How to Choose Blog Topics When You Could Write Anything

Open a blank doc with “write a blog post” on your to-do list and the hardest part hits you before the first sentence: you could write about anything. Here’s how to choose blog topics without the spiral: run every candidate through a three-part intersection test — what your audience actually asks about, what you can answer better than anyone else, and what serves your business. A topic that hits two of the three makes mediocre content. A topic that hits all three belongs in your catalog. That test, applied consistently, is most of what separates blogs that compound from blogs that churn.

And I want to say the quiet part out loud right away: topic choice is where your content strategy actually happens. Traffic, trust, and sales-relevance are mostly decided before you write a single word. A brilliant article on a topic nobody wants, that you had no business writing, aimed at readers who’ll never buy — that’s three months of effort spent beautifully decorating a dead end. Let’s make sure that never happens to you.

Why is topic choice where content strategy actually happens?

Okay, let’s be honest about something the “just write great content” crowd skips: quality is a multiplier, not a starting value. A 10x-better article on a topic with no audience demand multiplies zero. A perfectly optimized post on a topic where you have nothing new to say gets outranked by someone who does. And a traffic magnet that attracts readers who will never, ever need what you sell is a hosting bill with a vanity metric attached.

Each circle of the test guards against one of those failures:

• Audience demand guards against writing into the void. Someone has to be asking.
• Your edge guards against being a worse copy of an answer that already exists. You have to bring something.
• Business fit guards against building an audience your business can’t serve. The topic has to point somewhere.

Two out of three always produces a recognizable kind of mediocre. Demand + fit without edge is commodity content — technically relevant, instantly forgettable. Demand + edge without fit is a popular blog that never moves revenue. Edge + fit without demand is a beautifully argued piece read by your cofounder and your mom. The intersection is small, and that’s the point: it’s a filter, not a brainstorm. So let’s unpack each circle properly, because each one has traps.

How do you find out what your audience actually wants? (Circle one: demand)

Demand evidence comes in tiers, and I’d rank them honestly rather than the way most guides do — because most guides start with keyword tools, and keyword tools are tier two.

Tier one: your own customers’ questions

This is the gold vein, and it’s sitting in systems you already own. Support tickets. Sales call notes. Onboarding emails. The questions people ask in demos. The comments and DMs on your social posts — if you manage your channels from a unified inbox (SocialBlaze keeps comments from every connected network in one place, which makes this sweep a ten-minute job instead of an afternoon), scroll one month of them and write down every genuine question. These questions have a property no keyword tool can offer: you know real people with money-adjacent intent asked them, in their own words, recently. When a question shows up three times in a month, that’s a blog topic with proof of demand attached.

Tier two: search demand — with the honesty caveat

Search data widens the view beyond people who already found you. Three sources, cheapest first:

• Autocomplete. Type your seed topic into Google and read the suggestions. Each one is a phrasing real people type.
• People Also Ask. The question boxes under results are a map of the sub-questions searchers actually have.
• Keyword tool estimates. Useful for relative comparison — topic A likely gets more searches than topic B. But say it plainly: every volume number in every tool is an estimate, often a rough one, and two tools will give you two different numbers for the same phrase. Use them to rank candidates against each other, never as a promise of traffic. Nobody can promise you traffic, including me.

Tier three: community listening

Forums, subreddits, Facebook and Slack groups, Discord servers — wherever your audience talks to each other instead of to vendors. This is where the unfiltered phrasing lives: the frustrated version of the question, the “am I crazy or…” version, the comparison people make when no marketer is watching. Community listening rarely gives you volume data, but it gives you language and emotional context, which is worth more than you’d think when you sit down to write. If you want a deeper system for this kind of evidence-gathering, I’ve walked through the full process in my guide to how to do content research — topic selection is really just content research with a decision bolted on.

The no-demand trap

Here’s the trap this circle exists to catch: topics only you find interesting. The behind-the-scenes of your internal tooling. The industry debate that fascinates your team and precisely nobody else. These feel like great topics because your enthusiasm is real — but enthusiasm is not evidence. If you can’t point to a customer question, a search suggestion, or a community thread, the honest conclusion is that demand hasn’t shown up yet. Park it. (Occasionally you’ll publish one anyway as a bet — that’s allowed — but know you’re betting, and don’t bet weekly.)

