Go and read the top of your sales page right now. Then go and read the last three messages a member sent you. If those two things sound like they were written about different products, you are not bad at marketing. You are just guessing, and you have a room full of people who could tell you the answer if you asked them properly.
Member interviews are the cheapest research a creator can run. No tools, no budget, no audience size requirement. Thirty minutes on a call with somebody who already pays you will teach you more about your positioning than a month of studying competitors. And yet almost nobody does it, partly because it feels awkward and partly because the version most of us try is genuinely useless.
So let us fix the useless version.
Why your own words are the problem
You have been inside your topic for years. You use vocabulary your members do not use. You lead with the sophisticated benefit because the obvious one feels embarrassingly basic to you. You describe the transformation you find interesting rather than the one people are shopping for.
This is not a personal failing, it is just proximity. A furniture maker talks about joinery. The customer wants a table that does not wobble. Both are describing the same object. Only one of those sentences sells anything.
Interviews solve this by handing you language you would never have written. Not better language, exactly. Truer language. The phrase a member uses to describe her problem at eleven at night is the phrase that will make a stranger stop scrolling, because it is the phrase already running in the stranger's head.
Who to talk to, and it is not who you think
Most creators interview their favorites. The enthusiastic member who comments on everything, who calls the community life changing, who would happily talk for an hour. That person is lovely and will teach you almost nothing, because they have already fully adopted your worldview and your vocabulary. They will hand your own marketing back to you.
You want three groups, and the ratio matters. Talk to people who joined in the last thirty days, because they still remember what it felt like to be undecided, and that memory evaporates fast. Talk to people who have been around a year or more, because they can tell you what actually kept them, which is almost never what got them. And talk to people who left, if any of them will speak to you, because they are the only ones who can describe the gap between what you promised and what showed up.
Six to eight conversations is usually enough for the patterns to become embarrassingly obvious. You do not need a sample size. You need to stop guessing.
How to ask without making it weird
The invitation is where most attempts die, because creators oversell it. A long message about how much their input would mean, how you are on a journey together, how this will shape the future of the community. That reads as a favor request, and busy people decline favors.
Short and specific works better. Tell them you are trying to describe the community more accurately to people who have not found it yet, that you would like twenty five minutes to ask about their experience, and that you are asking a handful of members rather than running a survey. Offer two concrete time slots. Say plainly that there is nothing to prepare.
Do not offer money. It quietly changes the conversation into a transaction and people start performing helpfulness. If you want to give something afterwards, make it small and unrelated, and mention it only once the call is over.
Then tell them what will happen with what they say, before you start. You are going to use some of this to describe the community publicly. You will check with them before quoting anything. If they would rather stay anonymous, that is completely fine and you will honor it without asking why. That takes fifteen seconds and it changes how honestly people speak, because they know the rules.
The questions that actually work
Throw out anything that asks people to speculate. What features would you like, would you pay more for this, would you recommend us, how likely are you to renew. People are terrible at predicting their own behavior and wonderful at describing their own history. So ask about history.
Start with the moment before. What was going on in your life the week you joined. Not why did you join, which invites a tidy after the fact explanation. The week you joined. What was happening.
Then ask what they had already tried. This one is gold. The list of things somebody attempted before finding you is a map of your real competition, and it is usually not another membership. It is a book, a free video, asking a friend, or doing nothing while feeling guilty about it.
Ask what nearly stopped them from joining. Every single person has an answer. Price, time, a fear of looking foolish in front of strangers, a bad experience with a previous group. That list becomes the objections section of your sales page, written by the people who had the objections.
Ask what the first week was like, in detail. Where did they go first. What confused them. Did anyone speak to them. This is the most operationally useful answer you will get and it will probably sting a little.
Ask what they would tell a friend who was considering it. Note that this is different from asking for a testimonial, because you are asking them to imagine talking to a peer rather than to a camera. The language gets dramatically plainer.
Then ask the closing question, which is my favorite in any interview: what have I not asked about that I should have. Wait through the silence. Something like a third of the time, the best thing in the whole conversation arrives after that question.
The rule that makes all of this work: be quiet
You are going to want to explain. A member will describe a feature of your community incorrectly, or say something mildly unfair, or misremember how the pricing works, and every cell in your body will want to correct the record. Do not. You are not there to be understood. You are there to find out what people believe.
When they finish an answer, count to three before you speak. It feels unbearable and it is where the real material lives, because people tend to give a polite answer first and an honest one second. If you fill the gap, you get only the polite one.
The only follow up you need most of the time is a version of tell me more about that, or a plain repetition of the word they just used, turned into a question. If they say the community felt safe, say safe. Then stop. They will define it for you, and their definition is the copy.
What to do with the recording
Ask permission to record, every time, in words, at the start of the call. If they say no, take notes instead and do not make it awkward.
Then treat that recording like the private material it is. Store it somewhere only you can reach, do not paste it into random summarizing tools, do not drop clips into a group chat, and delete it once you have pulled out what you need. A member telling you about a rough year in her life is not content. Some of what you hear should never be used for anything, and knowing the difference is most of the job.
When you go through it, do not write a summary. Copy exact sentences into a document, word for word, including the ums and the sideways grammar. Sort them into buckets: the state they were in before, what they had tried, what nearly stopped them, what surprised them, what keeps them. After six interviews the same five sentences will show up in different mouths. Those are your headlines.
Turning it into copy without putting words in anyone's mouth
Here is the line I would not cross. You can use a member's language as raw material to write in plainer, truer terms about what you offer. You cannot attribute a polished sentence to a person who did not say it.
So there are two separate outputs. The first is your own writing, sharpened by what you heard, and that needs no approval from anyone. The second is quotations, which need explicit sign off on the exact wording, with the person's name shown exactly as they want it shown. Send them the quote, offer to trim it, accept a no without negotiation.
Never invent a testimonial. Never lightly improve one. Never combine two people into one convenient voice and present it as real. Beyond the obvious ethical problem in the United States and most other markets, made up praise is oddly easy to spot, and getting caught costs you the one thing that took years to build.
Also resist the urge to sand the quotes smooth. Slightly awkward, specific praise is far more persuasive than a perfect sentence, because it sounds like a person instead of a brochure.
When to do this
Twice a year is plenty, and once a year beats never. Good times to run a round: before you change your pricing, before you rewrite your sales page, after a quarter where churn ticked up, or when you catch yourself describing your community differently in two places and no longer know which one is true.
You should also do it when things are going well. That is the counterintuitive one. Interviews during a good stretch tell you what to protect, and protecting the right thing is how a community survives its own growth.
Do this this week
Pick three names. One person who joined in the last month, one who has been with you over a year, and one who left. Send all three a short note today, offering two specific time slots, and say plainly that there is nothing to prepare.
Write your seven questions on a single page so you are not improvising, and put a note at the top that just says be quiet. Ask permission to record, tell them how you will use what they say, and then let them talk while you take almost no part in it.
Afterwards, copy their exact sentences into a document. Do not paraphrase. Do not tidy. Just collect the words, and then go look at the top of your sales page again.
You will probably wince. That wince is the most valuable feeling in marketing, because it means you finally know something you did not know yesterday, and now you get to go fix it.
