What's the best process for content ideation?

TL;DR

The best content ideation process is not a single brainstorm — it is a small, repeatable loop with three stages: capture, validate, and prioritize. You capture ideas continuously from real sources (audience questions, competitor gaps, search data, your own published content), you validate each one against measurable demand and strategic fit before you commit to writing it, and you prioritize what survives into a ranked backlog. The stage most teams skip is validation — and skipping it is expensive. The loop closes because published content generates its own new inputs (reader questions, internal-link gaps, decaying pages), so a working ideation process feeds itself over time rather than needing a fresh "big idea day" every few weeks. Below: how the loop works, a scannable list of the techniques that fill it, and the worked example of how we planned this blog.

---

Ask ten content teams for their ideation process and most will describe the same clean line: brainstorm, pick, write. It fits on a whiteboard and it is easy to teach. It also quietly fails, because it treats ideation as a one-time event you do before production — a feast-or-famine cycle of one big idea day, a long list, a slow grind through it, then emptiness and another big day.

A better way to think about it: ideation is a loop you keep running, not a gate you pass through once. And the loop only has three moving parts.

What is content ideation, exactly?

Content ideation is the ongoing practice of sourcing, validating, and prioritizing the topics you will actually create. It sits between raw inspiration and a production-ready brief. Done well, it turns scattered signals — a support ticket, a keyword export, a competitor's thin page — into a ranked list of things worth writing, each attached to a reason.

The word doing the heavy lifting there is validating. Generating ideas is easy; almost any prompt or meeting produces a long list. The hard, valuable part is deciding which ideas deserve the days of work a real piece costs — and killing the rest before you spend anything on them. A process that only generates is a list. A process that generates and filters is a strategy.

The three-stage loop: capture, validate, prioritize

Here is the whole process. Everything else in this guide is detail hung on these three stages.

1. Capture — widely, with no judgment at intake. The top of the loop should be porous. Anyone on the team can drop a link, a customer quote, a half-formed question, or a keyword into one shared place. Systematic sources (SEO gap analysis, competitor audits, community listening) go in the same bucket as serendipitous ones. The only rule at intake is no filtering yet — filtering here kills the odd, specific ideas that often turn out best, and it slows capture to a crawl.

2. Validate — before you commit a single writing hour. This is the stage that separates a real process from a wish list. Each captured idea gets checked against two questions: does anyone actually want this, and does it fit what we are trying to build? "Want" usually means measurable search demand or a recurring real question from customers. "Fit" means it supports a topic you have chosen to own. Ideas that fail either check get cut — not parked, cut.

3. Prioritize — into a ranked backlog, not a flat calendar. What survives validation gets scored and ordered by opportunity, so you are always pulling the next-best idea, not the next-oldest one. A backlog ranked by opportunity is worth far more than a calendar filled by date.

The reason this is a loop and not a line: stage 3 output — published content — becomes stage 1 input. A live post attracts reader questions, exposes gaps in an existing topic cluster, and eventually decays and needs a refresh. Each of those is a fresh idea flowing straight back into capture. Publishing is your most reliable ideation source, which is exactly why "ideation as a one-time event" leaves so much on the table.

One-time brainstorm model	Capture → validate → prioritize loop
Goal: fill a list of topics	Goal: keep a ranked backlog of validated topics
Flow: linear, done before production	Flow: circular, published work feeds new ideas
Inputs: mostly a keyword export	Inputs: keywords + customer questions + your own published content
Selection: gut-check meeting	Selection: measurable demand + strategic fit
Output: a static calendar	Output: a backlog ranked by opportunity

Content ideation techniques (the ones worth using)

The loop above is the shape; these are the techniques that fill the capture and validate stages. Pick a handful and run them consistently — that beats knowing all of them and using none.

Keyword-gap analysis. Map the terms your site already covers against the terms your topic clusters should cover, and the holes are your idea list. This is how we source most of our own topics — it points you at demand you have a real chance of serving rather than whatever sounds clever that week. If you are new to the mechanics, our keyword research guide for beginners covers sourcing and validation from scratch.

"People also ask" and question mining. Search your core topic, open the "People also ask" box, and note the questions that appear — then click one and watch new ones expand. Three layers deep is often a full outline. These are the exact phrasings real people use, which makes them both good keywords and good headings.

Customer conversation mining. Sales calls, support tickets, onboarding questions, and community threads are the highest-signal source you own, because the person is telling you their problem in their words. A question that comes up three times in support is a validated topic before you have run a single search.

Competitor page-two audit. Look at what similarly sized sites — not the giant publishers — rank for on page two. Those are terms with demand that nobody has fully answered yet. A more complete, sharper page is a realistic way to take that spot.

Topic-cluster mapping. Instead of hunting single posts, pick a topic you intend to own and list the dozen specific questions orbiting it. One cluster of interlinked pages builds authority in a way scattered one-off posts never do. Our topic cluster strategy guide walks through structuring one.

Search-intent reframing. Take an idea and ask what the searcher actually wants — to learn, to compare, or to buy — then check the current top results to confirm. An idea can be a good topic and a bad post if you write the wrong format for its intent; our search intent guide covers how to read that signal.

Content-decay review. Periodically scan your analytics for pages losing traffic. A decaying post is often a better bet than a brand-new idea: the topic is proven, and a refresh usually returns faster than starting cold. This is the clearest example of the loop feeding itself.

Repurposing sparks. When a piece performs, ask what else its core idea could become — a companion post answering the top comment, a deeper cut on one section, a different format. One good idea rarely wants to stay one piece.

