How to Run Content Ideation That Produces 40 Article Ideas in 90 Minutes

TL;DR

Content ideation is the structured process of producing qualified article, video, and campaign ideas from four specific inputs: SERP data, sales and customer conversations, competitor gaps, and internal subject-matter expert interviews. A well-run 90-minute ideation session produces 40 ideas before any filtering. The ideas worth producing come from the intersection of real buyer queries and non-obvious angles your competitors haven't already written about.

Most content ideation sessions in B2B marketing are a whiteboard, a group of tired marketers, and 45 minutes of people throwing out whatever came to mind that morning. The ideas that come out of that process are predictable, mostly derivative, and produce content nobody needed.

I've run content ideation for enterprise tech brands for over a decade. The version that works is 90 minutes long, uses four specific inputs, and produces roughly 40 qualified ideas per session. Here's how to run it.

Why most content ideation produces ideas nobody needs

The failure mode of most content ideation is structural, not creative. Teams brainstorm in a vacuum, without SERP data, without sales conversation tags, without any structured input from customers or internal experts. The output is whatever the loudest person in the room happened to think of.

That process produces what I call regression-to-the-mean content. Every competitor in your category is generating the same articles, targeting the same keywords, making the same claims. The ideas aren't bad individually. They're just identical to what your audience has already seen ten times.

A good content ideation process starts with inputs that force differentiation. It doesn't start with a whiteboard.

The four inputs every content ideation session should start with

Before the meeting, the facilitator pulls four datasets. These inputs replace the blank-whiteboard problem.

Input one is SERP data for five to seven primary keywords the brand needs to own. Input two is a set of tagged customer conversations from sales, customer success, and support tickets. Input three is a competitor content map showing the top ten pieces each direct competitor has published in the last six months. Input four is a list of three to five internal subject-matter experts who will be interviewed briefly before the session.

The session doesn't start until all four inputs are in the room. If any are missing, the ideation reverts to guesswork.

SERP mining for the keywords buyers actually search

Open your keyword research tool. I use SEMRush. For each primary keyword, pull the top ten ranking pages, the People Also Ask questions, and the related keyword cluster. Read the headlines of the top ten pages. Note what they cover and, more importantly, what they don't cover.

The gaps in SERP coverage are where differentiated ideas live. If every top-ranking piece on content marketing ROI lists the same metrics, an idea like "the three content ROI metrics that survive a board meeting" is differentiated. If every piece on SEO strategy talks about keywords, an idea like "the SEO signals that matter more than keywords in 2026" is differentiated.

This step alone produces eight to twelve ideas per keyword for an experienced facilitator. Five to seven keywords means roughly 40 to 80 SERP-sourced ideas before any of the other inputs are even opened.

Sales and customer conversation tags as idea fuel

The second input is conversations. Not survey data, not personas, not market research. Actual transcripts, call notes, and support tickets from the last 60 to 90 days.

Ask sales to tag the five objections they heard most often. Ask customer success to tag the five questions new customers ask in onboarding. Ask support to tag the five problems that generate the most tickets. Fifteen tags, each one a potential article. Many of them are questions that never appear in keyword research tools because they're phrased in customer language rather than search-engine language, but they're exactly the kind of queries people now type into ChatGPT and Claude.

This input typically produces another eight to twelve ideas.

Competitor gap analysis without copying competitor output

The third input is a map of what your competitors are publishing. Not so you can copy them. So you can identify what they're not writing about.

Pull the top ten pieces published by each of your three to five closest competitors in the last six months. Read the headlines. What topics are they over-covering? What questions are they avoiding? What angles have they not taken? Competitors collectively reveal the shape of the content universe. The negative space is where your differentiation lives.

I've seen this input produce three to five ideas that nobody else has written. Those are often the highest-performing pieces over a 12-month window because they earn backlinks and citations without direct competition.

