"Operational audit" tends to conjure up images of consultants with clipboards, a stack of interviews, and a report that gets skimmed once and filed away. That's not what a useful audit looks like — and it's not what I deliver. A good audit produces a short, prioritised list of things to fix, in an order that makes sense, with enough detail that your team can act on it immediately.
Here's what that process actually looks like in practice.
Step 1 — Map the real workflow, not the documented one
Almost every business has a documented version of how a process works, and an actual version of how it works. They rarely match. The documented version says approvals go through a manager. The actual version is that the manager forwards it to someone else who actually checks it, because that's who has the context.
I sit with the people doing the work — not just the people managing it — and walk through recent, real examples. Not hypotheticals. The last five orders, the last five support tickets, the last five onboarding cases. That's where the truth lives.
Step 2 — Find where time and certainty disappear
Two things matter in every workflow: how long each step takes, and how confident people are that it was done correctly. Bottlenecks show up in both. Sometimes a step is slow because of genuine complexity. More often, it's slow because of waiting — waiting for approval, waiting for someone to check email, waiting for a system that doesn't talk to another system.
I'm also looking for steps where people manually double-check something because they don't trust the data. That's a different kind of cost — not time lost, but confidence lost, and confidence problems compound in ways that are hard to see until they cause a real mistake.
Step 3 — Separate "fix the process" from "automate the process"
This is the step most audits skip, and it's the one that matters most. Automating a broken process just makes the broken process faster. Some things genuinely need automation. Other things need a clearer decision-making rule, a single source of truth, or just one person taking ownership of a step that currently has none.
Automation is a multiplier. If the underlying process is wrong, automation multiplies the wrong thing faster.
Step 4 — Build a roadmap, not a wishlist
The output isn't a list of everything that could theoretically be improved. It's a short, sequenced plan: what to fix first, what depends on what, and what the expected impact is for each item. Usually 4–8 items, ranked by effort versus impact, not 40 items with no sense of priority.
What makes an audit actually useful
The test of a good audit isn't how thorough the document looks. It's whether someone on your team can read the top three items, understand exactly what to do, and start in week one without needing a translator. If a report needs another consultant to interpret it, the audit didn't do its job.
If you've never had a structured look at where your operational time actually goes, that's usually the highest-leverage place to start — before any tool, any automation, any new system. Know what's actually broken first. Everything after that gets easier.