How should end-of-quarter math handle deals that straddle close window boundaries?

The Two Numbers a Straddle Deal Touches

Every straddle argument I have ever refereed came down to two people using the word "revenue" to mean two different numbers. The controller means GAAP revenue on the income statement. The VP of Sales means bookings against quota. When a $120,000 annual contract signs on the last day of the quarter, the sales org wants to celebrate $120K of "closed-won," while finance may recognize $10K that quarter (one month of ratable service) and defer $110K to the balance sheet as deferred revenue. Both are correct. Neither is lying. They are answering different questions.

Write these definitions down and post them where the sales team and the finance team can both see them:

• Bookings (a.k.a. closed-won, ACV, TCV): the contractually committed value at signature. This is a *sales* number. It drives quota attainment, commission, and pipeline coverage. Close date owns it.
• Recognized revenue: the portion of the contract earned in the period per ASC 606. This is a *finance* number. It hits the P&L. The delivery pattern owns it, not the close date.
• Deferred revenue / contract liability: cash or invoiced amounts not yet earned, sitting on the balance sheet until performance obligations are satisfied.
• Billings / invoiced amount: what you actually invoiced this period. This can lead or lag both bookings and recognized revenue depending on payment terms.

Once you name these four numbers separately, the straddle question stops being "which quarter does the deal go in?" and becomes four cleaner questions: which quarter gets the booking, how much revenue is recognized in each period, how much sits in deferred, and when do we invoice. A deal can — and routinely does — split differently across all four. A $240K two-year prepaid deal signed on the final day of Q2 books $240K of TCV in Q2 bookings, recognizes roughly $10K in Q2 revenue, parks ~$230K in deferred, and bills $240K on day one if terms are prepaid. Four numbers, four answers, zero contradiction. The teams that fight about straddle deals are almost always the teams that never wrote down which of the four they are arguing over.

Set the Close-Window Boundary Rules Before the Quarter Starts

Ambiguity is expensive, and it is *always* resolved in the direction of optimism if you let humans resolve it in the final week. The only durable fix is to define the boundary rules at the start of the fiscal year, in writing, and make them mechanical. Pavilion, RevOps Co-op, and most sales-operations communities converge on the same core discipline: a deal moves buckets based on documented, verifiable state — not on how the rep feels on a Thursday night.

Here is a concrete, defensible ruleset you can adapt. The specific thresholds are illustrative; calibrate them to your own historical data, but keep the *structure* rigid.

Commit window (counts toward the number you tell the board): a deal qualifies only if all of the following are true —

1. Documented close date on or before the quarter's final calendar day.
2. Stage is at Negotiation/Commitment or later — not Proposal, not Discovery.
3. Buyer has signed a deal authorization, LOI, order form, or equivalent procurement sign-off; verbal-only does not qualify.
4. No open legal, security, procurement, or finance hold.
5. The AE certifies close risk below ~20% with a specific reason, not a vibe.
6. The opportunity was created in the CRM at least ~30 days before the close date (this filters out last-minute "pipeline plumbing" that inflates the number).

Best-case window (upside, not committed): close date within roughly 3 business days after quarter-end, stage at Proposal or higher, a concrete next meeting or countersignature pending, outcome expected within 48 hours.

Pipeline (everything else): any deal without a documented next step. No "feels close" deals allowed in either of the two windows above.

The single most important line: the quarter ends on a calendar date, and no CRM edit can move it. The classic credibility-killer is a rep dragging a close date from April 2 back to March 31 in the last hour. Lock close-date edits within the final 48 hours (require manager approval with a reason code), snapshot the pipeline daily so you can see exactly who moved what, and audit any close-date change inside the final week. Once these immutable rules exist, forecast variance visibly tightens — teams routinely report meaningfully lower miss rates simply because the borderline deals stop flipping in and out three times a week. The number stops being a negotiation.

A board-ready sentence falls out of this for free: *"Our commit forecast includes only deals the owning AE has certified as closing on or before the final day, with buyer sign-off in hand and no open holds."* That sentence is worth more than any dashboard because it tells leadership exactly what the number does and does not contain.

The Three-Bucket Framework for Straddle Deals

The most common mistake in handling straddle deals is treating them as binary — either fully in the current quarter or fully in the next. Binary treatment manufactures artificial volatility and rewards gaming. A three-bucket framework acknowledges that deals exist on a probability continuum and forces an honest, weighted call.

Bucket 1 — Locked (roughly 70–100% probability). The customer has signed, the PO or order form is in hand, and only delivery or onboarding remains. For SaaS this means an executed contract before the customer is provisioned; for services it means a signed SOW before kickoff. These book at 100% of value in the quarter they execute, because the *sales* event (the commitment) is complete — independent of when *revenue* is recognized.

