Malaysian tower company EDOTCO Group built NaPA GPT on Microsoft Azure so network planners can query operational and geospatial data in plain language and get ranked site shortlists without waiting on the data science team.
- First reported
- news.microsoft.com · 17 Aug 2026
- Added
- 13 Oct 2026
- Source checked
- 13 Oct 2026
What they did
EDOTCO first pulled planning, geospatial and operational data into one platform called NaPA, then added a conversational layer connected to those datasets. Planners type criteria such as coverage gaps or where investment would be strongest, get a ranked list of opportunity areas, and refine it with follow-ups like prioritising sites near transport corridors. Selected results are shared with government agencies through dashboards.
What happened
Microsoft says planners now get a ranked shortlist within minutes, where preparing the data used to take the data science team three days to a week.
Keep in mind
This is a vendor feature article. It does not say how often the shortlists match planners' final choices or which model powers the tool.
Try it with your agent
Here are our datasets: <coverage data>, <site list> and <demand or population data>. Rank the top <number> areas where a new <site or branch> would add the most value, using <criteria>. Show the score breakdown for each area, note any data gaps, and let me refine the ranking with follow-up conditions.
Replace the <angle brackets> with your own details.
Original source
news.microsoft.comVendor case studyPublished by the company that makes the agent. Read it as a claim.
Tools for this kind of work
- RepoExtract tables and text from complex or scanned PDFs with Docling
- RepoSearch the web and scrape sites into clean data with Firecrawl