What’s your edge — and why does it matter more than ever? (Circle two)

Demand tells you people want an answer. The second circle asks a harder question: why would your answer be the one worth reading?

I call it the better-answer test, and it has three honest passes. You have an edge on a topic when at least one of these is true:

• Experience you have. You’ve actually done the thing — shipped it, failed at it, fixed it. You can write the paragraph that only someone who’s been there can write.
• Data you own. Your product, your customers, your operations generate observations nobody else has. Even small proprietary data beats recycled third-party claims.
• A take you’ll defend. A genuine position — one you’d argue for on a call with a skeptic — on a question where the consensus answer is weak or lazy.

Now the context that makes this circle urgent: commodity answers are free now. Anyone can ask an AI assistant for a competent generic explainer on almost anything, instantly. That flood didn’t make content marketing pointless — it made the generic middle worthless and raised the premium on everything the flood can’t produce: lived experience, proprietary data, a named author willing to take a position. Your edge isn’t one advantage among several anymore. It’s close to the only moat. (This is the same logic I lean on in ranking without a big backlink profile: when you can’t outspend, you out-specific.)

And the honesty this circle demands: no edge means earn it or skip it. If a topic scores high on demand and fit but you’ve got nothing — no experience, no data, no defensible take — you have two legitimate moves. Earn the edge: run the experiment, collect the data, do the thing, and write about it afterward. Or skip it and let someone better-positioned have that topic. The illegitimate move is the one most blogs make: write it anyway by paraphrasing the current top results, adding a unit of sameness to a pile of sameness. That’s the content the flood already makes for free.

How do blog topics connect to actual business results? (Circle three)

The third circle is the one marketers whisper about, as if admitting the blog should help the business is gauche. It isn’t. It’s the reason the blog gets budget.

Map topics to the funnel

For every candidate, ask: who searches for this, and what might they buy? Some topics attract people with the exact problem your product solves — these are rarer and more valuable than their search volume suggests. Some attract people adjacent to your buyer: right industry, wrong role, or right role, wrong stage. And some attract tire-kickers, students, and the merely curious. None of these are bad people — they’re just not all equally worth acquiring with your limited publishing slots.

The adjacent-topic honesty

The seductive failure here is the “viral-adjacent” topic: broad, high-estimated-volume, loosely related to your space. A project management tool writing “best motivational quotes for Monday.” The demand is real! The edge might even exist! But the audience it attracts costs real money to host and serve, and almost none of them will ever become customers. Traffic that can’t convert isn’t neutral — it’s an expense wearing a success costume. When a topic’s connection to your buyer requires a three-step story to explain, believe the three steps, not the story.

The sells-without-selling test

My favorite practical check: does this topic let your product appear naturally? Not a bolted-on pitch paragraph — a moment in the genuine answer where the honest advice is “and this is the kind of thing a tool handles for you.” If your product can walk on stage as part of the actual answer, the topic has business fit. If mentioning your product would feel like a cutaway ad, it probably doesn’t — and you’ll feel that awkwardness every time you write the draft.

How to choose blog topics in practice: the intersection workflow

Theory’s lovely; here’s the Tuesday-afternoon version. Four steps, about ninety minutes the first time, faster every time after.

Step 1: The candidate sweep

Collect roughly 20 candidates from the demand sources, in tier order: sweep a month of customer questions (support, sales, social comments), pull autocomplete and People Also Ask for your two or three seed topics, and skim your best community spaces for repeated frustrations. Write each candidate as a question or a promise, not a keyword — “how do we know which metrics actually matter” travels better through the next steps than “marketing metrics.”

Step 2: The three-circle scoring

Score each candidate 1–3 on each circle. Keep it crude on purpose — this is a sorting tool, not a science:

• Demand: 1 = I think people care. 2 = search/community evidence exists. 3 = my own customers asked this, repeatedly.
• Edge: 1 = I’d be paraphrasing. 2 = I have real experience or a defensible take. 3 = I have experience and data or a strong position.
• Business fit: 1 = vaguely our world. 2 = our buyer would read this. 3 = our product appears naturally in the honest answer.

