When businesses decide they want to understand their customers better, the first instinct is usually to send a survey. It’s quick, cheap, and you can reach lots of people at once.
But surveys and interviews answer very different questions. Pick the wrong one, and you’ll end up with plenty of data that doesn’t tell you what to do next.
Here’s how to choose.
New to customer research?
Start with our guide on how to understand your customers.
The short answer
Surveys tell you what and how many. “60% of customers find checkout difficult.”
Interviews tell you why. “Customers find checkout difficult because they can’t see the delivery fee until the last step, and it feels like a trick.”
Surveys measure. Interviews explain. Most businesses need both, but in the right order.
When to use a survey
Surveys work well when you already know what you’re looking for and want to measure it. Use one when you want to:
Check how widespread a problem is. You’ve heard that delivery is slow. How many customers feel the same?
Track satisfaction over time. Comparing scores month to month shows if things are improving.
Prioritise between known options. Which of these three features matters most to customers?
Collect basic information. Who your customers are, how they found you, how often they buy.
When surveys let you down
The biggest weakness of surveys is that you can only learn what you thought to ask. If the real problem is something you’ve never considered, a survey won’t reveal it.
Surveys also struggle with:
Vague answers. “The website could be better” doesn’t tell you what to fix.
Polite answers. People tend to rate things higher than they really feel. (More on this in how to get honest customer feedback.)
Low response rates. Often only your happiest or most frustrated customers reply, which skews the results.
No context. A 3 out of 5 rating tells you something’s wrong, but not what.
When to use customer interviews
Interviews are best when you don’t yet know what the problem is, or you know it exists but not why. Use them when:
Sales or sign-ups have dropped, and you don’t know why
You’re about to make a big change, like a redesign, new product or pricing update
Your survey results are confusing or don’t match what’s actually happening
You want to understand how customers make decisions, including what nearly stopped them
A good interview lets you ask “why?” and “can you show me?” These follow-ups are where the real insight lives. (Need help with what to ask? See our 15 customer interview questions that get honest answers.)
When interviews aren’t the right fit
Interviews have limits too. They’re not ideal for:
Proving how common something is. Five people mentioning an issue is a strong signal, but not a statistic.
Measuring change over time, like tracking satisfaction every quarter.
Very simple questions, like “What’s your preferred payment method?”
The best approach: use both, in the right order
The most useful research usually combines the two.
Option 1: Interviews first, then a survey
Talk to a handful of customers to discover what the real issues are. Then survey a larger group to see how common each issue is. This means your survey asks the right questions.
Option 2: Survey first, then interviews
If you already have survey data showing a problem, like low satisfaction with checkout, follow up with interviews to find out why.
Either way, each method fills the other’s gaps.
A quick example
A (fictional) online pet supplies shop sent a survey asking customers to rate their experience. Most gave 4 out of 5. Repeat purchases, though, were low.
They then spoke to six customers who hadn’t reordered. Four said the same thing: they’d forgotten which food they bought last time and couldn’t find their order history easily, so they bought from whichever shop came up first on Google.
The shop then added a follow-up survey question, “Was it easy to reorder?”, and found the issue affected far more customers than expected. A simple “buy again” button followed.
The survey spotted that something was off. The interviews found what. The second survey confirmed how big the problem was.
Quick decision guide
Ask yourself:
“Do I know what the problem is?”
No → Start with interviews.
Yes → A survey can measure it.“Do I need numbers to convince someone?”
Yes → Survey (ideally after a few interviews).“Do I need to know what to fix?”
Yes → Interviews.
Not sure which you need?
Choosing the right method, asking the right questions, and making sense of the answers takes time and experience. Getting it wrong can mean weeks spent on research that doesn’t lead anywhere.
At Kleaarity, we help you pick the right approach, then run it for you, speaking to your real customers and turning what we learn into clear, prioritised recommendations. Real customers. Real conversations. No AI-generated feedback.
Want a fresh look at where customers might be getting stuck?
Frequently asked questions
Are surveys or interviews better for small businesses?
For most small businesses, interviews are the better starting point. You’ll learn more from five good conversations than a survey with a handful of responses.
How many survey responses do I need?
It depends on your customer base, but aim for at least 50 to 100 responses before reading too much into percentages. With fewer, treat results as a rough guide.
Can I add open-ended questions to a survey instead of doing interviews?
Open questions help, but you still can’t ask follow-ups. Answers are usually short and lack the detail you’d get from a conversation.
How long does each method take?
A simple survey can be set up in a day and run for a week or two. A small interview study typically takes two to three weeks from planning to recommendations.
What about online reviews? Do they count as feedback?
Reviews are useful but skewed towards very happy or very unhappy customers. Treat them as clues to explore further, not the full picture.