Today
Q3 OKR draft feedback Active
Refining the section on revenue targets…
SQL query for churn cohort
Generated WITH RECURSIVE CTE for 90-day…
Refactor billing module
Reviewed 3 approaches, recommended strategy…
Yesterday
Onboarding email for Acme
3 variants generated, picked option B…
Summarize customer interviews
14 transcripts, 6 recurring themes…
Last 7 days
Pricing page rewrite
Comparison table and FAQ updated…
Incident postmortem draft
5-whys analysis from timeline…
Q3 OKR draft feedback
Started 14 min ago · 6 messages · 2,840 tokens
AI
Hi Avery — I've read the Q3 OKR draft you shared. There's a lot to like, but I noticed the revenue objective is missing a measurable key result. Want me to suggest two alternatives based on your historical close rate?
AV
Yes — and also flag any objective where the key results don't ladder up cleanly. Use the framework we discussed last quarter.
AI
Got it. Here's a summary table of objectives vs. measurable key results. Two objectives (Revenue, Customer Health) need sharper numbers:
Objective Key result Measurable?
Revenue growthIncrease ARR by 30%Vague
Customer healthImprove NRRVague
Product velocityShip 4 major features by Sep 30Clear
Team growthHire 6 engineers in Q3Clear
For Revenue, try: "Grow ARR from $4.2M to $5.5M by Sep 30, with at least 40% from expansion." Want me to draft the Customer Health KR in the same shape?
AV
Yes. Also show me the SQL you'd run against the warehouse to validate the Q2 baseline numbers I'm citing.
AI
For Customer Health, suggest: "Raise NRR from 108% to 118% and cut gross churn from 2.4% to 1.5% by Sep 30."
Here's the SQL — it joins subscriptions with invoices and rolls up MRR per cohort month:
WITH cohort AS (
  SELECT account_id,
         DATE_TRUNC('month', started_at) AS cohort_month,
         SUM(amount) AS starting_mrr
  FROM subscriptions
  WHERE status = 'active'
    AND started_at >= '2024-04-01'
  GROUP BY 1, 2
)
SELECT c.cohort_month,
       SUM(c.starting_mrr) AS baseline_mrr,
       SUM(CASE WHEN i.paid_at < '2024-07-01' THEN i.amount END) AS realized_mrr,
       ROUND(SUM(CASE WHEN i.paid_at < '2024-07-01' THEN i.amount END)
             / NULLIF(SUM(c.starting_mrr), 0) * 100, 1) AS nrr_pct
FROM cohort c
LEFT JOIN invoices i ON i.account_id = c.account_id
GROUP BY 1
ORDER BY 1 DESC;
Run that against warehouse.production and you'll get one row per cohort. Want me to wrap it in a dbt model?
AI
Drafting dbt model…
Press Enter to send · Shift+Enter for new line Tokens used: 0 / 128,000