We're part of an 80-person development agency that juggles various client projects, and honestly, our support workflow is quite chaotic. Right now, here's where we stand: CloudWatch/monitoring alerts land in email inboxes, often belonging to developers who haven't worked on the project for months or have left the company entirely. Clients can't submit tickets directly; instead, they text whoever they remember—sometimes a former developer, an old project lead, or even our CEO. This has turned into a game of 'telephone' to figure out who should handle each issue. Clients also lack visibility into their infrastructure status, leaving them in the dark and frustrated. We're missing critical alerts and wasting hours just trying to identify the right person for the job. My questions for you all are: How do you handle alert notifications from client infrastructure? What's your process for clients to report issues? How do you route relevant alerts and requests to the appropriate team members? And what tools do you use? Is it as chaotic as what we're dealing with?
4 Answers
The approach can greatly depend on your ability to implement changes within the agency. If you can showcase how lack of incident response is costing money, you might secure some budget for improvements. The tech stack also matters—a custom solution can be created based on your specific needs. I've set up tailored incident processes in the past and have seen that user adoption is crucial for success.
We use LogicMonitor for setting up customer-specific alert rules and routing. For our incoming tickets, we rely on JIRA Service Desk, which integrates well with our workflow. JIRA tickets paired with PagerDuty helps us in routing and notifying the right internal teams effectively.
It sounds like you may be facing the consequences of relying too much on developer memory for client assignments. After delivering projects, consider offering support contracts so that you're better prepared with resources. The way you’re currently operating could lead to more chaos, and it's clear that some restructuring is needed.
What you're experiencing seems like a classic case of communication breakdown. The solution, though not exciting, is effective: establish a dedicated on-call rotation and implement a proper ticketing system instead of relying on emails. Think about using tools like PagerDuty or OpsGenie for alerts and a ticket management system like JIRA or Linear. Setting up a client portal for tracking tickets and status can also enhance transparency and organization.
Related Questions
Can't Load PhpMyadmin On After Server Update
Redirect www to non-www in Apache Conf
How To Check If Your SSL Cert Is SHA 1
Windows TrackPad Gestures