Most businesses answer this with a survey, get a set of pleasant responses, and learn almost nothing.
The problem is not the survey. It is that people are unreliable narrators of their own behaviour, especially when asked politely, in writing, by a business they like.
There are better methods and they are cheaper.
Why are surveys so weak?
Three reasons that compound.
People answer the question they wish they had been asked. Ask what matters most when choosing a supplier and almost everyone says quality and reliability. Nobody says price, and price moves most decisions.
Only certain people respond. Your happiest customers and your angriest ones. The large middle, which is where most of your revenue sits, ignores it.
Hypotheticals are worthless. Would you buy this. Would you pay more for that. People genuinely do not know, and their answers correlate poorly with what they later do.
Surveys are useful for one thing: measuring something you already understand, across a lot of people. They are poor at discovering what you do not know.
What works better?
Talking to eight people for twenty minutes each.
Not a focus group, individual conversations. Recent customers, ideally, and one or two who chose somebody else. Twenty minutes, on a call, asking about what actually happened rather than what they think in general.
Eight sounds too few and it is not. The same three themes will come up by the fifth conversation, and by the eighth you will be hearing repeats. That is the signal that you have found the pattern.
What do I ask?
Ask about the last time, not about preferences.
“Walk me through what happened when you decided to sort this out.” Where were they, what triggered it, what did they do first. This is where you find out that half your customers started on a phone at ten at night, which changes your website more than any opinion would.
“Who else did you look at?” You will learn who you are actually competing with, and it is frequently not who you assumed. Often it is doing nothing.
“What nearly stopped you?” The single most valuable question. Every answer is an objection your marketing should be handling and probably is not.
“What were you worried about after you paid?” Tells you where reassurance is missing in your process, which is usually cheap to fix.
Notice that all four are about a specific past event. That is the whole trick.
Will customers actually agree to this?
More readily than people expect, if you ask properly.
Say you are trying to do the job better, say it will take twenty minutes, and do not turn it into a sales call. Most people are pleased to be asked. Recent customers are especially willing, because the experience is fresh and they already like you enough to have paid.
The ones who chose somebody else are harder and more valuable. Some will not reply. The ones who do will tell you exactly what lost you the work, which is information you can otherwise never buy.
What else is worth looking at?
Three sources you already have and probably ignore.
Your own enquiry history. Every question anyone has asked before buying, collected in one place, is a map of what your website fails to answer.
Your competitors’ reviews. Not their marketing, their reviews, especially the three star ones. People are far more specific about what disappointed them than about what pleased them, and it is free.
What people actually search for. Search tools show real phrases in real volumes. That is behaviour rather than opinion, and the language people use is usually plainer than the language the industry uses.
How do I know when I have enough?
When you stop being surprised.
If the next two conversations tell you nothing new, you have the pattern. Adding more interviews at that point buys confidence rather than information, and confidence is cheaper to buy by acting and watching what happens.
Research has a point of diminishing returns and it arrives much sooner than the research industry suggests.
What do I do with the answers?
Change one thing, and watch.
The most common failure is not the research, it is that the findings go into a document and the website stays the same. Eight interviews that produce one changed headline are worth more than a comprehensive report nobody acts on.
Pick the objection that came up most, address it directly on the page where it matters, and see whether enquiries change. That is the whole loop, and it can run in a fortnight.
What would I do this week?
Call three recent customers and ask what nearly stopped them.
That is an hour of your time and it will tell you more than a quarter of guessing. Most business owners have never once asked that question, and every one who does comes back with something they did not know.
We do short, focused research that ends in a decision rather than a report. Tell us what you are working on.