Sevens and up go on the shortlist. Everything else — and this matters — doesn’t get deleted. Scored candidates are exactly what a healthy content backlog is made of: today’s six might be next quarter’s eight once you’ve earned the edge or the product grows into the fit.

Step 3: The cluster check

Before a shortlisted topic gets a slot, ask: does it serve a pillar? Blogs build authority in clusters — a pillar topic surrounded by spokes that link up to it and to each other. A high-scoring topic that belongs to no cluster is an orphan: it might do fine alone, but it dilutes your focus and strengthens nothing around it. Orphans aren’t forbidden; they’re just expensive. If you keep generating great orphans in the same neighborhood, that’s a signal to found a new cluster there deliberately, with a pillar — not to scatter one-offs.

Step 4: The cannibalization check

Last gate, two minutes: do we already rank for this? Search your site (site:yourdomain.com plus the phrase) and check Search Console. If an existing post already covers the topic, the right move is almost always to improve the incumbent — refresh it, deepen it, re-promote it — rather than publish a competitor to your own page and make Google choose between your children. A new post wins only when the angle or audience is genuinely different enough to deserve its own URL.

How do you shape the topic once you’ve picked it?

Choosing the territory isn’t the end — the same territory can produce wildly different articles. Three shaping decisions, made before the outline:

Pick the altitude

Every topic exists at multiple altitudes: the beginner version (“what is X and why should I care”), the practitioner version (“the exact workflow, with the mistakes”), and the decision-maker version (“how to evaluate, budget, and justify X”). These are different articles with different readers, and altitude is one of the most underused forms of differentiation — when the top results all sit at beginner altitude, the practitioner version of the same topic can win on depth alone. Choose the altitude where your edge is strongest, not the one with the biggest estimated volume.

Match the format to the topic

Some topics are tables — comparisons, option matrices, spec-driven decisions. Some are stories — lessons, post-mortems, experience pieces. Some want to be tools — templates, checklists, calculators, worksheets. And some aren’t blog posts at all; a messy multi-step process often teaches better as video, which is its own planning discipline — my guide to building a video content strategy covers how to decide what deserves the camera. Forcing a table-shaped topic into 2,000 words of prose is how good topics become bad articles.

The title-first test

Before you outline, draft the actual title — three or four honest versions. If no honest title excites you, the topic is weaker than its score suggested, and it’s better to learn that now than 1,800 words in. A strong topic produces a title that makes a specific promise you can keep; a weak one produces titles that need hype to sound interesting. The title is the topic’s first stress test. Let it fail cheap.

How do you balance a blog’s topic portfolio?

Here’s the part nobody tells you: knowing how to choose blog topics one at a time isn’t enough, because even perfect individual choices can add up to a lopsided blog. Topic selection is a portfolio problem, and a healthy portfolio holds three kinds of positions:

• Demand-bankers: search-evergreen how-tos and explainers. They compound slowly and bring strangers in the door.
• Edge-flexers: opinion, experience, and data pieces. They rarely win search but they’re why anyone remembers you, subscribes, or shares — and increasingly, they’re what earns citations in AI-generated answers.
• Business-servers: comparisons, alternatives, “how we do X” pieces. Small audiences, high intent, closest to revenue.

A sane cadence mixes all three — many teams land somewhere near half demand-bankers with the rest split between the other two, then adjust by what their goals need most. The trap worth naming is the all-SEO-topics blog: every post a keyword-shaped explainer, nothing with a pulse. It can rank and still be a blog nobody could love — no voice, no point of view, nothing a human would subscribe to. Your blog is allowed to have a personality. Strategically, it needs one.

One more portfolio rule: trends are garnish, not the meal. Timely topics earn occasional slots — they show you’re awake and they catch attention spikes — but they decay fast, and a blog built on them has to re-earn its traffic every month forever. Season with trends; cook with evergreens.

What should you not do when choosing blog topics?

• Don’t copy a competitor’s topic list. Their topics passed their three circles — their customers’ questions, their experience, their product. Their edge isn’t yours, so the same topic that works for them produces commodity content from you. Study their gaps, not their greatest hits.
• Don’t volume-chase zero-fit keywords. A big estimated number next to a topic your buyer doesn’t care about is a trap with a bow on it. Revisit the adjacent-topic honesty above; it never stops applying.
• Don’t let shiny objects skip the line. The idea that arrives at 11pm feeling urgent goes through the backlog like everything else: scored, cluster-checked, scheduled. If it’s really a nine, the scoring will say so in the morning.
• Don’t write to a score. Optimization tools that grade drafts against the current top results will happily guide you toward topics and outlines that are averages of what already ranks — content built for a grader, with no actual reader in mind. Tools advise; the three circles decide.

