Multisource Content-Gap Research Workflow
Mine proven content and audience conversations to find differentiated angles.
- Difficulty
- Moderate
- Time to result
- ~days to results
- Steps
- 7
- Confidence
- 96%
The Multisource Content-Gap Research Workflow begins with videos or channels that have already performed in a target vertical, then expands beyond those sources. An agent extracts the source content and researches related discussions across YouTube comments, Reddit, and other relevant locations. It clusters audience pain points, questions, objections, and overlooked use cases while also documenting what existing creators failed to address. The creator then chooses a differentiated angle positioned at the intersection of demonstrated interest and incomplete coverage. The same research package can include a proposed script, visual references, and a B-roll gallery, making the output immediately useful for production. Once refined, the process can be converted into a reusable skill so each new topic receives the same broad research treatment rather than an improvised search.
Origin
Kevin Hudson demonstrated this Manus-based content research process on Marketing Against The Grain.
Core principles
- 01Start with content that has already demonstrated audience interest.
- 02Audience comments reveal needs that published content may not satisfy.
- 03Use several source types to avoid copying one creator's perspective.
- 04Look for omissions and unresolved pain points, not merely popular keywords.
- 05Translate research into a differentiated angle and production-ready assets.
- 06Save a successful research process for repeated topic analysis.
How to run it
- 1
Seed with Proven Content
Provide the agent with relevant videos, channels, or topics that have demonstrated audience interest. Define the target vertical and intended audience.
Pro tip Choose several sources with different perspectives rather than one market leader.
Watch out Popular content is an input to research, not permission to copy its angle.
- 2
Extract the Existing Coverage
Summarize the claims, examples, tools, structure, and conclusions already present in the seed content. Identify areas requiring additional verification.
Pro tip Ask for a source-linked research report that remains easy to audit.
Watch out Transcript extraction may miss important visual evidence from the videos.
- 3
Expand Across Audience Sources
Search related Reddit discussions, YouTube comments, and other audience conversations. Capture specific pain points, questions, disagreements, and overlooked use cases.
Pro tip Preserve representative comments alongside each insight.
Watch out A loud comment is not necessarily a widespread customer problem.
- 4
Map Content Gaps
Compare audience needs with what the existing content covered. Highlight unresolved questions, missing examples, weak explanations, and underserved viewpoints.
Pro tip Prioritize gaps supported by repeated signals across independent sources.
Watch out Novelty without audience relevance is not a useful gap.
- 5
Choose a Differentiated Angle
Develop a content premise that addresses a meaningful gap while remaining connected to proven audience interest. Explain clearly how it differs from existing coverage.
Pro tip Frame the angle around a specific audience problem or decision.
Watch out Changing the headline without changing the substance is not differentiation.
- 6
Prepare Production Assets
Turn the research into an outline or script and gather suggested visual evidence, examples, and B-roll. Keep every major claim traceable to the research package.
Pro tip Generate B-roll ideas during research so visuals support the argument.
Watch out Do not let generated visuals imply evidence the sources did not establish.
- 7
Save the Workflow
After refining the report through feedback, convert the process into a reusable skill. Use a new topic as the input for each subsequent run.
Pro tip Retain the source-diversity and content-gap requirements in the saved skill.
Watch out A reusable workflow still requires editorial judgment before publication.
In the wild
A creator supplies several successful videos about AI agents. Manus extracts their coverage, reads related subreddits and video comments, identifies repeated questions and pain points, and lists concepts the videos did not adequately cover. It then proposes a differentiated episode structure and visual assets.
→ The creator receives a research-backed angle rather than a generic imitation of existing videos.
A marketing team supplies a channel and asks an agent to analyze its performance over time while comparing topics, audience responses, and gaps across the vertical. The team uses the resulting dashboard to identify themes with proven demand but weak current coverage.
→ Editorial planning is informed by both channel performance and audience needs.
Common mistakes
Researching Only Published Content
Videos show what creators chose to discuss, while comments and community conversations reveal what audiences still need.
Confusing Popularity with a Gap
A popular topic may already be saturated; the workflow must compare demand with incomplete or weak coverage.
Accepting the Agent's Angle Uncritically
Generated differentiation still needs human review for originality, factual support, and relevance to the creator's audience.
Is it for you?
Best for
It is best for marketers and creators developing videos, articles, or campaigns in active categories with accessible audience conversations.
Not ideal for
It is not ideal for confidential or extremely new topics with little public content, commentary, or audience evidence.
From the transcript
“it goes and reads a bunch of different subreddits, finds YouTube comments across all these different videos and isolates pain points from all of those”
“And there's like content gaps, uh like what people didn't cover.”
“You're finding where the arbitrage and the market is for what to create, and then you're also using the same tool to kind of repurpose…”
From the episode
The AI Skill Ladder (Beginner → Workflow Builder)