Operations Manager

Operations management owns the ongoing, repeatable delivery of a product/service at the required quality and volume — distinct from Project Management's bounded, one-time work. Operations is what runs continuously after a project delivers a new capability into production/business-as-usual.

Process design and improvement

• Process mapping — documenting the actual steps a process follows (not the idealized version) is the prerequisite for improving it; a surprising amount of operational waste is only visible once the real process, including its informal workarounds, is written down honestly.
• Bottleneck identification — in any multi-step process, the slowest step caps the whole process's throughput regardless of how fast every other step is; improving a non-bottleneck step doesn't increase overall throughput, the same underlying idea Critical Path applies to project schedules.
• Continuous improvement (Kaizen, lean, Six Sigma) — different named traditions, sharing the core idea of incremental, evidence-based process refinement over time rather than one-off redesigns, with a bias toward eliminating waste (unnecessary steps, waiting, rework) rather than just working faster within a flawed process.

Capacity planning

Matching resource availability (staff, equipment, infrastructure) to expected demand ahead of time — under-provisioned capacity causes missed SLAs and burnout; over-provisioned capacity is a standing cost with no offsetting value. Demand is rarely flat, so capacity planning usually needs to account for known seasonality/peaks, not just an average.

SLAs and service quality

A Service Level Agreement (SLA) states a measurable commitment (response time, uptime, resolution time) and what happens if it's missed — useful only if the metric it measures actually reflects the outcome that matters to the customer/user, not merely what's easy to measure. Distinguish an SLA (a commitment, sometimes contractual) from an SLO (an internal target, often stricter than the external SLA, giving margin before an actual SLA breach) and an SLI (the specific measured indicator behind both).

Vendor and supplier management

Ongoing operational dependencies on third parties need active management, not just a signed contract: tracking actual performance against agreed terms, a defined escalation path when a vendor underperforms, and contingency planning for a critical vendor's failure or exit — a single point of failure in the supply/vendor chain with no contingency plan is a common, avoidable operational risk.

Common pitfalls

• Improving a non-bottleneck step and reporting it as a win — doesn't move overall throughput, and can even make the true bottleneck's queue longer by feeding it faster.
• An SLA that measures the wrong thing — e.g. average response time when the customer experience is actually driven by worst-case (tail) response time; hitting the average SLA while a meaningful fraction of cases badly miss it can still mean real dissatisfaction.
• Capacity planned to the average, not the peak — leads to predictable SLA breaches during known busy periods that a slightly more conservative plan would have absorbed.
• No contingency for a single critical vendor — an operational dependency with no fallback plan turns a vendor's outage or exit into the organization's own crisis with no prepared response.
• Process documentation describing the idealized process, not the real one — makes any improvement effort built on it miss the actual workarounds and waste that exist in practice.

Learn more

• ITIL — a widely-used IT service-operations framework covering much of this territory for technology operations specifically.
• Critical Path for the bottleneck/dependency logic operations shares with project scheduling.
• Project Management for the bounded-work counterpart to ongoing operations.