How to Justify Maintenance Fees When the System Runs Smoothly

Question: A client (the "甲方") feels the system is so stable that there is no work for the operations team and wants to reduce or eliminate the maintenance fee. Below are several practical approaches shared by Zhihu users.

Make your presence felt. Set up alerts (e.g., disk‑usage warnings at 95%) that force the client to contact you when thresholds are crossed, ensuring you remain indispensable.

Generate workload artificially. Introduce occasional faults or scheduled service interruptions so the client must call for assistance, thereby creating visible work.

Adjust SLA and manpower metrics. Quantify the human‑day effort behind the service, present clear SLA targets, and negotiate reductions in manpower only if service levels can still be met.

Provide regular reports. Deliver system‑stability analyses that list data volume, incident counts, resolution times, and fault causes, highlighting the reliability you provide.

Offer value‑added services. Include proactive security updates, optional feature upgrades, or custom monitoring tools (e.g., a script that randomly stops a service to trigger a support call).

Know when to walk away. If the client refuses reasonable fees, consider terminating the contract to avoid unprofitable engagements.

Additional anecdote: One contributor described how a SQL log that filled the disk caused frequent alerts; by adding a cloud‑disk‑full reminder and handling the cleanup, the client remained dependent on the operations team.