An AI Tool That Analyzes and Compares Multiple Sources — Here's How to Use It

Compare YouTube videos, find conflicting opinions, and uncover useful insights without manually analyzing every source.

You have five YouTube videos on the same topic. Each is an hour long, and every creator has a slightly different opinion.

You could watch all five, take notes, and try to compare everything yourself.

Or you could use Google NotebookLM to analyze all five sources together.

Here's a practical example of how it works.

I uploaded five YouTube courses about affiliate marketing in 2026 into NotebookLM.

The videos covered similar topics: choosing profitable niches, finding affiliate programs, driving traffic, and generating sales.

I wasn't looking for five separate summaries.

One comparison prompt generated a structured report showing agreements, differences, and unique insights across the sources.

Josh Burns Tech reported that video content generates more than 90% of his $40,000 monthly affiliate revenue.

Marketer Milk argued that SEO-driven blogs can be more effective because YouTube viewers sometimes search for products directly instead of clicking affiliate links.

Alex Micol presented a different approach: selling through TikTok Shop without building a traditional website.

Three different strategies, presented side by side.

Instead of simply summarizing what each creator said, NotebookLM made the differences between their approaches easier to identify.

Adam Enfroy criticized Amazon's relatively low commissions and short attribution window.

Josh Burns Tech recommended Amazon Associates as a starting point for beginners because of its accessibility and established customer base.

The difference becomes clearer when you compare their priorities: higher commissions versus a lower barrier to entry.

Neither position automatically applies to every business. But the comparison reveals what each recommendation is based on.

Santrel Media and Marketer Milk emphasized the importance of owning a website for credibility, affiliate approvals, and building an email audience.

Alex Micol described an approach that relies on TikTok Shop instead.

NotebookLM identified these competing recommendations without requiring me to manually organize the five courses.

These findings reflect the claims identified in the uploaded video transcripts, not independently verified earnings or business results.

The analysis highlighted several additional techniques:

Marketer Milk differentiated between products people would like to have and products that solve an immediate, important problem.

Several creators discussed contacting brands directly to negotiate custom affiliate commissions instead of relying exclusively on public affiliate networks.

Santrel Media described using ManyChat to automatically deliver resources and affiliate links when users comment specific keywords on short-form videos.

Adam Enfroy discussed structuring article headings around specific questions to make content more discoverable in AI-powered search.

Josh Burns Tech highlighted the importance of presenting audience demographics and performance metrics when pitching brands.

These ideas were scattered across different courses. NotebookLM organized them into one comparison, making it easier to identify which techniques deserved a closer look.

You can repeat this workflow with almost any research topic.

Go to notebooklm.google.com and create a new notebook.

Choose three to five YouTube videos about the same topic.

Click Add sources → YouTube and paste the video URLs.

NotebookLM uses available transcripts, so choose public videos with accessible captions.

Once your sources have been processed, paste the following prompt into the NotebookLM chat.

Role: Act as a research analyst specializing in comparing multiple sources. Context: I have uploaded several sources covering the same topic. Task: Analyze all sources together and identify: 1.

Ideas and recommendations that multiple sources agree on. 2. Direct contradictions between sources.

3. Unique insights mentioned by only one source. 4.

Important details that some sources overlook. 5. Practical conclusions supported by the material.

Constraints: - Use only the uploaded sources. - Include source citations for major findings. - Do not invent missing information. - Distinguish actual contradictions from different perspectives. - If evidence is insufficient, say so explicitly. Output Format: 1. Comparison table.

2. Shared conclusions. 3.

Contradictions with supporting evidence. 4. Unique insights.

5. Key takeaways. 6.

Questions worth researching further. Enter fullscreen mode Exit fullscreen mode Step 4. Review the Results

NotebookLM will organize the findings into a structured comparison.

Use the source citations to verify important conclusions before making decisions based on them.

This workflow isn't limited to YouTube courses or affiliate marketing.

Compare lectures from different educators to understand how they explain the same concept.

Analyze several sources to identify recurring arguments, conflicting evidence, and missing information.

Compare product reviews to see which advantages and disadvantages multiple reviewers consistently mention.

Compare expert strategies, case studies, or industry analyses before deciding which approach deserves further investigation.

Analyze multiple tutorials or articles to identify useful perspectives and research gaps before developing your own content.

The most useful part of NotebookLM isn't simply getting a shorter version of a long video.

It's being able to ask one question across multiple sources and see how their answers compare.

“What do these sources agree on, where do they disagree, and what important information am I missing?”

That turns a collection of videos into a structured starting point for research.

And you can apply the same process to almost any topic.

Want to discover more useful ways to work with AI?

Join WORKKIT on Telegram to explore practical AI tools, ready-to-use prompts, step-by-step workflows, and productivity shortcuts.

From research and content creation to design, freelancing, and everyday tasks — find AI solutions you can actually put to work.