What's the right attainment formula for lower-is-better OKR KRs?

Use `=(2*Target - Actual)/Target`. For a churn target of 1.8% with actual churn of 2.0%, this gives `(2×0.018 − 0.020)/0.018 = 88.9%` — correctly reflecting a miss. The standard `=Actual/Target` inverts the result, showing 111% attainment when you actually underperformed. Add `=MAX(..., 0)` as a floor to prevent negative scores on catastrophic misses, and `=MIN(..., 1.5)` on higher-is-better KRs to cap upside at 150%.

How do I pull OKR actuals from my P&L without manual entry?

Use a date-bounded SUMIFS anchored to your Assumptions tab: `=SUMIFS('P&L'!C:C,'P&L'!A:A,"Revenue",'P&L'!B:B,">="&Assumptions!$B$3,'P&L'!B:B,"<="&Assumptions!$B$4)`. When you update quarter dates in Assumptions, OKR actuals refresh automatically with the rest of the model. Never manually type actuals into an OKR sheet — they go stale within days and you can't reconcile them at quarter-end.

What should I do when OKR targets change mid-quarter?

Don't overwrite the original target. Maintain an `OKR History` log with columns for KR number, revision date, old target, new target, reason, and approver. Reference the live target on the main OKR tab using `=MAXIFS('OKR History'!D:D,'OKR History'!A:A,A2)`. Report both the revised attainment (against the new target) and the restated attainment (against the original) in the board pack. The history log provides full traceability for auditors and investors.

How many KRs should sit inside the financial model?

Track 5–8 KRs directly in the model — the ones with quantitative targets you can formula-drive from the P&L, headcount tab, or CRM export. Google's re:Work guidance recommends 3–4 key results per objective, typically 9–16 KRs at the company level per quarter. Most of those have a qualitative or lagging component that doesn't belong in a financial model. For the financial model, stick to KRs where the Actual column can be a formula, not a typed number.

Should OKR attainment scores cap at 100%?

No — cap at 150%, not 100%. Google's OKR framework treats consistent 100% attainment as evidence that targets were set too conservatively. A 70–85% range on ambitious targets is the intended outcome. John Doerr's *Measure What Matters* (2018) makes this explicit: "If you're consistently hitting 100%, your OKRs aren't ambitious enough." The 150% cap prevents a single blowout KR from masking underperformance elsewhere in the average.

OKR Sheet for FP&A: Wire It Into Your Model

Q: What should I do when OKR targets change mid-quarter?

Don't overwrite the original target. Maintain an `OKR History` log with columns for KR number, revision date, old target, new target, reason, and approver. Reference the live target on the main OKR tab using `=MAXIFS('OKR History'!D:D,'OKR History'!A:A,A2)`. Report both the revised attainment (against the new target) and the restated attainment (against the original) in the board pack. The history log provides full traceability for auditors and investors.

Q: How many KRs should sit inside the financial model?

Track 5–8 KRs directly in the model — the ones with quantitative targets you can formula-drive from the P&L, headcount tab, or CRM export. Google's re:Work guidance recommends 3–4 key results per objective, typically 9–16 KRs at the company level per quarter. Most of those have a qualitative or lagging component that doesn't belong in a financial model. For the financial model, stick to KRs where the Actual column can be a formula, not a typed number.

Q: Should OKR attainment scores cap at 100%?

No — cap at 150%, not 100%. Google's OKR framework treats consistent 100% attainment as evidence that targets were set too conservatively. A 70–85% range on ambitious targets is the intended outcome. John Doerr's *Measure What Matters* (2018) makes this explicit: "If you're consistently hitting 100%, your OKRs aren't ambitious enough." The 150% cap prevents a single blowout KR from masking underperformance elsewhere in the average.

This article covers four things: the full column layout with real KRs wired up, how to pull actuals from the P&L automatically, how to score churn and burn rate without getting nonsense results, and how to log target resets without destroying the audit trail.

Why Most OKR Sheets Break for Finance

The template that circulates in strategy decks has three columns: Objective, Key Result, Progress. Progress is manually typed. That works for a 10-person team tracking headcount goals. It falls apart the moment you need to reconcile it with an $18.4M ARR target already tracked in five other tabs.

There are 2 failure modes that actually matter for finance:

Manually entered actuals. Someone fills in "87%" in the Attainment column from a dashboard screenshot. By the next board meeting the number is stale and doesn't tie to anything. Finance can't sign off on it.

Flat attainment math. =Actual/Target produces nonsense for KRs where lower is better — churn, burn rate, days sales outstanding. Google's published OKR guidance (re:Work, last updated 2024) flags this as one of the most common scoring errors teams make when moving from qualitative to quantitative KRs. A churn rate that improved from 2.1% to 1.8% against a 1.8% target shows 100% attainment with the right formula and 117% with the wrong one.

OKR Sheet Layout: The Full KR Table

This is the tab structure that works. Five KRs covering the metrics that appear in a typical board pack — revenue, headcount, churn, burn rate, and pipeline. Every Actual is formula-driven except where noted.

#	Objective	Key Result	Owner	Direction	Target	Q2 Actual	Attainment	Status
1	Grow revenue	ARR at quarter-close	AE Team	Higher	$18.4M	$16.2M	88.0%	🟡
2	Scale team efficiently	Engineers hired in H1	People	Higher	12	9	75.0%	🔴
3	Improve retention	Monthly gross churn rate	CS	Lower	1.8%	2.0%	88.9%	🟡
4	Control burn	Monthly net burn	Finance	Lower	$1.20M	$1.35M	87.5%	🟡
5	Build pipeline	Qualified pipeline at EoQ	Sales	Higher	$9.6M	$11.2M	116.7%†	🟢

†Capped at 150% per scoring policy.

