The Scenario
You're the lead data architect at a SaaS company and your team is migrating to a new dashboard infrastructure next month.
Before anything moves, you need a complete inventory of the 25 cached datasets defined in your Keen IO project — dataset names, index fields, query types, timeframes, creation dates. The purpose is a formal review: which datasets are still actively useful, which ones were created for a dashboard that no longer exists, and which ones are so old they should be rebuilt from scratch rather than migrated.
Nobody sent you a list. There's no internal documentation. The last person who knew these configurations left eight months ago, and what they left behind is a Confluence page that says "see Keen IO for details."
The bad version:
- Open the Keen IO API docs, find the cached datasets endpoint, and write a GET request to list all of them.
- Parse the response JSON — each dataset has a nested "query" object with its own subkeys — and copy the relevant fields into a workbook manually.
- Calculate creation dates, figure out which ones are more than 6 months old, and mark them by hand.
The migration kickoff is in three weeks. You have a backlog of infrastructure work that's already competing for your time, and a manual audit of 25 API objects is the last thing that should be on your plate.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Excel workbook. It connects to Keen IO and can fetch structured API responses — like a cached dataset list — and write them to your workbook with whatever conditional logic you need applied inline.
Open a blank audit worksheet and try this:
List all cached dataset definitions in my Keen IO project and write each one to this sheet with columns for dataset name, index field, query type, timeframe, and creation date. Flag any dataset created more than 6 months ago by marking column F as "Stale—Review", leaving all others blank.
What You Get
- One row per cached dataset across all 25 definitions
- Columns: Dataset Name (A), Index Field (B), Query Type (C), Timeframe (D), Creation Date (E), Status Flag (F)
- Column F populated with "Stale—Review" for any dataset whose creation date is more than 6 months before today, blank for the rest
- If a dataset's configuration is missing a field (e.g., no explicit timeframe set), SheetXAI writes "—" rather than leaving the cell empty, so your audit grid stays readable
What If the Data Is Not Quite Ready
You need to filter the audit to only "count" and "sum" query types
List all cached dataset definitions in my Keen IO project. Write only datasets with a query type of "count" or "sum" to this sheet — include dataset name, index field, query type, timeframe, and creation date.
You want the full list plus a summary table on a second worksheet
List all cached dataset definitions in my Keen IO project and write the full details to Sheet1 with columns for dataset name, index field, query type, timeframe, and creation date. On Sheet2, write a summary showing how many datasets exist per query type, sorted by count descending.
Some datasets have the same index field — you need to group and flag duplicates
List all cached dataset definitions in my Keen IO project and write them to this sheet. In column G, mark any dataset that shares its index field with another dataset as "Duplicate Index Field—Review". Leave others blank.
Kill chain: pull list, flag stale, flag duplicates, and write a migration recommendation column
List all cached dataset definitions in my Keen IO project and write each one to this sheet with dataset name (A), index field (B), query type (C), timeframe (D), and creation date (E). In column F, mark any dataset older than 6 months as "Stale". In column G, mark any dataset sharing an index field with another as "Duplicate Index". In column H, write "Rebuild" if a dataset is both stale and has a duplicate index field, "Review" if only one flag applies, and leave blank if neither.
One prompt covers the full decision tree — so what you get back is an audit-ready document, not a raw list you'd spend another two hours annotating.
Try It
Get the 7-day free trial of SheetXAI and open a blank Excel workbook, then ask it to pull and annotate your full Keen IO cached dataset inventory before your next migration planning session. For related Keen IO tasks, see the schema registry spoke or the full hub overview.