18 March 2026 · Management reporting
Why a two-week pack beats a perfect month
Most growing companies already have the numbers. They have a ledger, a bank feed, a billing export and a media sheet. What they do not have is a pack the room will open before the next decision. The work stretches because every stakeholder wants one more reconciliation, one more cut, one more week “so it is clean”. The month that arrives in week three is accurate in a narrow sense and useless in the only sense that matters: it can no longer change a hire, a campaign or a purchase order.
A two-week pack is not a lower standard. It is a different standard. You lock the dictionary on days 3–5. You name the gaps in the movement note instead of hiding them in a plug. You ship on day 14 with every material figure recoverable from a source file. If a line cannot be recovered, it stays out. That rule feels severe in week one. It is the reason the board still has a month left to use.
The usual objection is that the data are not ready. They rarely are. Readiness is what you get after you have agreed cut-off rules and stopped reopening them for convenience. A delayed pack does not create readiness. It creates a second, quieter close that happens in Slack the night before the meeting, with no dictionary and no date.
Ship the board pack in two weeks. Then decide, in writing, whether the next month is a retainer or a stop. Do not keep a “nearly ready” file in draft for a quarter. A draft that never releases is just another workbook on the pile.
If you want the sprint mechanics, they are on How It Works. If you want a planning range in S$, see Engagements.
9 May 2026 · Cash
Build a 13-week cash view operators will open
Annual cash forecasts die in the drawer because they are written for a budget ritual. Operators need a view they can open on a Tuesday when a supplier asks for an earlier payment or a customer slips a milestone. Thirteen weeks is long enough to see a payroll, a tax date and a slow cohort. It is short enough that the assumptions can still be evidenced.
See cash 13 weeks out — but only if the receipts follow real terms. A model that books next month’s collections at last month’s revenue is a hope file. Start from open invoices and contracted payout lags. Marketplace channels often pay on a delay the P&L never shows. Annual SaaS prepay looks like a feast in week one and a gap in week twenty; the 13-week view has to treat that as timing, not as run-rate.
Keep three paths in one file: base, slower, stretch. The stretch path is where unevidenced receipts live. If a salesperson is “confident” about a transfer that is not in the bank calendar, it does not enter the base case. The pack should say so in a sentence, not in a colour-coded dashboard that implies precision.
Update without rebuilding. The horizon walks forward each week. A model that needs a weekend to refresh will not be opened mid-week, which means the operating team will keep a private sheet — and you will be back to two versions of cash.
Pair the cash view with the management pack, not as a rival artefact. Profit can look fine while cash is tight. The room needs both on the same release day, with the same dictionary, or the conversation splits into “finance cash” and “ops cash”.
2 July 2026 · Unit economics
Unit economics that survive a channel mix change
Blended contribution is how a growing company talks itself into a channel that does not pay back. Own-site, marketplaces and paid social do not share a cost structure. When the mix shifts — and it will, the first time a platform changes a take-rate or an ad auction — a single “unit margin” number becomes a souvenir.
Build the model at the grain you will actually manage. For commerce that is often channel and, if the catalogue is wide, a small set of categories, not every SKU on day one. For subscriptions it is plan and cohort, not a lifetime value that assumes the current price holds for three years. For project firms it is project type and utilisation, not a blended day-rate that hides unpaid scope.
Cost to win a customer must include the costs that actually move with the win: media, take-rates, payment fees, variable fulfilment, success commissions. Office rent does not belong in the unit until you are making a capacity decision. Putting every cost into the unit is how the model becomes a second P&L that nobody trusts.
When a channel mix changes, the model should move in the same week. That is a design choice. If the refresh needs a rebuild, the number will be late and the campaign will already have run. Lock the definitions. Wire the feeds. Leave a clearly marked “not yet” where the extract cannot carry the claim — for example, a new marketplace with two weeks of data.
Illustrative scenarios for how this looks in a month sit on Playbooks. They are worked examples, not results we claim from a named client.
14 September 2026 · Close
Close the month while it still matters
A close that lands after the next campaign has started is a filing exercise. The operating team has already spent the money. The board meeting becomes a review of history. If you want the month to still belong to the people who can change it, put a date on the pack and defend that date.
After handover we aim for a working-week close: collect on a named morning, draft inside the week, release on a published day. Five working days is typical when the dictionary is stable and the extracts arrive. It is not a promise that every group will hit five days in month one. It is the pace the calendar is built around.
Late feeds are declared, not waited for. Moving the release to “next Tuesday, when inventory lands” is how week three happens. A pack that says “inventory not in; cash and contribution exclude that cut” is more useful than a silent delay. The room can still decide. The missing cut can join the next cycle.
Keep the review short. The meeting is for movement in defined measures, not for redesigning tabs. New questions are welcome; they become a defined cut or they wait. A pack that grows a new sheet every month is a pack that nobody can learn.
None of this is investment, tax or legal advice. It is the operating craft of getting a usable month onto the table while the month can still be used. If that is the gap in your calendar, send a brief.
