Monday stand-up, by tool
The head of ops runs a stand-up. Each person reports on their queue: onboarding referrals from the review tool, monitoring alerts from the vendor's screen, disputes from the processor portal, complaints from the helpdesk, returns from the provider, data requests from a spreadsheet. Each person gives a rough number.
Nobody can say how old the oldest item across everything is, whether any queue is getting worse week on week, or whether the team is on the right work. When the CEO or the partner bank asks how operations are doing, the head of ops spends an afternoon building slides.
Why there is no single view
Each queue was set up when it was needed, with whatever tool was to hand. Nobody designed the whole.
- Different tools have different ideas of open, pending and closed.
- Some queues have timestamps for received and resolved; some do not.
- Targets exist informally, in people's heads.
- Ownership is visible in some tools and not others.
- Reporting in each tool is basic and does not combine with the others.
Managing without seeing
Without a combined view, the head of ops manages by anecdote. Busy queues get attention because people complain; quiet ones age without anyone noticing. Staff planning is guesswork. Reports to leadership and partners are slow to produce and hard to repeat consistently.
An ops dashboard across every queue
What we build brings each queue's items into one place without moving the work out of the tools people use.
- Each queue source is connected through its API or export: your review tool, monitoring vendor, card processor, helpdesk, provider reports and any spreadsheet that has to stay for now.
- Items are mapped to a shared model: type, received time, status, owner, customer and link back.
- Statuses from each tool are mapped to shared states such as waiting on us, waiting on customer, and done, agreed with you.
- Targets for each queue type are stored as settings you control, so ages can be shown against them.
- The dashboard shows each queue's volume, oldest item, items past target, and ownership, and the trend over recent weeks.
- Clicking an item opens it in its own tool.
- A weekly summary can be sent to leadership, or exported for a partner pack.
| Queue | Source | Shared measure |
|---|---|---|
| Onboarding referrals | Review tool | Age since referral, owner |
| Monitoring alerts | Monitoring vendor | Age since raised, owner |
| Card disputes | Processor or dispute tool | Days to next stage date |
| Complaints | Helpdesk or complaints log | Age against your policy's deadlines |
| Returned payments and unallocated funds | Provider data and ops queue | Age since received |
If you already use a BI tool such as Metabase or Looker Studio, the dashboard can live there. If not, a simple web page is often enough.
Stand-up with one screen
Getting the shared states right is most of the work. A dispute waiting on the card scheme is not the same as a complaint waiting on a final response letter, and a referral waiting on the customer should not count against your team. We agree these definitions with you queue by queue, write them down, and show them on the dashboard, so a number on the screen always means the same thing to everyone reading it.
The stand-up starts with the dashboard. Everyone sees which queues are growing and which items are oldest. The head of ops moves people to where the work is. Leadership and partner reporting uses the same numbers, produced from the dashboard rather than rebuilt in slides.
Is this your operations view?
- Queue numbers are reported verbally at stand-up.
- Nobody knows the oldest item across all queues.
- Targets are informal.
- Ops reports for leadership take hours to build.
- Some queues only get attention when someone complains.