothing repeats after you've talked to everyone, that's an answer too: the question was too broad, or you talked to the wrong group.
Can I use a survey instead?
Yes, if you want to know how many. But a survey can only answer what you already thought of. To learn why customers do what they do, you need to talk and follow up. A few conversations can tell you why, but not how many people think the same.