What can I sell to my audience?

Why "what should I sell?" is the wrong first question

Search that question and you will find lists. Two hundred digital product ideas. Fifty things to sell in 2026. Templates, presets, ebooks, memberships, challenges, cohorts.

The lists are not wrong, exactly. They are just answering a question you do not have. A list tells you what can be sold. It cannot tell you what your people will buy, because it has never seen your comments section.

Here is the more useful framing. A product is a bet that a specific group of people have a specific problem, badly enough to pay to make it go away. You already have privileged access to the only evidence that settles that bet — years of published work, and an audience that has been telling you what they are stuck on, unprompted, for free.

So the job is not invention. It is extraction: going back through what already exists and reading it as data rather than as content.

The four places the evidence already exists

1. Your own back catalogue

Look for repetition, not performance. Which subject have you returned to five, ten, twenty times without planning to? That recurrence is usually a signal that you have more to say than the format allows — and that is exactly the shape of a product.

Views are a weaker signal than they look. A post can travel because it was funny, or timely, or because it annoyed people. Reach tells you what spread. Repetition tells you where your expertise actually is.

2. The questions you keep answering

Comments, DMs, email replies, the same question asked after every talk. Most creators have a handful of questions they have answered so many times they have stopped noticing. Those are the most valuable sentences you own.

Pay particular attention to questions that arrive after someone has consumed a lot of your work. Those people have moved past curiosity into implementation, and implementation is where people pay for help.

3. Problems described in their words, not yours

You have a professional vocabulary. Your audience does not. Collect the phrasing they actually use — "I keep starting over," "I don't know what to charge," "I built it and nobody came." The gap between how you name a problem and how they name it is often the reason a good product fails to sell: it was described in the wrong language.

4. What they already pay for

This is the one most creators skip, and it is the closest thing to hard evidence available before a product exists. What software, tools, courses, services, or supplies does your audience already buy in order to do the thing you talk about?

Existing spending establishes two things at once: that the problem is worth money to them, and roughly how much. A group that already pays for something adjacent has cleared a hurdle that a group who has never paid for anything has not.

Interest is not evidence someone will pay

This is the part that costs creators the most money, so it is worth being blunt about.

When you ask people whether they would buy something, you are asking them to predict their own future behavior about a thing that does not exist yet, while talking to someone they like. They will be encouraging. Encouragement is not demand.

What matters is not what people say, but what they have already done — and how much the action cost them. Attention is cheap. Typing is cheaper than paying. A useful way to think about it is a ladder, where each rung costs the person something more than the last:

• Weak Likes, saves, follows, "this is so needed!"

  Costs nothing and is often social politeness. Treat as ambient interest in the topic, not in a product.

• Weak "I would definitely buy that"

  A prediction made to please you. Not worthless — but it is the weakest form of stated intent, and it should never be the thing you build on.

• Better A specific, detailed question

  Specificity costs effort and reveals a real situation. "How do I handle X when Y?" is worth more than fifty enthusiastic comments.

• Better Giving you an email address for this thing specifically

  A small but real cost, and — crucially — it is a decision about this product rather than about you in general.

• Better Telling you what they currently pay for, and what it costs them

  Established spending on an adjacent solution. Evidence about the category, not yet about you.

• Strong Booking time, applying, or joining something with a limit

  Scarcity forces a real decision. People opt out of things they do not actually want when a slot is at stake.

• Strong Money changing hands — a deposit, a pre-order, a first sale

  The only evidence that settles the question. Everything above it is a forecast; this is a result.

None of this means you should ignore the top of the ladder. Weak signals are useful for generating candidates — they tell you where to look. They are just a poor basis for committing resources. Use the weak signals to decide what to test, and the strong ones to decide what to build.

One caution in the other direction: a failed test is not always proof that nobody wants the product. It may mean the offer was described badly, priced wrongly, or put in front of the wrong part of your audience. A single weak result narrows the question; it does not close it.

The exercise: 90 minutes, no tools

You need a document and your own archives. Nothing else. Do it in one sitting if you can — the value is in seeing the clusters, and that is hard when the work is spread over weeks.

1. Collect 30 real questions (30 minutes)

   Go through comments, DMs, email replies, and any place people ask you things. Copy 30 questions verbatim. Do not paraphrase, tidy, or translate them into your own vocabulary — the exact wording is the data. If you cannot find 30, that itself is a finding, and it is worth knowing now rather than after a launch.

2. Cluster them (15 minutes)

   Group the questions by the underlying problem rather than the surface topic. Two questions that sound different often turn out to be the same person stuck at the same place. You are looking for three to five clusters. If everything lands in one, you have a very clear product direction. If nothing clusters, your audience may be following you for the personality rather than the subject — useful to know, and a different business.

3. Score each cluster (15 minutes)

   Rate each cluster one to five on four axes, and write one line of justification for every score:

   • Frequency — how often does this come up, unprompted?
   • Urgency — is this blocking them right now, or is it someday-maybe?
   • Specificity — can you name exactly who has this problem, or is it "anyone interested in the topic"?
   • Proximity to money — do these people already pay for something to address this?

   Urgency and proximity to money are the two that most often predict whether something sells. Frequency alone is seductive and can simply mean the question is easy to ask.

4. Write the promise in one sentence (10 minutes)

   For your highest-scoring cluster, complete this sentence using their words: "This helps [specific person] go from [specific stuck point] to [specific outcome], without [the thing they dread]." If you cannot fill every bracket without vagueness, the opportunity is not ready. That is not a failure of the exercise — it is the exercise working.

5. Design the cheapest real test (20 minutes)

   Pick the smallest thing that would move someone two rungs up the ladder. Options, roughly in order of cost to you:

   • Publish one piece of content that is the promise, and see who replies asking for more.
   • Offer a limited number of paid one-to-one sessions on exactly that problem.
   • Open a waitlist that asks one qualifying question, not just an email.
   • Pre-sell a defined version at a defined price, with a stated delivery date and refunds if it does not ship.

   Note that every one of these produces evidence before the product is built. That is the entire purpose. Building first and testing afterward inverts the risk onto the most expensive part of the process.

What this method cannot tell you

An honest method states its own boundaries, so here are the four that matter.

• It cannot prove a product will sell. It improves the quality of the bet by grounding it in behavior instead of hope. It does not remove the bet. Products built on good evidence still fail, for reasons ranging from pricing to timing to execution.
• It is limited by what you can observe. The people who ask questions are not a representative sample of your audience — they are the most engaged slice. The quiet majority may be stuck somewhere else entirely, and you will not see it here.
• It says nothing about whether you want to run the business. A product is an ongoing obligation: support, updates, refunds, marketing, tax. A strong signal for something you would resent operating is not a good outcome.
• It does not account for your capacity to build it. The highest-scoring opportunity may require software, ongoing service delivery, or an upfront investment you are not in a position to make. That is a real constraint, not a reason to pick a weaker idea by default — but it has to be faced honestly before you commit.

---