Bucket 2 — Committed (roughly 40–69% probability). Verbal commitment exists, legal has reviewed terms, but the signature has not landed. These are the true straddlers. Assign a weighted value equal to your historical close rate for deals *at that specific stage in your cycle*. If deals that reach legal review historically close 55% of the time, put 55% of the deal value in the current-quarter forecast and roll the remaining 45% to next quarter. Do not use a gut number — pull the conversion rate from your CRM's stage-history report.

Bucket 3 — Pipeline (roughly 10–39% probability). Still in active negotiation with no committed close date. Never split these. They belong entirely to the quarter where your CRM's historical stage-to-close timing says the signature is most likely to land.

The non-negotiable rule: no deal sits in more than one bucket at a time. This kills double-counting and forces a single honest assessment. Review bucket assignments weekly through the last four weeks of the quarter, adjusting weights based on *actual movement*, not renewed optimism.

Operationally, build a "straddle tracker" — a CRM report or a Friday-afternoon spreadsheet — that auto-computes weighted value from stage, deal size, and the rep's own historical accuracy. When the quarter closes, lay your weighted forecast next to actuals. That variance *is* your calibration data for next quarter's weights. Over three or four quarters this converges: your Committed-bucket weight stops being a guess and becomes an empirically justified number you can defend line by line in a QBR.

A worked example. You enter the final four weeks with these straddlers: a $90K deal at signature-pending (Locked, weight 100% → $90K), two deals totaling $200K in legal review (Committed, historical 55% → $110K forecast, $90K rolled), and a $150K deal mid-negotiation with no close date (Pipeline, lands next quarter → $0 this quarter). Your straddle contribution to commit is $90K + $110K = $200K, with $90K explicitly forecast as slipping and $150K explicitly excluded. Nothing is hidden, nothing is double-counted, and every figure traces to a documented rule.

The 48-Hour Hard Stop and Soft-Landing Protocol

The most dangerous window for straddle math is the final 48 hours. Pressure peaks, judgment degrades, and deals that should slip get crammed into the wrong bucket. A structured protocol removes the discretion at exactly the moment discretion fails.

Hard stop — 48 hours before quarter-end. No new deal enters Locked or Committed without a *signed document*. Verbal commitments, "we're basically there" emails, and handshake deals auto-move to next quarter's pipeline. This is enforced without exception, announced at the start of the quarter, and reminded at the two-week mark. The reason it works is that it is mechanical: the rule, not the manager, says no, so no relationship is damaged and no argument is winnable.

Soft landing — the first 10 business days of the new quarter. Create a "carryover window" where deals that were in the Committed bucket at the hard stop can finish with minimal friction:

• No new discovery or demos — only completion of in-flight negotiations.
• Legal and procurement resources are pre-blocked for these specific deals.
• Commission credit is auto-split in favor of the prior quarter (see the next section).
• The CRM tags these as "carryover closes" so reporting stays clean.

This prevents the quarter-end hangover where reps spend two or three weeks of the new quarter closing what should have closed in the old one — and it hands you clean diagnostic data. A deal that was "Committed" at the hard stop but does *not* close inside the soft-landing window was never truly committed, which tells you your bucket criteria are too loose.

Measure the straddle conversion rate: the percentage of Committed-bucket deals at the hard stop that actually close within the soft-landing window. A healthy band is roughly 60–80%. Below ~50% means your Committed criteria are too permissive — tighten them. Above ~90% means you are being too conservative and leaving forecastable revenue in Pipeline where the board never sees it.

Escalation rule: if a rep carries more than ~30% of their quarterly quota into the carryover window two quarters running, that is a pipeline-management problem, not a straddle-math problem. Require a coaching plan focused on earlier qualification and deal acceleration so the straddle framework never becomes a crutch for chronically late-stage pipelines.

Commission Splitting Without the Cliff

The all-or-nothing commission cliff at 11:59 PM on the last day of the quarter is the single biggest behavioral distortion in straddle deals. It drives frantic, margin-destroying discounts to drag a deal one day earlier, and it drives reps to *hold* deals so they land in a quarter where they are behind quota. Both behaviors are bad for the company and both are created entirely by the comp design, not by the reps.

Decouple compensation from the accounting entirely and design the split to reward the *right* behavior. A graduated model works well:

• Deal fully negotiated in the prior quarter but signed in the first ~5 business days of the new quarter: 70% credit to the prior quarter, 30% to the current.
• Deal signed in the last ~5 business days of the current quarter but initiated earlier: 50/50 split — this specifically removes the incentive to burn margin closing one day sooner.
• Deal that closes cleanly in the middle of either quarter: full credit to the quarter of close.

