01The blank-calendar problem
Staring at an empty content calendar, most people brainstorm — they list topics that sound interesting to them and hope some land. It’s a guessing game, and the guesses are biased toward what the writer finds interesting rather than what the audience is actually searching for.
There’s a better starting point: the questions your audience already asks in public, unprompted, on repeat. You’re not inventing topics — you’re reading the ones they’ve already told you matter.
02Your audience already wrote the calendar
Every community has a set of questions that come up again and again — the ones regulars are tired of answering. Each recurring question is a piece of content with proven demand: people are actively looking for the answer, in their own words, often enough that it repeats.
The phrasing matters as much as the topic. The exact words people use to ask are the exact words they’ll later type into a search box — so a question mined from a thread is both your headline and your keyword research in one.
03What to mine for a calendar
Reading a community for content ideas, you’re collecting five things:
- Recurring questions — the ones asked weekly, each a proven-demand topic
- The exact phrasing — how they ask, which becomes your title and your keyword
- The “nobody talks about…” gaps — under-served topics with an audience waiting
- The debates — recurring disagreements that make for high-engagement pieces
- What gets upvoted — the answers the community already rewards, worth expanding
04From threads to a calendar
The pipeline turns this into a repeatable step rather than a manual skim. Search queries are built from the actual title phrases people post in your chosen communities, with frequency counts — so you see which questions recur most, not just which ones you happened to read. Content ideation is a first-class use-case the classifier tags for, so topic-shaped threads surface instead of getting lost.
The output is a ranked list of real questions with real frequency behind them — a calendar sourced from demand, not from a brainstorm.
See the full workflow
Mining an audience for content ideas is a use-case the pipeline is built around end to end.
Audience & content research →05A worked example: from scattered questions to a plan
Collect the raw questions
Over a couple of weeks of reading the communities your audience lives in, you jot down the things people actually ask — “why does my roast taste flat”, “how do I know first crack”, “cheap roaster or popcorn popper to start”, “how long do I rest beans before brewing”. You keep the wording verbatim, because the wording is the lead.
Group them by theme and intent
Tasting problems cluster together. Equipment-for-beginners is its own cluster. Process timing — first crack, resting, cooling — is a third. Intent matters too: a “how do I start” question wants a beginner guide, while “why does mine taste flat” wants a troubleshooting piece.
Map each cluster to one piece
The equipment cluster becomes a starter-gear comparison. The tasting cluster becomes a troubleshooting guide framed around the exact symptoms people described. The timing cluster becomes a single explainer of the roast curve.
Title each piece in their words
The flat-taste question becomes the headline almost unchanged — the phrasing people used to ask is the phrasing they will later search for, so you have written the title and done the keyword work in one move.
Frequently asked questions
Stop brainstorming topics that interest you and start reading the questions your audience already asks in public. The ones that repeat are pre-validated demand — people are actively looking for those answers. Collect the recurring questions over a few weeks, note the exact wording, and each one becomes a content idea you know someone is searching for.
Spend time where your audience gathers and log the questions that come up unprompted, keeping the phrasing verbatim. Watch for the ones regulars are tired of answering — those repeat because demand is steady. Then group the questions by theme and by intent, so a cluster of related asks points clearly at one piece you should write.
Read the threads, comments, and questions in the communities your audience already uses, rather than guessing from your own assumptions. The signal you want is repetition — a question asked once is a whim, but a question asked weekly is a topic with proven pull. Capture the exact words; they double as your future headline and search terms.
Collect the recurring questions, group them into themes and by what the asker actually wants — a beginner guide, a troubleshooting fix, an explainer. Then map each cluster to a single piece that answers it in the audience’s own language. The result is a calendar sourced from demand instead of a brainstorm, with each title already phrased the way people search.