The problem
An automation is live and the team has moved on. Nobody is watching the failure queue or the model output quality.
A document format changes, an API version is deprecated, or a model is updated, and the workflow starts producing bad writes.
The person who built the automation is the only one who can fix it. Every change becomes a bottleneck.
The business case was based on hours saved, but nobody measures whether those hours are still being saved.
What you get
Outcomes
- Monitoring and a monthly report on volume, failures and manual touches.
- Exception review with a prioritised fix list.
- A small monthly improvement block for prompts, rules and integrations.
- A documented change log and fewer single-person dependencies.
Deliverables
- Workflow monitoring and an error review.
- A monthly report: volume, success rate, manual touches and time saved.
- Exception triage and a prioritised backlog.
- Up to one agreed improvement block per month.
- A change log and updated runbook.
- A quarterly review of whether the workflow should be simplified or retired.
Fit and boundaries
This is a good fit when
- A live automation with a real business dependency.
- A team that wants monitoring and a monthly improvement cadence.
- A named owner on your side.
- A workflow that already has logs and a runbook.
This is not a fit when
- A project that has not launched yet. Start with the build.
- A request for 24/7 incident response or a formal SLA. That is a different support contract.
- An automation with no logs or ownership. Fix the handover first.
- A one-off task with no ongoing volume.
How it works
- 01
Handover check
The runbook, logs, alerts and ownership are confirmed before the retainer starts. If the handover is incomplete, the first month fixes that.
- 02
Monitor the live workflow
Volume, failures, manual touches and latency. The report is short and written for the person who owns the process.
- 03
Review exceptions monthly
The cases that needed a human are the best source of the next improvement. They are reviewed, grouped and prioritised.
- 04
Improve or retire
Each month has a small improvement budget. Every quarter we check whether the workflow still earns its place.
Pricing and timing
Fixed-price engagement
€750 to €1,500 / month Monthly · EUR, VAT not applicable at present
Monthly retainer, excluding third-party API and infrastructure costs. 30 days notice.
- Written scope and acceptance criteria before work starts.
- 40 to 50 percent deposit reserves the calendar slot.
- Milestone invoices and 14-day payment terms.
- A checkpoint where either side can stop with paid work settled.
Questions
What is included each month?
Monitoring, exception triage, a short report and one agreed improvement block. The block is scoped at the start of the month so it does not expand indefinitely.
What is the response time?
Business-hours response, normally within one working day for a production issue. This is not a 24/7 SLA. If you need one, it is a separate agreement.
Can we cancel?
Yes, with 30 days notice. The runbook and change log stay with you, so cancelling does not strand the workflow.
Do you take over an automation someone else built?
Sometimes. I start with a short readiness review of the logs, the runbook and the failure paths. If the system is not observable, the first task is to make it observable.
Page last updated: October 2026