A defensible formula for the split basis is (days of the negotiation that fell in Quarter A / total negotiation days) × deal value, with a minimum 30% floor for whichever quarter did the heavy lifting so a deal that dragged is not stripped of all credit from the quarter that earned it.

Worked example: a $100K deal spends 45 days in negotiation across Q1 and signs on day 3 of Q2. Under a 70/30 carryover split the rep books $70K of quota credit in Q1 and $30K in Q2. The manager's forecast shows $70K as expected carryover in Q1 and $30K as an early close in Q2 — no double-counting, no argument, and critically no incentive to have discounted the deal to force a March 31 signature.

Two implementation rules make this stick. First, publish the split in the comp plan at the *start* of the fiscal year; changing splits mid-quarter destroys trust faster than almost anything else you can do. Second, keep it mechanical in the CRM/comp tool so the split computes automatically from the negotiation-start and close timestamps — the moment a human has to adjudicate a split by hand, the cliff behaviors come back.

What Finance Actually Requires: ASC 606 and Period Cut-Off

Sales-ops leaders get into trouble at audit time when their "straddle rules" quietly contradict what the controller is required to do. You do not need to be an accountant, but you need to know the four moving parts so your forecast and your books reconcile.

The five-step model (ASC 606 / IFRS 15). Revenue is recognized by (1) identifying the contract, (2) identifying the performance obligations, (3) determining the transaction price, (4) allocating that price to the obligations, and (5) recognizing revenue as each obligation is satisfied. Step 5 is the straddle crux: recognition follows *satisfaction of the obligation*, which is either at a point in time (control transfers on a date) or over time (control transfers continuously — most subscriptions and many services).

Point-in-time vs. over-time. A perpetual software license or a shipped physical good typically recognizes at the point control transfers — often the close/delivery date, which *does* land in one quarter. A SaaS subscription or a retainer recognizes *over time*, ratably, so the close date barely touches the current period. A fixed-fee services engagement that straddles the boundary recognizes using an input measure (cost or hours incurred to date over total estimated) or an output measure (milestones accepted), and only the earned slice hits the current period.

Cut-off is the audit risk. "Cut-off" is the accounting term for making sure a transaction is recorded in the correct period. Recording revenue a few days early to make a quarter — pulling a January-delivery deal into December — is precisely the practice the SEC's Staff Accounting Bulletins (notably SAB 101/104 on revenue recognition) and countless enforcement actions target. Your sales-side straddle rules must never pressure finance to book revenue before the obligation is satisfied. This is exactly why the *booking* rule (close date owns it) and the *recognition* rule (delivery owns it) must stay separate: the sales org can celebrate a Q4 booking while finance correctly recognizes most of it in Q1, Q2, and beyond, and the audit stays clean.

Practical reconciliation. Once a quarter, run a reconciliation that ties bookings → billings → recognized revenue → deferred balance and confirms the movements agree. When they diverge, the difference should be fully explained by timing (deferred revenue that has not yet been earned), never by a deal that was double-counted or booked in the wrong period. That reconciliation is your early-warning system: if sales bookings and finance recognition drift apart in a way *timing* cannot explain, a straddle deal has been mishandled and you want to find it before the auditor does.

Putting It Together: A Quarter-End Operating Checklist

Assemble the pieces into a repeatable close routine so nothing depends on heroics in the final week:

1. Start of fiscal year: publish the bucket definitions, the boundary rules, and the commission-split schedule in writing. Lock them.
2. Start of each quarter: re-confirm the historical conversion weights for each stage from last quarter's actuals-vs-forecast reconciliation.
3. Final four weeks: review the straddle tracker every Friday; move deals between buckets only on documented state changes; snapshot pipeline daily.
4. Two-week mark: remind the team of the 48-hour hard stop and the close-date-edit lock.
5. 48 hours out: hard stop engages. No unsigned deal enters Commit. Close-date edits require manager approval with a reason code.
6. Quarter-end: book only executed contracts; roll unsigned Committed deals into the soft-landing window with the carryover tag and the 70/30 split.
7. First 10 business days of new quarter: work the carryover window; no new discovery, only completions.
8. Post-close: reconcile bookings → billings → recognized → deferred; compute the straddle conversion rate; feed the variance back into next quarter's weights.

Run this loop for two or three quarters and the straddle problem largely dissolves. The rules do the arbitrating, the weights become empirical rather than emotional, and the two numbers — bookings and recognized revenue — reconcile cleanly with the difference sitting exactly where it should: in deferred revenue, fully explained by timing.

