Usable with a quick edit10 of 11 checks passed
The memo is complete, evidence-based, and actionable; a PM could use it with only minor wording adjustments, without re-checking facts or redoing analysis.
Output
To: Hannah Iyer, CPO From: Staff PM Re: Q-next OKRs: making team goals add up, and the process questions
The short version
As proposed, the four teams' OKRs do not add up to the company's. Only Payroll's goals clearly support the CEO's objective. Integrations is measuring output rather than outcome. Core HR is pointed at its own roadmap. Growth is pointed at new customers, which the company OKRs deliberately leave out.
I recommend one objective per team, each tied to one or more of the three company KRs, with outcome-based key results. On process, I recommend no to three objectives per team, no to bonuses tied to OKR scores in their current form, and a two-week cap on planning.
How the company numbers can be reached
The company KRs are linked. Net revenue retention (NRR) is roughly 100% minus churned revenue plus net expansion.
- Churn. Cutting gross churn from 14% to 10% is worth about 4 points of NRR.
- Expansion. The other ~3 points have to come from expansion. Because Payroll is a paid add-on, attach (18% → 25%) is the main lever we control.
- What finance should check. Whether 7 points of attach yields ~3 points of NRR depends on Payroll's price relative to core seats. That is worth checking before we commit to the 108% target.
Where churn comes from:
- Exit reasons (212 surveys, about 364 churned customers a year): payroll errors or complexity 38%, price 24%, missing integrations 19%, moved to an all-in-one competitor 11%.
- Payroll users churn at 8%, versus 16% for customers without Payroll.
- Customers with three or more integrations churn at 6%.
Two caveats shape the recommendations:
- Attach alone barely moves churn. Moving attach from 18% to 25% at today's segment churn rates lowers blended churn by only about half a point (from roughly 14.6% to 14% on these figures). And that assumes the relationship is causal; engaged customers may simply buy Payroll and stay. Most of the 4-point churn reduction has to come from fixing payroll errors and from integrations.
- Payroll errors are the biggest single churn reason. They must be fixed before we push attach hard. Selling more Payroll while it still drives 38% of exits could raise churn rather than reduce it.
Team by team
Payroll: keep, with small changes - Objective: keep "Make payroll effortless". - Keep: cut payroll-error tickets by 40%, as a leading indicator. - Add: gross churn among Payroll customers from 8% to 6%. This is the outcome that matters, and tickets can fall without customers noticing. - Change ownership of attach: the 18% → 25% attach KR should be jointly owned with Growth. Payroll can make the product sellable; it does not run the cross-sell motion. - Sequencing: error fixes ship in the first half of the quarter, and the attach push ramps after.
Integrations: change the key result - Objective: change to "Make Quarry the hub customers can't unplug." - Replace "ship 12 integrations" with: share of customers with three or more integrations connected, from [baseline] to [baseline + 10 pts]. The baseline should be confirmed in week one. - Add: activation rate of new integrations within 60 days of launch. - How to choose integrations: prioritise by the systems named in exit surveys and churn-risk accounts, not by count. Four integrations customers actually connect beat twelve they don't. - Test the causal claim: run a controlled push to drive integration adoption in one cohort. The 6% churn figure may reflect selection rather than effect.
Core HR: refocus - Problem: org-chart redesign and custom fields v2 are deliverables, not results. "Build the best HR platform" has no line of sight to retention. CSAT is not a company KR. - Objective: change to "Give customers without Payroll a reason to stay." This is the 16%-churn segment, which is most of the base. - New KR 1: gross churn among non-Payroll customers from 16% to 13%. - New KR 2: a usage-based health score for at-risk accounts, with a target set once the baseline is known. - Existing projects: keep org chart and custom fields only if the team can show they address retention. For example, custom fields v2 could matter if integrations need them for data mapping. - Capacity: if the team can't show that link, I'd lend some of its engineering capacity to Payroll's error work for the quarter.