Your topic-selection toolkit

Four pieces, all reusable. Copy them into whatever you plan in.

The three-circle worksheet

Circle	The question	Evidence that counts	Kill signal
Audience demand	Who is asking for this, and where?	Repeated customer questions; autocomplete/PAA phrasings; community threads	Only you find it interesting
Your edge	Why would our answer be the one worth reading?	Experience you have, data you own, a take you’ll defend	You’d be paraphrasing the current top results
Business fit	Does the reader this attracts ever buy?	Buyer-relevant problem; product appears naturally in the honest answer	A three-step story is needed to connect topic to buyer

The scoring card

Candidate topic	Demand (1–3)	Edge (1–3)	Fit (1–3)	Total	Decision
Example: “How do we pick which metrics matter?”	3 — asked in 4 support threads	2 — strong take, no data yet	3 — product shows the metrics	8	Shortlist
Example: “Industry history deep-dive”	1	3	1	5	Backlog
Rule of thumb: 7+ shortlists, any circle at 1 triggers the earn-it-or-skip-it conversation.

The portfolio-mix planner

Position	What it is	What it’s for	Example share of cadence
Demand-bankers	Evergreen how-tos, explainers	Compounding search discovery	Roughly half
Edge-flexers	Opinions, experiments, owned data	Memorability, subscriptions, citations	A steady quarter-ish
Business-servers	Comparisons, alternatives, how-we-do-it	High-intent, revenue-adjacent readers	The remainder
Trend garnish	Timely reactions	Attention spikes, freshness	Occasional — seasoning only

The candidate-sweep source list

Source	Where to look	What it gives you
Support & sales	Tickets, call notes, demo questions	Proven questions from money-adjacent people
Social comments & DMs	One month of comments across your channels (a unified inbox makes this one sweep)	Real phrasing, recency, engagement signals
Autocomplete & PAA	Google suggestions and question boxes for seed terms	How strangers phrase the search
Keyword tools	Any estimator you have	Relative demand ranking — estimates only, never promises
Communities	Subreddits, groups, Discords, forums	Unfiltered frustration and comparison language
Your own analytics	Search Console queries, top posts’ comments	Demand you’re already half-serving

FAQ: how to choose blog topics

How many blog topic ideas should I generate before choosing?

Sweep for about 20 candidates before you score anything. Fewer than that and you’re choosing from whatever you happened to think of; the volume forces you past the obvious ideas. Score all of them 1–3 on demand, edge, and business fit, shortlist the 7+ totals, and keep the rest in your backlog — scores change as your experience and product grow.

Should I choose blog topics based on keyword search volume?

Use search volume as one input, never the decision. Every tool’s volume number is an estimate, and tools disagree with each other on the same phrase. Volume is useful for ranking candidates against each other, but a high-volume topic with no edge and no business fit still fails the intersection test — and a low-volume topic your customers keep asking about often outperforms its numbers.

What if I have no unique edge on a topic people clearly want?

Earn it or skip it. Run the experiment, gather your own data, or do the thing and write from experience afterward — that’s earning the edge, and it usually takes weeks, not years. Writing it anyway by paraphrasing the current top results just adds another commodity answer to a pile that AI tools already generate for free.

How do I know if a blog topic will attract buyers instead of tire-kickers?

Apply the sells-without-selling test: in a fully honest answer to the topic’s question, does your product appear naturally as part of the solution? If yes, the readers it attracts live near your funnel. If mentioning your product would feel like an ad break, the topic likely attracts an audience you’ll pay to host but never convert.

Is it bad to write about a topic a competitor already covered?

Not at all — most good topics are covered somewhere. What matters is whether you can pass the better-answer test: different experience, your own data, a sharper take, or a different altitude (the practitioner version where theirs is beginner-level). Copying their topic list wholesale is the mistake, because their topics were chosen for their edge and their buyers, not yours.