FAQ

What exactly counts as a "straddle" deal in end-of-quarter math?

A straddle deal is any opportunity whose commercial activity crosses a quarter boundary — most commonly one where the expected close date sits inside the quarter's final booking window (often the last 7–14 days) while either the signature or the delivery slips past the boundary. It requires special handling because ordinary pipeline math tends to either count it fully (inflating the current quarter) or drop it entirely (understating momentum), and because its *booking* quarter and its *revenue-recognition* period frequently differ.

Should I split a deal's value across quarters if it closes in the window but delivers later?

Separate the two questions first. For sales attribution, most teams assign 100% of the booking to the quarter the contract is executed, even when delivery is later, because quota and commission track the close. For revenue recognition, you follow ASC 606: if it is a ratable subscription or an over-time services engagement, you recognize only the earned portion in each period and defer the rest — that is a genuine split on the P&L. So the *booking* usually is not split, but the *recognized revenue* very often is. Don't force the two to match.

How do I keep straddle deals from inflating my end-of-quarter forecast?

Flag any deal whose close date sits in the final window as a "straddle" in the CRM, then forecast it with a weighted probability drawn from your historical stage-to-close conversion — commonly landing in the 40–70% range for legal-review-stage deals — instead of full value. Track straddlers as a distinct category, enforce the 48-hour hard stop so unsigned deals cannot enter commit, and reconcile weighted forecast to actuals each quarter to keep the weights honest. That preserves legitimate late-quarter momentum without letting optimism inflate the commit number.

What's the best way to present straddle deals to leadership in a QBR?

Show them as their own category, never blended into clean commit. Use a simple table — deal name, expected close date, total value, bucket, the percentage you are counting this quarter, and the amount rolled to next quarter. Pair it with your straddle conversion rate (what share of last quarter's Committed straddlers actually closed on time). Transparency here builds forecast credibility: leadership stops being surprised, because you have already told them what could slip and by how much.

Can I use historical data to predict how many straddle deals will close?

Yes, as a calibration guide rather than a guarantee. Pull the last four to six quarters and compute the conversion rate for deals at each stage in the final window — you will typically see a wide band (often 10–30% for early-stage deals in the window, higher for signature-pending ones) that varies with deal size and complexity. Use that band to sanity-check the current quarter and to set your bucket weights, but never treat it as a hard prediction; each quarter's deal mix is different, which is exactly why you reconcile and re-weight every quarter.

What if my sales team pushes to count a straddle deal as fully closed when it isn't?

Hold the line by tying booking credit to actual contract execution — signed and countersigned — not verbal commitment, and by making the rule mechanical rather than a manager's judgment call. Combine that with the graduated commission split so the rep is not facing an all-or-nothing cliff; the 70/30 carryover credit removes most of the incentive to misrepresent status, because the rep still gets the majority of the credit for a deal that slips a few days. When the rule says no and the comp plan is fair, the gaming pressure largely evaporates.

Sources

• Financial Accounting Standards Board (FASB) — Accounting Standards Codification Topic 606, *Revenue from Contracts with Customers*: https://www.fasb.org/page/PageContent?pageId=/standards/accounting-standards-codification.html
• IFRS Foundation — IFRS 15, *Revenue from Contracts with Customers* (over-time vs. point-in-time recognition and allocation): https://www.ifrs.org/issued-standards/list-of-standards/ifrs-15-revenue-from-contracts-with-customers/
• U.S. Securities and Exchange Commission — Staff Accounting Bulletins on revenue recognition and period cut-off (SAB 101/104): https://www.sec.gov/interps/account/sabcodet13.htm
• American Institute of CPAs (AICPA) — Revenue recognition resources and industry implementation guidance: https://www.aicpa-cima.com/topic/audit-assurance/audit-and-assurance-greater-than-revenue-recognition
• Harvard Business Review — analysis of sales forecasting discipline and earnings-timing pressure: https://hbr.org/2015/07/how-to-really-motivate-salespeople
• Investopedia — reference explanations of deferred revenue, bookings vs. revenue, and quarter-end accounting cut-off: https://www.investopedia.com/terms/d/deferredrevenue.asp

Related on PULSE

• [How should forecast models handle multi-year deals that straddle revenue recognition boundaries?](/knowledge/q306)
• [How should a 2027 sales org draw boundaries between deal desk and RevOps?](/knowledge/q12607)
• [How do you coach reps to handle end-of-quarter procurement pressure?](/knowledge/q13926)
• [How do you operate sales during a system outage during end-of-quarter in 2027?](/knowledge/q12414)
• [How should a 2027 RevOps leader define boundaries with the data team?](/knowledge/q12615)