Case study #3
How I Use AI to Research Content for YouTube and Blog Posts
Project Summary: AI Content Research
The challenge
Creators know that in-depth research is the key to standing out and growing an audience. But it takes so much time.
The solution
A structured AI research workflow using Claude Cowork to gather, analyse, and synthesise information from multiple sources into a single comparison document, ready to form the basis of a blog post or YouTube video.
Tools used
Claude Cowork, Claude
The Problem I was Solving
Good content requires good research. If you want to produce a YouTube video or blog post that genuinely stands out, you need to go deeper than the surface-level articles that already exist. That means comparing multiple sources, checking what real users are saying, and identifying angles your competitors haven’t covered.
The problem is that thorough research takes hours. Visiting multiple websites, reading reviews, cross-referencing information, and then organising it all into something usable is a significant time investment before you’ve written a single word.
I wanted to find a faster way to do this without cutting corners on quality.
My Approach
Step 1 — Define the research brief
Before touching any AI tool, I like to write a clear brief describing exactly what I need. In this case I wanted a comparison of eight Etsy SEO and product research tools, covering what each one does, its key features, pricing, and what real users are saying on review sites and forums. Having a specific, well-defined brief is what separates useful AI research from vague, generic output.
Step 2 — Set up the Cowork task
I opened Claude Cowork on the desktop app and pasted my brief as a single task. Cowork immediately broke it down into three steps: research all eight tools, compile the findings into a Word document, and verify the accuracy of what it had gathered. The important part of this step is clearly explaining in plain English what is required and then letting it get to work.
Step 3 — Cowork researches in parallel
This is where the time saving becomes real. Rather than visiting eight tool websites one by one, Cowork ran parallel web searches across all of them simultaneously, gathered user sentiment from review platforms, and pulled together a 21-page document full of structured research. A task that would have taken several hours manually was completed in minutes.
Step 4 — Review and refine the output
Once Cowork had produced the comparison document, I reviewed it in Claude to sense-check the findings and add critical context. One honest limitation worth noting: Reddit was not directly accessible during the research session, so community sentiment was inferred from blogger summaries rather than live forum data. For a published piece, I would verify the most important claims manually before hitting publish.
The Results
What the content research process produced:
1. The Report Output
In under 30 minutes, Cowork produced a structured 21-page comparison report covering all eight tools. For each one it documented what the tool does, its key features, current pricing, and an honest assessment drawn from independent review sites and seller communities.
2. Key Findings
eRank is the strongest all-round choice for most sellers. It has the best free tier of any dedicated Etsy SEO tool, the widest feature set at its price point, and is trusted by over one million sellers.
Alura is the best all-in-one platform for sellers who want everything in one subscription, including A/B testing — a feature no direct competitor currently offers.
EverBee is the best tool specifically for product research and niche discovery, with revenue estimates that are among the most accurate in the category.
Etsy’s own Marketplace Insights is worth using alongside any third-party tool because it draws on native Etsy search data — the only source of truly authoritative keyword figures.
ProfitTree fills a gap none of the others address: real-time profit tracking that factors in all fees, ad spend, and production costs automatically.
3. Recommendation Table
The report also produced a recommendations table mapping each seller profile — from brand-new sellers on a tight budget through to high-volume print-on-demand shops — to the specific tool combination that best fits their situation.
That kind of structured, actionable output would typically take several hours to produce manually. Here it emerged as part of the same research task. Read the full research report.
What You Could Do With This
If you create content for a blog or YouTube channel and find that research eats up most of your preparation time, this approach is worth exploring. A clear brief, a single Cowork task, and a review pass in Claude can take you from blank page to structured research document in under an hour.
If you’d like help building a content research workflow for your own business, get in touch here.
