For L2/L3 support and on-call engineers
Investigate customer incidents across databases
A customer issue rarely lives in one table. DataTug is building investigations that keep the order, the payment, the carrier’s answer and the log line in one case, so the trail survives the shift change.
Starts here: DevelopersHelpdeskSRE and on-callEngineering managers
Also useful: AI agent builders (also useful)Security and compliance (also useful)QA and testers (also useful)


The problem
The order is in one database, the payment in another, the carrier booking behind an API and the error in a log nobody has opened. The investigation lives in six tabs and a chat thread, with IDs copied between them by hand.
When the shift ends the next engineer starts again, and when the incident is closed its findings stay in a ticket that nobody will find the next time the same thing happens.
Illustration · sample story, not a capture
Following order 11045 to the cause
- Northwind · Orders11045 · BOTTM · not shipped
- PaymentsCaptured 18 MayOK
- Carrier APIbooking: nullproblem
- Logsroute-17-2 expired 19 Mayproblem
- FindingPaid, not booked: the route had expiredneeds attention
- CaseEvidence and finding saved for the team
Static illustration from the sample story of order 11045, not a capture of the product.
Before and after
Who: L2/L3 support and on-call engineers
Today
3 tools2 hand-offsno shared password
- Copy the order ID from the support tool into one SQL client, then another.
- Rebuild the joins between orders, payments and shipments from memory.
- Paste screenshots and query results into the ticket as evidence.
- Close the ticket; the reasoning behind the fix is lost.
With DataTug
1 workspace0 hand-offsno shared password
Some steps are planned or in development.
- Start from the order and follow its related records, keeping the identifier and the joins in view.Related records in the web appAvailable
- Read the logs where they live, filtered to the same order and time window.Log viewerPreview
- Keep queries, evidence and notes in one case that anyone on the team can resume.Detective CasesIn development
- Carry the case’s context to the next tool or agent instead of re-explaining it.Context BagIn development
- Close with an incident history that shows what the team knew at each step.Incidents in DataTug projectsIn development
Each step names the part of DataTug it relies on and that part’s status today. “Planned” and “In development” steps describe where the product is going, not what you can use now.
Where it meets the other solutions
Data Understanding MeaningGraph is planned to recognise Orders.OrderID and the carrier’s order_id as the same order, so following related records can cross systems. Today you map that meaning by hand with column meanings.
Safe Data Access The Perimeter is designed to let an investigation read production, masked and audited, without anyone holding a standing admin login.
Data Quality & Monitoring An investigation’s root cause is meant to become a monitoring rule, so the same problem is spotted before the next customer has to ask.
Try it on sample data
The public demo project opens in the browser with sample Northwind and Chinook data. No sign-up.
Explore the demo project

