A predictable week is a debugging tool. When something feels off, you can usually spot which ritual broke first. The standup got cancelled twice. The retro became a status update. The Friday close-out became silence. Each pattern is a signal about where the team has stopped repairing itself, before it is a signal about where the work is failing.

The cadence below is what I run with most pre-seed teams. Hold it for a quarter unchanged. Then either deepen the rituals or cut the ones that did not earn their slot.

Mondays: open with intent

0900: 15-minute team standup. Three questions: what is the one thing we are committing to ship this week, what is the one decision we owe the team, what is blocking us. No status updates, no demos.
0915: Founder solo block, 90 minutes. The week is planned now, in the calendar, before slack reopens. If this slot is missed, the week is reactive by lunchtime.
1100: First customer call of the week. Always Monday morning. Sets the tone before the inside work starts.

Tuesday to Thursday: deep work + one ritual per day

Tuesday afternoon: pairing or design review block. 90 minutes. Whoever is shipping the hardest thing this week leads it.
Wednesday afternoon: one external conversation. Investor update prep, advisor call, customer interview, partnership conversation. Pick one type per quarter and rotate.
Thursday morning: writing block. Founder writes, in long form, the decision-of-the-week or the post-of-the-month. If you cannot remember what to write, the week was reactive.

Fridays: close with an artefact

1500: Team async close-out, written, shared in the same channel each week. Three lines per person: what shipped, what slipped, what I learned. No emoji-react acknowledgement is fine; the artefact itself is the point.
1600: Founder reads the close-outs and writes the weekly note. Public to the team, and the basis for the monthly investor update on the first Monday of the next month.
1700: Calendar audit. Five minutes. Anything that landed without a clear purpose this week gets cut from next week, before it becomes recurring noise.

The monthly tether

On the first Monday of every month, the cadence pauses for a 60-minute monthly review. Three questions, twenty minutes each.

What did we commit to last month and what did we ship?
What number from the runway planner do we put in the investor update subject line?
What one ritual, hire, or decision does next month need that this month was missing?

The quarterly tether

At the end of each quarter, two artefacts. A team-wide retro (90 minutes, off-site if you can) and the schlep inventory (separate worksheet, run independently). Together they reset the cadence for the next quarter — what to keep, what to retire, what to deepen.

How to read a broken cadence

Mondays cancelled twice in a row → the founder has lost the week before it started. Recover by re-blocking Monday morning before any other meeting moves.
Friday close-outs going silent → the team has run out of things they feel safe naming as slipped. Add a one-on-one before the next close-out.
Monthly review skipped → either the cadence is not earning its slot or the founder is hiding from a number. Both diagnoses are urgent.
Quarterly retro becomes a status meeting → kill it for one quarter. Bring it back only when there is a question worth ninety minutes off-site.

A startup's cadence is the cheapest dashboard it owns. The team's relationship to its rituals tells you more about the company's state than the metrics do, and it tells you a fortnight earlier.

Template from Marga Haus · margahaus.com/resources · Adapt and use freely. Attribution appreciated, not required.

Weekly cadence template