Quality committee, first Monday of the month. The technical director knows many reports get cancelled. She doesn’t know why.
Her challenge is deciding where to act: train, change a process or fix the app? The villain is the spreadsheet: months of cancellations exported and sorted by hand.
And the twist: the problem wasn’t technical everywhere. In one branch it was. In two others, it was data entry.
The question
«Group this year’s cancellations by reason and show me how they evolve.»
You can also ask
- «Break them down by branch and by inspector.»
- «Which reason grew the most in the last three months?»
How the assistant answered
Group this year’s cancellations by reason and show me how they evolve.
Account, policy or user data: 25%. Meter data: 22%. Unclassified: 14%. CO measurements: 10%. Appliance data: 9%.
Data-entry errors add up to more than half. CO measurements are the fastest-growing reason. “Unclassified” drops as the field gets used.
Here is the monthly chart of the top five.
Based on a real case; names and figures changed
What it answers and why it matters
- 1
Reasons with their weight and trend
Cancellations are grouped into 14 base reasons, such as meter data, tightness test or CO measurements. You see how much each weighs and how it changes month by month.
- 2
Where each problem concentrates
The breakdown by branch and inspector separates technical from administrative. In the example, the South branch concentrates CO, isometric and meter; North and Center, account or policy errors.
- 3
Concrete actions, with a clear limit
Reinforce CO measurement in one branch, validate the account in the app, or ask whoever cancels to pick the reason. The reason is what the person who cancelled chose, not a technical audit.
Read-only and with your same permissions. Gas users’ personal data never leaves RETGAS. See privacy
The director didn’t need more data. She needed to know where to act first. The assistant draws the charts with the data it queries in RETGAS.
