The Vacation Test: How to Build a Business That Runs Without You (A 12-Month Playbook)

Most founders fail the vacation test and then explain it away. "The team needs more experience." "We're in a critical period right now." "I like being involved."

The explanations aren't wrong. They're just not the full story.

The full story is that every significant decision in your company routes through your head. When your head is on a plane to Greece, those decisions queue up. By day three of the holiday, people are pinging you "just a quick question." By day five, you've given up and you're on Slack from a café. By day ten, you're booking the flight home early.

This is not a time-management problem. It's a systems problem. And systems problems take systems to fix — not willpower, not better delegation advice, not another productivity book.

The playbook below is what actually works. Not a framework. A sequence. Each month builds on the one before it. Skip the early months and the later ones collapse.

The honest timeline

Twelve months sounds long. It isn't. Here's what actually takes time:

• Installing decision rules takes weeks; getting people to use them takes months.
• Writing metrics is a weekend; getting them measured honestly takes a quarter.
• Delegating a function is a conversation; trusting the outcome takes a year.

The playbook accounts for this. The first three months look like you're doing nothing. You're not. You're laying foundation no one sees. By month six, things visibly change. By month twelve, the vacation test is passable. By month fifteen, it's genuinely easy.

Month 1: Audit what's in your head

Do nothing else in month 1 except this: track every decision that came to you in one week. Every Slack DM, every "got a sec?", every 9pm text, every escalation on a customer call.

Categorise them afterwards into four buckets:

1. Decisions only you could make (strategic, existential, legal, new territory)
2. Decisions someone else should make but the rule wasn't written down
3. Decisions someone else should make but they didn't have the data
4. Decisions that shouldn't have been made at all — a process would have handled it

Most founders discover that bucket 1 is 10–15% of inbound. The other 85–90% is fixable. That ratio is your starting point. The goal for month 12 is to flip it: 85% of decisions handled by the system, 15% by you.

Months 2–3: Install decision rights and weekly metrics

Two artefacts, nothing else.

Decision rights matrix. One page. Columns: decision type, budget threshold, who decides, who must be consulted, who must be informed. Start with the ten highest-frequency decisions from your month 1 audit. Share it, discuss it, revise it. Add more as you go.

Weekly metrics scorecard. Five to seven numbers your leadership team reviews every week. Each number has an owner, a target, and a reason it matters. Resist the urge to put twenty on it. Five to seven is the sweet spot — enough to see the business, few enough to read in ninety seconds.

The point of these two artefacts isn't the artefacts. It's the conversations they force. You'll discover that three people thought they owned pricing decisions (and none of them really did). You'll discover no one tracks your most important metric because nobody was told to. Those discoveries are the value.

Months 4–6: Delegate the first three functions

Pick the three functions that consume the most of your time and where the decision quality doesn't require you. For most SMEs that's: customer operations, internal operations, and some slice of finance.

For each function, do the same thing:

1. Name the person who owns it (may already exist; may need to be hired).
2. Write the one-page charter — scope, authority, metrics, boundaries.
3. Run a 30-day handover period where they shadow you and you shadow them.
4. After day 30, stop being copied on routine decisions. This step is the hardest.

If you're still being copied on every invoice approval at month 9, the delegation didn't happen. Pull yourself off the CC list. If the world doesn't end, you're fine. If it does end, the right answer isn't for you to come back — it's to fix the system that made the world end.

Months 7–9: Document the ten processes you care about most

Not forty processes. Not a full wiki. Ten.

For each one, a one-page document that answers:

• When does this process start? (trigger)
• What are the steps?
• Who is involved at each step?
• What's the output?
• What goes wrong, and what to do when it does?

Most SMEs over-engineer this phase. They hire a consultant, buy software, start a Notion workspace, then lose momentum halfway through. One-page docs in a shared folder are enough. You can formalise later if needed.

The ten processes to document first: customer onboarding, customer support escalation, sales handoff, hiring, onboarding a new employee, monthly financial close, weekly leadership meeting, supplier payment, incident response, performance review. These are the ones that hit quality hardest when they're tribal. We go deeper on this in Document Your First 20 Processes.

Months 10–12: Install the leadership layer

The last step is the one most founders get wrong. They try to install a leadership layer in month 1 — promoting people, hiring heads of — before the systems exist to support them.

Do it last. By month 10, you've got decision rules, metrics, three delegated functions, ten documented processes. Now the leadership roles have something to plug into.

You need either:

• Two to four direct reports who own clearly defined areas and meet weekly with a structured agenda, OR
• One senior operator (a real COO, not a glorified project manager) who runs the machine while you focus on the one or two things only you can do.

The second option is faster but costlier. The first is cheaper but slower. See COO vs Fractional COO vs Consultant for how to decide.

The vacation test, retaken

At month 13, run the vacation test for real. Book ten consecutive working days off. Don't say you're available for emergencies. Don't check Slack. Turn off email notifications. Let it run.

If something breaks in those ten days, note it — but don't fix it while you're away. Let the system handle it as best it can. You'll return to a specific, documented list of what failed. That list is more valuable than anything you'd have accomplished if you'd stayed working through the holiday.

Fix the failures. Run the test again six months later. Over time, the list shrinks to nothing.

What this is NOT

This playbook is not: a path to passive income, an exit plan, an excuse to stop caring about your company, a replacement for strong founder judgement on the decisions that still matter.

The vacation test isn't about being absent. It's about the business being robust enough that your absence doesn't break it. That's a fundamentally different thing — it's what makes you able to focus on the 10–15% of decisions that actually need you, without being buried in the 85% that don't.

What to do next

Three options:

1. Run the month-1 audit this week. Track decisions for seven days, categorise them. No other commitment. The exercise alone will show you where to start.
2. Hire a fractional COO to compress the timeline. A good operator can do months 2–9 in about four months. Costs €6K–€10K/month. Makes sense if you have the budget and the urgency.
3. Run a Business Pulse diagnostic. We'll tell you honestly whether your problem is actually structural (this playbook) or something upstream — product-market fit, positioning, financial model — that no operating system will fix.

Whatever you choose, start month 1 this week. The hardest part of this playbook isn't any individual month. It's resisting the urge to skip ahead.

Frequently asked questions