7 min read · 1,369 words
At a glance
- By the second Tuesday of the month, the books in most funded tech companies are roughly closed. What gets produced in the next 48 hours determines whether finance defends the next call or explains the last one.
- The deliverable should be three artifacts on three pages: the forward capital register updated, the next call named with a three-path model attached, and the runway truth under each path.
- If the second Tuesday produces those three, the variance memo writes itself. If the variance memo comes first, the close is doing the wrong work.
Three artifacts, each one page, due the second Tuesday of every month: the forward capital register, updated for the month that just closed; the next call, named, with a three-path model attached; and the runway truth under each of the three paths. That is what the second Tuesday of the close should produce. Not a variance memo. The memo comes later, and by then it mostly writes itself.
What does the standard close calendar look like?
For most venture-backed tech companies between $1M and $8M ARR, the close runs across the first ten business days of the month. Days 1 to 3 are mechanical: bank reconciliation, AR and AP cutoffs, payroll tie-out, the credit card and expense feeds. Days 4 to 7 are judgment work: revenue recognition, accruals, the deferred revenue roll-forward, the handful of entries that require someone to actually decide something. Days 8 to 10 are review: the controller ties the statements, the senior finance person reads them, and somebody opens a doc titled “variance commentary.”
The second Tuesday of the month lands on business day 7 or 8 in most months. By then the picture is roughly set. Revenue is known within a point or two. Burn is known almost exactly. Whatever the month did to the company has already happened and is now visible in the numbers.
This is the most valuable window in the finance calendar. The numbers are fresh, the month’s context is fresh, and nothing has been committed to a narrative yet. What the team chooses to produce in the next two days determines whether the close was worth running.
How did the variance memo become the default deliverable?
Because the close grew up inside accounting, and accounting answers a backward question: what happened, and is it stated correctly? The variance memo is the natural extension of that question. Plan said one number, actuals said another, here are four bullets on why.
Boards reinforced the habit. Investors ask what happened to sales and marketing spend this month, and a memo answers that cleanly. Auditors reinforced it too; commentary that explains movements is exactly what a reviewer wants in the file. So the memo became the deliverable, then the habit, then the unstated definition of a finished close.
The problem is not that variance commentary is wrong. It is that variance commentary is retrospective by construction. Most variances worth explaining trace back to decisions made under a context that has since moved: a channel that converted at one cost and now converts at another, a pricing premise from the last raise, a hiring plan built against a pipeline that changed shape. The decisions weren’t wrong. They were right for a context that no longer exists. A memo can describe that drift. It cannot act on it.
What should the second Tuesday produce instead?
Three deliverables, in this order.
- The forward capital register, updated. The register is the live list of the company’s material deployed dollars, each scored against current context. The close is the monthly trigger to re-score it: which dollars are now deployed against a context that died this month, and which are still defendable. A vendor renewal priced against last year’s usage, a channel whose unit economics moved, a headcount plan built for a pipeline that changed. The register update is the close translated into capital terms.
- The next call, named. The biggest unresolved decision the company has to make this month, written as one sentence, with the three-path model attached: Keep, Kill, Restructure, with the dollars under each path, a name defending each number, and a decision date. The math, the names, the dates. One decision, not eight. The others get scheduled.
- The runway truth, under each of the three paths. Not the runway slide. The runway slide is one number on one chart, and it assumes the company makes no decisions at all. The runway truth shows what cash does under each path of the named call: out-of-cash date under Keep, under Kill, under Restructure. The decision and the runway are the same conversation, rendered on one page.
If the second Tuesday produces those three things, the variance memo writes itself, because the commentary is no longer trying to carry the story. It cites the register, points at the named call, and moves on. If the second Tuesday produces a memo and nothing else, the close did compliance work and called it finance.
What does each artifact look like on one page?
The register page is a table, not a memo. One row per material deployed dollar, with the monthly amount, the context score (defendable, dead, or under review), the next move, the defender, and the re-decide date. The rows that changed this month are flagged. A board member should be able to read it in 90 seconds and know exactly which dollars the company is still willing to defend.
The next-call page is the Decision Slide: the decision in one sentence, the three paths in a phrase each, the numbers under each path in the same units, the finance function’s recommendation in one sentence, and the date and owner. If it runs past those five lines, it is not ready; everything beyond them is narrative that softens the recommendation.
The runway-truth page is three rows of monthly cash, one per path, with the out-of-cash date marked on each row. It is the Runway Spine re-cut against the live decision instead of against generic base, downside, and upside cases. Same discipline, sharper question.
How does the Financial Rhythm System schedule this?
The Financial Rhythm System is the monthly cadence that makes the second Tuesday repeatable instead of heroic. Its three instruments (the Decision Slide, the Runway Spine, the Signal Dashboard) are produced against the close, every month, in the same order; the full cadence is laid out in the Financial Rhythm System overview.
The second Tuesday deliverables are how the rhythm consumes the close. The register update is the raw input. The named call becomes the month’s Decision Slide. The runway truth is the Spine, re-cut. The FRS then puts one standing meeting on the calendar within 48 hours of the second Tuesday, where the three pages get read and the named call either resolves or gets a date. The variance memo is written after that meeting, not before it, and it quotes the three artifacts rather than originating the story.
That ordering is the whole trick. Close first, decide second, explain third. Most stalled portcos run it backward: explain first, and the deciding never quite gets a slot on the calendar.
What is the Decisive Finance role in this?
The 14-day Decision Diagnostic builds the first version of all three artifacts: the initial forward capital register, the first named call with its three-path model, and the runway truth under each path. The second month of an engagement installs the cadence so the company produces them itself, every second Tuesday, without us in the room. Guaranteed 3x your Diagnostic fee in recoverable value, in 14 days. Typical 5x to 10x. We don’t audit the past. We justify each next call.
Where to go from here
If your last close produced a variance memo and nothing else, the next second Tuesday is less than four weeks away, and three pages is a reachable target.
Read the guarantee Book the call
Related reading: