The Scenario
You're a market analyst at a fintech consultancy. A client has asked for a briefing on "AI-powered drug discovery clinical trials." Specifically, which publications are covering it, which angles are getting traction, and which articles are worth reading.
The problem is keyword search on this topic returns noise. Articles that contain "AI" and "drug discovery" and "clinical" without being about the intersection at all. What you need is semantic relevance — the 30 articles that are most conceptually on-topic.
Your client call is in 90 minutes.
The bad version:
- Run a Perigon keyword search with broad terms. Get 300 results. Open each one. Skim to assess relevance. Build a shortlist of 30 manually.
- Copy titles, sources, and URLs into an Excel worksheet. Format columns. Remove duplicates.
- Realize the session took 70 minutes. You have 20 minutes left to actually read the articles.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Excel workbook. Through its Perigon integration, it can run a vector search against Perigon's news index — returning results ranked by semantic relevance — and write them directly into your workbook.
Perform a Perigon semantic search for "AI-powered drug discovery clinical trials" and populate my Excel Research Digest sheet with the top 30 matching article titles, sources, dates, and URLs.
What You Get
- 30 rows in the Research Digest worksheet, each representing one semantically relevant article.
- Column A: title. Column B: source name. Column C: published date. Column D: URL.
- Results ranked by semantic relevance to your query, not by keyword overlap.
- Ready for review and delivery without additional filtering.
What If the Data Is Not Quite Ready
The query should come from a cell in the workbook
Read the research query from cell B1 of the Config worksheet. Use Perigon semantic search with that query and write the top 30 results to the Research Digest worksheet with title, source, date, and URL.
You want to filter to a specific date window
Use Perigon semantic search for "AI-powered drug discovery clinical trials". Filter results to articles published in the last 60 days. Write the top 30 matching results to the Research Digest worksheet with title, source, date, and URL.
You need to deduplicate against articles already in the workbook
Use Perigon semantic search for "AI-powered drug discovery clinical trials" and retrieve the top 50 results. For each result, check if the URL already exists in column D of the Existing Research worksheet. Write only new articles to the Research Digest worksheet — title, source, date, and URL.
Pull semantic results, deduplicate by source, and rank by coverage depth in one pass
Use Perigon semantic search for "AI-powered drug discovery clinical trials". Retrieve the top 100 results. Deduplicate by source domain — keep only the highest-ranked article per source. Write the final set to the Research Digest worksheet with title, source, date, and URL. In column E, note how many total results that source had in the original 100.
Ask for the search, deduplication, and source frequency analysis in one prompt.
Try It
Get the 7-day free trial of SheetXAI and open any Excel workbook where you're building a research brief or monitoring a technical topic, then ask it to run a Perigon semantic search and populate the results. See the full Perigon integration overview or explore importing Perigon story clusters for content planning.