You answer to your board and your regulator for systems you don’t run by hand. Your workspace puts the questions you ask first at the top: whether the applications are healthy, what needs your decision, who is handling the incident and whether the evidence is current.
Kestrel Pay, on a Friday morning.
Ngozi is the managing director and owner of a dedicated cell in Lagos. Approve a decision, follow the change her approvers are deciding, book the overdue drill or send yourself the report. Nothing you do here leaves the page.
CloudIllustrative Kestrel Pay scenarioFriday 6 November, updated 10:40
Good morning, Ngozi.
Are my applications healthy?
3 of 4 healthy. merchant-portal is slow.
- Healthy
payments-api
Shifts and settlement runs
Responding in under a second. Last change approved on 4 Nov.
- Degraded
merchant-portal
Field incident reports
Reports are slow to submit. Aspian is leading the incident.
- Healthy
alerting-bus
Alerts to the on-call team
In October, 99.6% of alerts were delivered within 60 s and 97.4% acknowledged within 5 min.
- Healthy
client-portal
Site reports for Kestrel Pay merchants
Last report run finished at 07:30 with no errors.
What needs my decision?
1 waiting for you. 1 change you follow.
- Waiting for T. Adeyemi and K. Okaforpayments-api, production
Add partial refunds to payments
Lets merchants refund part of a payment, so a partial refund no longer needs a full refund and a new charge.
Asked by D. Eze, developer. Approvers: T. Adeyemi and K. Okafor, each with a passkey
- Due by Friday 13 NovemberKestrel Pay billing
Raise the messaging usage cap to ₦300,000
Gives the extra December payments alerts room, so messaging stays inside its cap. The usage cap is ₦250,000 today; the fixed fee doesn’t change.
Asked by T. Adeyemi, IT lead. Approvers: You and K. Okafor
Who is handling the incident, and when is the next update?
Field reports are slow to submit on merchant-portal.
- Led by
- Ifeoma N., Aspian
- Working with
- Chidi A., Kestrel Pay operator
- Started
- 09:12 today
- Updates
- Here, and on WhatsApp as alert code KP-0412 with a link
Next update in
By 10:58
09:12 Detected
Alerts showed reports taking over 8 seconds to submit.
09:19 Escalated to Aspian
The cause was unclear after Chidi’s triage, so Ifeoma took the lead.
09:47 Mitigated
Aspian added database capacity. Reports submit in under 2 seconds again.
10:58 Next update
Ifeoma confirms the cause and who owns the fix.
Serious and unclear incidents are led by Aspian until responsibility is clear.
Is our evidence current?
Two items need attention.
- Stale
Export drill for client-portal is 41 days old
Its window is 30 days. The last drill on 26 Sept ran the export on an independent server in 52 minutes.
- Missing
No access review recorded for merchant-portal this quarter
Covers named users and operator grants. The quarter closes on 31 December.
Everything else is current: data location, change approvals, uptime and restore drills. Open the Trust centre
Is our team ready to run it?
Two operators authorised, one supervised, one in training.
| Operator | Runbooks | Status | Next |
|---|---|---|---|
| Chidi A. | Restart and scale, incident triage, database failover | Authorised | Renewal 14 Feb 2027 |
| Amaka O. | Restart and scale, settings changes | Authorised | Renewal 2 Mar 2027 |
| Bayo K. | Restart and scale | Supervised Write actions wait for Aspian approval | Assessment due this month |
| Funmi E. | In training: 3 of 5 practicals passed | In training | Next practical 27 Nov |
In October your team handled 11 routine incidents within its runbooks. 2 were escalated to Aspian: one serious, one with an unclear cause.
Authorisation is Aspian’s own competency assessment, not an external certification. It never lets anyone approve their own change or skip two-person approval.
Your October service report
- Uptime
- 99.97%payments-api
- Production changes
- 37each approved by two people
- Incidents
- 1311 handled by your team, 2 escalated
- Alerts delivered
- 99.6%by delivery receipt, within 60 s
- Alerts acknowledged
- 97.4%by a person, within 5 min
- Restore drill
- 9 min 12 s48 s of data loss
Effort by layer, 31 hours
- Platform operations 14 h
- Application operations 11 h
- Application changes 6 h
Spend in November so far
₦6,538,420
Fixed fee plus messaging usage, billed in naira. Messaging is at 15% of its ₦250,000 cap.
- Dedicated cell, fixed fee
- ₦6,500,000
- Messaging usage so far
- ₦38,420
- Usage cap
- ₦250,000
The cell is a fixed fee and is never paused for billing. Messaging usage can reach at most its cap plus 20%. Billing never deletes data.
Oversight you can act on in minutes.
The workspace is built for the accountable owner, not the engineer on call. Every panel ends in one clear next step, and every consequential action asks you to confirm it.
- 1
A named lead for every serious incident
Aspian leads serious and unclear incidents, and tells you who is on it and when you hear next.
- 2
Decisions written for the person deciding
What it does, the risk, the disruption and how it recovers. The technical detail is one tap away.
- 3
Evidence that flags itself
You see only what is stale or missing, with the step that fixes it. The rest is on the Trust Page.
Who answers for what
The same four layers appear in your contract, your tickets and your monthly report.
| Layer | Who handles it |
|---|---|
| Platform operationsRegions, backups, recovery, patching | Aspian runs and answers for it |
| Application operationsRestarts, scaling, first response | Your authorised operators, with Aspian oversight and step-in |
| Application changesWorkflow fixes, new features | Scoped separately, built by Aspian or your developers |
| Business decisionsApprovals, risk, spend | You decide. Aspian gives advice and evidence |
Want to see the evidence behind it? Visit the Trust centre or ask us for a walkthrough.
See what shipped without approval in your last 90 days.
Install the GitHub App and the scan arrives in minutes: every repo scored, every unreviewed merge and unapproved deploy named. Nothing is blocked until you choose.
Ship like a startup. Prove it like a bank.