What steps or processes do you use to identify areas where you can make a meaningful improvement at your company? Do you follow industry trends and look for ways to introduce new tools, or do you start with a specific problem and then find the right solution? I feel this is one of the biggest gaps holding me back from advancing in my career, so I'd appreciate practical advice on how to spot and prioritize worthwhile improvements.
5 Answers
Use data to validate what you hear. Depending on your role, measure things like delivery lead time, deployment frequency, failure and recovery rates, alert volume, ticket counts, downtime, compute spend, or developer time spent on manual work. Rank opportunities by measurable business value, then start with a small proof of concept before expanding the project.
Start by talking to the people who do the work every day. Keep notes on recurring frustrations, delays, confusing procedures, and dependencies between teams. The most valuable opportunities are often not flashy projects—they’re the small problems that repeatedly slow people down or make routine work unnecessarily difficult.
Begin with an existing pain point rather than a trending tool. Collect complaints and support requests for a week or two, then identify which issue consumes the most time, creates the most risk, or affects the most people. Only after defining the problem should you evaluate tools or automation. Examples include reducing review bottlenecks, speeding up recurring jobs, replacing manual requests with self-service, improving security, and removing unnecessary dependencies.
Document the baseline before making a change. Write down how long the process takes, how often it fails, how many tickets it generates, or how much alert noise it creates. After the improvement, compare the new numbers with the old ones. That proves the impact even after everyone gets used to the better process and forgets how painful it used to be. Recurring incidents, alert fatigue, toil trackers, and post-incident action items are especially good sources of improvement ideas.
Value-stream mapping is a useful exercise for common workflows such as onboarding, releases, new service creation, and integrations. Look for handoffs, queues, long waiting periods, and repetitive low-value tasks. Improvements usually come from removing unnecessary handoffs, shortening lead times, and making routine work safer and more self-service. Also review the features in tools your company already pays for—sometimes a manual process or add-on can be replaced by an existing capability.
Exactly. Understanding how people actually work, rather than just looking at systems and dashboards, often reveals the biggest opportunities.