The Direction column is the switch that controls which attainment formula fires. For KRs 1, 2, and 5 (Higher is better), attainment is =Actual/Target. For KRs 3 and 4 (Lower is better), attainment is =(2*Target - Actual)/Target. One conditional in the Attainment column handles both:

=IF(D2="Lower", MAX((2*E2-F2)/E2, 0), MIN(F2/E2, 1.5))

The MIN(..., 1.5) caps upside at 150%. The MAX(..., 0) floors catastrophic misses at zero so you don't get negative attainment scores. John Doerr's Measure What Matters (Portfolio/Penguin, 2018) recommends the 150% cap as standard — consistent 100% attainment is a signal the target was too easy, not a reason to let scores run to 200%.

Status is conditional formatting on the Attainment column: red below 0.70, yellow 0.70–0.99, green at 1.00 and above. Google's re:Work guidance uses 0.7 as the lower threshold, and it holds in practice.

The Attainment Calculation Most Templates Get Wrong

The standard formula =Actual/Target only works when higher numbers are better. Run it on a churn rate that missed by 20 basis points and you get 111% — implying outperformance when you actually underperformed.

The fix for lower-is-better KRs:

=(2*Target - Actual) / Target

With real numbers. Target churn: 1.8%. Actual churn: 2.0%.

=(2 * 0.018 - 0.020) / 0.018
= 0.016 / 0.018
= 88.9%

That's the right answer: you missed by 11 basis points, so attainment should be below 100%. At exactly 1.8%, attainment is 100%. At 1.6% (beat by 20bps), attainment is 111%. Same math applies to burn rate — KR 4 targets $1.20M monthly net burn, actual came in at $1.35M: =(2*1.20-1.35)/1.20 = 87.5%. Correctly below par.

Wiring the OKR Sheet to Your Financial Model

The Actual column is where most OKR sheets fail. It should never be a manually entered number.

For revenue KRs, the Actual cell pulls directly from the P&L using a date-bounded SUMIFS:

=SUMIFS('P&L'!C:C,'P&L'!A:A,"Revenue",'P&L'!B:B,">="&Assumptions!$B$3,'P&L'!B:B,"<="&Assumptions!$B$4)

Assumptions!$B$3 and $B$4 are your quarter start and end dates — the same cells driving every other time-bound formula in the model. Reforecast Q2 end date in one cell, and OKR actuals refresh with everything else.

For gross margin targets (say, 67.5%), the Actual cell pulls the ratio directly:

=SUMIFS('P&L'!C:C,'P&L'!A:A,"Gross Profit",'P&L'!B:B,">="&Assumptions!$B$3,'P&L'!B:B,"<="&Assumptions!$B$4)/SUMIFS('P&L'!C:C,'P&L'!A:A,"Revenue",'P&L'!B:B,">="&Assumptions!$B$3,'P&L'!B:B,"<="&Assumptions!$B$4)

For headcount KRs, pull from the Headcount tab:

=COUNTIFS('Headcount'!D:D,"Engineering",'Headcount'!E:E,"<="&Assumptions!$B$4,'Headcount'!F:F,"Active")

For pipeline, SUMIFS against the CRM export tab filtered by stage and date. The pattern is identical across all 5 KRs: anchor the date range in Assumptions, reference the source tab, never type a number.

Resetting OKR Sheet Targets Mid-Quarter Without Losing the Audit Trail

Targets change. A Series B closes and the $19.2M ARR target gets reset to $18.4M because the pipeline shifted. If you overwrite the target, you lose the history and the board asks why attainment jumped from 73% to 88% between updates.

The pattern that works: a target history log, either on the OKR tab itself or a companion OKR History tab.

KR #	Revision Date	Previous Target	New Target	Reason	Revised By
1	2026-02-14	$19.2M	$18.4M	Pipeline revision post-Series B	CFO
3	2026-03-01	2.0%	1.8%	CS capacity added in February	VP CS

The live target on the main OKR tab then references the most recent row for each KR:

=MAXIFS('OKR History'!D:D,'OKR History'!A:A,A2)

This pulls the highest (most recent) target value for that KR without hardcoding anything. For a restated attainment against the original target:

=F2 / MINIFS('OKR History'!C:C,'OKR History'!A:A,A2)

The ARR KR came in at $16.2M. Against the revised $18.4M target, attainment is 88.0%. Against the original $19.2M, restated attainment is 84.4%. Both numbers belong in the board pack — one shows current performance, the other shows how far you moved the goalposts.

Handling Non-Financial KRs

NPS, product velocity, customer health scores — these don't live in the P&L. As of May 2026, there's no clean automated pull for most of these without an integration project.

The practical approach: a Raw Data helper tab with a structured input area. Each non-financial KR gets a labeled row with data validation constraining inputs to expected formats. The Actual cell on the OKR tab references it:

=IFERROR(INDEX('Raw Data'!B:B,MATCH(A2,'Raw Data'!A:A,0)),0)

For KRs that do have an upstream source — qualified pipeline from HubSpot, MRR from Stripe, NPS from a survey tool — ModelMonkey can write a refreshable table into the Raw Data tab directly. Type "pull qualified pipeline from HubSpot for Q2 2026" and it creates a table the SUMIFS in KR 5 can reference. That closes the last manual-entry gap without a custom integration.