The 30-day systems audit

A short, honest look at how the work really gets done, so you know what to build first.

Most owners know something’s wrong but not exactly where. Things slip, the same problems come back, and the business slows down when they step away. An audit turns that feeling into a list: which processes exist, who owns them, where they live, and which ones still depend on you.

It isn’t a consultant’s report or a software review. It’s four weeks of watching the work, writing down what you find, and deciding what to fix. You can run it yourself, with your team, or with outside help.

Someone close enough to see the real work, and honest enough to write down what they find.

• The owner

  Works well whenYou have an hour a day for four weeks and want to see it for yourself.

  Watch out forYou’re too close to it. The things only you do can feel normal.

• An operations lead

  Works well whenSomeone already runs part of the day-to-day and knows the team.

  Watch out forThey need the authority to name gaps, including the owner’s.

• An outside operator

  Works well whenYou don’t have the time, or want fresh eyes and a build plan at the end.

  Watch out forPick someone who builds, not someone who hands you a report.

Whoever runs it, the team should know why it’s happening: to find what’s stuck and fix the process, not to judge people. People show you the real work when they know it’s safe to.

Five things show how the business really runs.

• The processes that stop when you’re unavailable.

  The clearest sign of where the business still runs through you.

• The numbers the team looks at every week.

  Is there a KPI dashboard? Does anyone own each number? Is it reviewed on a fixed day?

• The tools the work actually lives in.

  Not the software you pay for, the places work really happens: inboxes, spreadsheets, sticky notes, someone’s head.

• The owners of each core process.

  One role per process. Shared ownership usually means no ownership.

• The exceptions only you know how to fix.

  The odd customer, the tricky supplier, the workaround that lives in your memory.

Watch, name, write, decide: one step per week.

1. 01

   Week 1: watch the work.

   Follow the core processes from start to finish: a sale, an order, a delivery, an invoice. Write down every place work stops when you aren’t there.

2. 02

   Week 2: name the owners and the tools.

   For each process, write down who owns it and where the work lives. Mark the gaps: no owner, two owners, or no clear place.

3. 03

   Week 3: write the missing processes.

   One page each, using a one-page SOP. Start with the ones that stop without you.

4. 04

   Week 4: decide what to build first.

   Rank what you found by how much time it costs you and how much risk it carries. Pick the first three.

Keep everything to one page per process. If writing the processes feels like the hard part, the free SOP template gives you the format and a filled example.

If you don’t have these five things, the audit isn’t finished.

• A list of what still runs through you.

  Every process that stops when you’re unavailable.

• The owners.

  One role named for each core process.

• The tools.

  Where each process lives, and who has access.

• The first three things to build.

  Ranked, with a reason for each.

• A clear next step.

  Who starts the first build, and by when.

Start with what costs you the most time and carries the most risk.

• Stops when you’re away

  What it tells youThe business depends on you for it.

  PriorityHigh: build this first.

• Breaks every month

  What it tells youThe process is unclear or has no owner.

  PriorityHigh, especially if customers feel it.

• Takes hours of your week

  What it tells youYour time is stuck in the middle of the work.

  PriorityHigh if someone else could do it.

• Works, but nobody wrote it down

  What it tells youIt runs on one person’s memory.

  PriorityMedium: write it before that person leaves.

Auditing how it’s supposed to work, and stopping at the list.

• Auditing the org chart, not the work.

  Follow real orders and real customers.

• Blaming people instead of processes.

  If it breaks, the process isn’t finished.

• Buying software first.

  A tool can’t fix a process nobody has defined.

• Stopping at the list.

  The audit is only worth something if the first build starts the week after.

Pick one process that stops when you’re not available. Write it down and name the owner. That’s the start of the audit. Then move to the next one.

If you’d rather have it done with you, that’s where operations consulting starts: we find what’s stuck, then build it. See also Systems & SOPs.

Scope your project →