The 70-20-10 allocation. This is a well-known industry heuristic for spreading your bets, not a data finding: put roughly 70% of your effort into proven, core topics that support your pillars, 20% into adjacent or slightly newer angles, and 10% into genuinely experimental ideas. It is a guardrail against two failure modes — writing only safe posts, or chasing only shiny new ones.

Notice that most of these techniques feed capture, and only intent-checking and demand validation feed the validate stage. That is the usual imbalance: teams have five ways to generate ideas and no discipline for killing them. The next section is about fixing that.

The stage everyone skips: validation

Generating ideas feels productive, so teams over-invest there and under-invest in the filter. The result is a calendar full of topics nobody asked for.

We know the cost of skipping validation because we paid it. When we audited our own early posts, 43% of the keywords we had targeted turned out to have zero measurable search volume. The phrases read like real keywords — they fit our topics, they sounded plausible — but nobody was searching them. We had written for an audience that did not exist. Every one of those posts would have been caught by a single validation step: confirm the demand before you write.

A workable validation filter asks two things of every surviving idea:

• Is there real demand? For search-driven topics, that means confirming the phrase has measurable volume — a free keyword planner is enough; the discipline matters more than the tool. For audience-driven topics, it means the question actually recurs, not that it came up once. (Scoring ideas against real search volume before writing is one of the things our own tool, Writesy, does — but the point is to run the check somehow, not with any particular product.)
• Does it fit what we are building? An idea can have demand and still be wrong for you if it does not support a topic you have chosen to own. Chasing unrelated high-volume terms is how a blog ends up broad, shallow, and unranked.

If you find yourself validating idea-by-idea and getting stuck, the deeper method is worth reading: how to validate content ideas without becoming an SEO tool addict covers doing this without over-relying on the tools, and when keyword research actually helps covers when the data is worth checking at all versus when it just generates noise.

A worked example: how we planned this blog

We do run this loop on ourselves, and we documented it. When we sat down to plan this blog, the obvious move was to find high-volume keywords and write comprehensive posts against them. We deliberately rejected that, because an SEO-first list tends to produce content that looks like everyone else's.

Instead we started from topics we wanted to own — a small set of clusters where our expertise and positioning overlap — and only then sourced specific ideas within them. We validated each idea against demand and fit, and we cut the ones that failed: a "content calendars" piece we had no distinct take on, an "AI content detection" piece that was reactive rather than strategic, and a couple of adjacent-but-off-cluster topics. Cutting is uncomfortable, and it is most of the value.

The full breakdown — the clusters we chose, the topics we excluded, and the decisions we got wrong — is in we planned this blog using our own content system. It is the same three-stage loop described above, run once, in public.

Turning the backlog into a plan

Ideation ends where planning begins: a validated, prioritized backlog is the input to a calendar, not a substitute for one. Once you have a ranked list, sequence it against your real capacity — usually fewer posts than you think, done well, beats more posts done thinly. Our guide to building a content calendar covers turning the backlog into a realistic schedule.

The bottom line: a perfect idea list is useless in a vacuum. The best ideation process is the one that connects capture to validation to a backlog you actually work from — and that quietly refills itself from the content you publish.

FAQ

What is the best content ideation process?

The most reliable process is a three-stage loop: capture ideas continuously from real sources (audience questions, competitor gaps, search data, and your own published content); validate each idea against measurable demand and strategic fit before committing to write it; and prioritize what survives into a backlog ranked by opportunity. The stage most teams skip is validation, which is also the one that saves the most wasted effort. Because published content generates new inputs — reader questions, internal-link gaps, decaying pages — the loop feeds itself rather than needing a fresh brainstorm every few weeks.

What are the best content ideation techniques?

The techniques worth running consistently are: keyword-gap analysis (map what you cover against what your clusters should cover), "People also ask" and question mining, customer-conversation mining from sales and support, competitor page-two audits, topic-cluster mapping, search-intent reframing, content-decay reviews of your own analytics, and repurposing sparks from posts that performed. Most of these feed idea generation; pair them with a demand-and-fit validation step so you are also killing the weak ideas, not just collecting more.

What is the 70-20-10 rule in content?

The 70-20-10 rule is a well-known industry heuristic for allocating content effort, not a research finding. It suggests putting roughly 70% of your effort into proven, core topics that support your main pillars, 20% into adjacent or newer angles that build on those themes, and 10% into genuinely experimental ideas — new formats or audiences that might fail. It is a guardrail against two opposite mistakes: only ever writing safe posts, or only ever chasing shiny new ones.

How is content ideation different from content strategy?

Content strategy decides what you should be known for — the audience, the topics you will own, and how you will measure success. Content ideation is the narrower, ongoing practice of sourcing and validating the specific topics that fit that strategy. Strategy is the boundary; ideation fills inside it. Ideating without a strategy is how you end up with a broad, shallow blog, because every idea looks equally valid when there is nothing to test fit against.

How often should you do content ideation?

Continuously for capture, on a schedule for validation. Treat capture as always-on — a shared place where ideas land the moment they appear — and set a recurring slot (weekly or every two weeks) to validate and prioritize the intake into your backlog. This beats the "big idea day" pattern, which produces a burst of ideas that go stale before you work through them and leaves you empty until the next session.

Further Reading

• How to Decide What Content to Create (Without Guessing)
• Why SEO Tools Don't Help You Plan Content
• Content Strategy: The Go-To First Step