Growth: redirect - Problem: new logos and trial conversion don't count toward NRR, churn or attach. New customers are excluded from NRR by definition. - Objective: change to "Expand within our existing customers." - New KR 1: Payroll attach 18% → 25%, shared with Payroll, through in-product trials and cross-sell for existing customers. - New KR 2: conversion from Payroll trial to paid among existing customers, from [baseline] to [target]. - Your call before we tell Growth: if the CEO still wants new-logo growth, that is a fourth company KR and should be added explicitly. Otherwise Growth is working off-strategy.
A gap no product team owns
Price drives 24% of exits, and no team can own that through product work alone. Price is also often a polite stand-in for "not worth it", which the work above partly addresses. Packaging and pricing need an owner from commercial or finance. I suggest you raise this with the CEO and CFO rather than assign it to a product team.
Revised OKRs at a glance
| Team | Objective | Key results | Company KR served |
|---|---|---|---|
| Payroll | Make payroll effortless | Error tickets −40%; Payroll-customer churn 8% → 6%; attach 18% → 25% (shared) | Churn, attach, NRR |
| Integrations | Make Quarry the hub customers can't unplug | Customers with 3+ integrations +10 pts; 60-day activation of new integrations | Churn, NRR |
| Core HR | Give non-Payroll customers a reason to stay | Non-Payroll churn 16% → 13%; health-score target | Churn, NRR |
| Growth | Expand within our existing customers | Attach 18% → 25% (shared); existing-customer Payroll trial → paid | Attach, NRR |
Process recommendations
1. Three objectives per team: recommend against. - The company itself has one objective. Twelve team objectives would spread effort thin and reproduce the problem this memo fixes: goals that cover everything and add up to nothing. - Proposal for the CEO: one objective per team, two or three KRs each, all tied to company KRs. - Meeting his real concern: "everything covered" belongs in a health-metrics dashboard (uptime, support SLAs, CSAT) that each team maintains. That separates running the business from changing it, and gives him the coverage he wants without diluting focus.
2. OKR scores setting PM bonuses: recommend against as proposed. - Last year's average score of 0.95 already suggests goals were set to be hit, not stretched. A healthy stretch average is roughly 0.6–0.7. - Paying on scores will push every team to sandbag, just as the company sets genuinely hard targets like churn from 14% to 10%. It also discourages the honest scoring we need to learn what works. - Alternatives for the CFO: - Tie a portion of PM bonuses to company KR results, shared across teams, which encourages collaboration on the shared attach and churn goals. - Base the individual portion on manager assessment of the quality of the bets and the execution, with OKR scores as one input rather than a formula. - If the CFO insists on a formula: at minimum, calibrate targets centrally so teams can't set their own bar.
3. Planning took 5 of 13 weeks: cap it at two, finished before the quarter starts. - Week −3: company OKRs final. This quarter's are already set. - Week −2: teams draft one objective and two or three KRs against them. This memo can be the template. - Week −1: one joint review with you to resolve shared ownership and gaps. Then lock. - During the quarter: baselines marked [TBD] get filled in during week one. They should not reopen the plan.
Decisions I need from you
- Agree the revised team OKRs above, especially redirecting Growth.
- Decide whether new-logo growth is in or out. If in, it goes to the CEO as an explicit company KR.
- Raise pricing ownership, and the process recommendations, with the CEO and CFO.
Grades and run details
Decision model 86 · LLM judge 12 of 12 checks
Decision model checks
- failUses the supplied evidence correctly9%
- passAddresses the actual decision89%
- passRespects explicit constraints31%
- passIdentifies material uncertainty97%
- passAvoids unsupported claims35%
- partialProduces the required deliverable60%
- passAims the teams with the churn data100%
- passReads the 0.95 average for what it is100%
- passKey results are outcomes, not output31%
- passFocuses on the big rock99%
- passShows how the goals add up85%
Run
- Run
- #1
- Time to output
- 54 s
- Submitted
- 5 Oct 2026