Internal SME interviews that surface non-obvious angles

The fourth input is the one most B2B teams skip. Before the ideation session, the facilitator conducts 20-minute interviews with three to five internal subject-matter experts. These are the engineers, product leads, solutions architects, and customer-facing specialists who know things the marketing team doesn't.

Two questions per interview. What is something you wish customers understood better before they bought from us? What is something you've changed your mind about in the last year?

Those two questions produce angles that nobody on the marketing team could have generated alone. They're the source of the thought leadership pieces that actually sound like leadership rather than summary.

The 90-minute session structure that produces 40 ideas

With the four inputs in the room, the session itself runs on a tight structure.

Minutes zero to 15, review the inputs as a group. Minutes 15 to 45, silent idea generation. Each participant writes ideas on sticky notes or in a shared doc, referencing the inputs, aiming for 15 to 20 ideas per person. Minutes 45 to 60, group clustering of ideas by topic and format. Minutes 60 to 75, discussion of the most differentiated clusters. Minutes 75 to 90, initial shortlisting of ideas worth scoring later.

A team of three to four people running this structure produces 40 to 60 ideas. Some will be weak. Most will be better than anything a blank-whiteboard session would have produced. The filtering happens next.

Scoring and shortlisting after the session

The 40 ideas from the session are raw. They need scoring before anything gets written. I use a five-dimension scorecard that covers search intent fit, differentiation, evergreen potential, distribution hook, and sales conversation value.

The point of the scorecard isn't to pick favorites. It's to kill weak ideas fast. Out of 40 ideas, 30 typically fail the scorecard. The remaining 10 become the editorial roadmap for the next quarter.

Better to produce 10 strong articles than 40 average ones. The ideation session exists to make sure the 10 you keep are the right 10.

But hey, we understand that this isn't second nature for everyone. Which is why we're here to help. If your team needs help with content ideation or more, don't hesitate to find some time to connect.

Frequently asked questions

What is content ideation?

Content ideation is the structured process of generating qualified article, video, and campaign ideas for a content marketing program. It combines SERP research, customer conversation data, competitor gap analysis, and internal subject-matter expert interviews to produce ideas grounded in real buyer queries rather than internal guesswork.

How do you run a content ideation session?

A structured 90-minute content ideation session follows five phases. Fifteen minutes to review four prepared inputs, thirty minutes of silent individual idea generation, fifteen minutes to cluster ideas by topic, fifteen minutes to discuss the most differentiated clusters, and fifteen minutes for initial shortlisting. This produces roughly 40 ideas for a team of three to four people.

What inputs should a content ideation session start with?

A productive content ideation session starts with four inputs prepared in advance. SERP data for five to seven primary keywords, tagged customer conversations from sales and support, a competitor content map of the last six months, and internal SME interviews with three to five subject-matter experts. Sessions without these inputs revert to guesswork.

How many content ideas should a single ideation session produce?

A well-run 90-minute content ideation session typically produces 40 to 60 raw ideas for a team of three to four people. After scoring and filtering, roughly 10 ideas survive as viable production candidates. The goal is quantity in the session and quality after the scorecard pass.

What's the difference between content ideation and content strategy?

Content strategy defines what topics, formats, and channels a brand will invest in and why. Content ideation is the tactical process of generating specific article, video, and campaign ideas within that strategy. Strategy sets the boundaries; ideation fills them with executable ideas.

How often should a B2B team run content ideation sessions?

Most B2B content teams benefit from a formal content ideation session once per quarter, producing the editorial roadmap for the following three months. Between sessions, lightweight ideation (SERP checks, sales conversation reviews) should happen continuously to capture timely topics that emerge.

Can AI tools replace a structured content ideation process?

AI tools can accelerate specific steps within content ideation (SERP analysis, phrasing variations, outline expansion) but they cannot replace the four-input structure. Customer conversations, competitor gap analysis, and SME interviews produce ideas that AI cannot generate without being grounded in real inputs first.