Optimizing Service Delivery
The Platform Engineering team of a telecom client trusted us to enhance their container-as-a-service offering used by more than 20 tenants. Most of their customers were happy with the quality of service they received, but this was not the case for the time taken to resolve their issues.
Before engaging us, several tactics were used:
- Priority Classification: an "inner circle" of tenants was established, with the expectation that engineers would address their requests with heightened urgency due to their privileged status
- Streamlined Communication: a new communication channel was established to ensure tenants connected with the right individuals from the outset. Engineers were asked to redirect tenants to a dedicated team consisting of Product Owners, Scrum Masters, and a Service Lifecycle Manager
- Process Refinement: robust processes were established to guide tenants when they need support from the platform team, employing the classic approach of raising a ticket for every task
The Problem
Despite these proactive measures, over 1,200 unresolved work items were unresolved when we conducted our initial assessment. Shockingly, some of these issues had been lingering for as long as 33 months. The average age of these unresolved items reached a staggering 135 days.
The Solution
In the face of this daunting challenge, we decided on a multi-faceted approach to regain control:
- Prioritization Revisited: a concerted effort was made to re-focus the team on long-standing customer issues. Many of these were already resolved but had slipped through the cracks without proper closure
- Decluttering Effort: identified and closed items that were no longer relevant, immediately reducing the size of the backlog
- Tagging Accuracy: recognizing the impact of incorrect tags, the team implemented a comprehensive dashboard aimed at detecting and rectifying mislabelled issues promptly, avoiding future missteps
- Governance Dashboard: a dedicated dashboard and regular reporting were introduced to monitor the service requests, keeping cycle time and average age at the forefront of the team's attention as two crucial KPIs
- Agile Planning: to deal with the incoming requests effectively and with a sense of urgency, the frequency of planning meetings was intensified from twice a week to five times. This enabled the swift review, clarification, approval and allocation of incoming service requests to the most suitable teams
- Architect Engagement: architects were brought into the planning meetings at least once a week, reducing wait times for their input and expediting request processing
- Service Request Guides: to streamline the request process, guides were crafted for frequently requested services, ensuring that tenants knew precisely what information to provide, thus minimizing clarification time
The Result
In just three months, the Platform Engineering Team achieved remarkable results. The average cycle time was slashed by 35%, bringing down from 135 days to 87 days. Within the subsequent couple of months, they further improved this figure to a staggering 73% of the original, reducing it to a mere 36 days. The average backlog size plummeted from over 1,200 work items to fewer than 300.
This success story stands as a testament to the team's dedication and adaptability in the pursuit of optimizing service delivery, ultimately leading to more satisfied tenants and a more efficient operation.