Monday funnel diagnosis delivered before standup.
Cortex growth agents run your weekly CAC reports, alert on channel anomalies, log experiment results to institutional memory, trace content to pipeline, and deliver attribution rollups — so you compound on what works instead of re-learning it.
“Across the DTC brands I've worked with, the throughline on every team that scaled is the same: operational rigor. The A-teams build their own institutional memory over time and pull away. AI doesn't flatten this. It amplifies it. But when capture is automatic, operational rigor stops being a function of discipline and starts being a function of infrastructure.”Read the founding thesis →
Jobs your agent handles.
Monday funnel diagnosis delivered before standup — 3 actions, not a data dump.
Scheduled Monday 8am recipe across CAC, conversion, and campaign performance data.
Workflow recipe (marketing.cac_report + marketing.campaign_performance): weekly funnel diagnosis with 3 prioritized actions
This week's run parameters, any data sources that returned stale results.
Weekly CAC baseline and conversion benchmarks per channel — updated automatically. When CAC drifts more than 15%, the next run surfaces the trend, not just the current number.
Monthly attribution rollup — which touchpoints drove closed deals, where to shift budget.
Scheduled first Monday of the month recipe. Ready for the growth team's monthly review.
Workflow recipe (marketing.attribution_rollup): connect channel touchpoints to closed deals, surface budget re-allocation calls
The specific month and any attribution model adjustments the team applied.
Monthly attribution results by channel — accumulated over quarters. After 3 months, trend lines emerge automatically: 'LinkedIn content consistently drives enterprise deals; paid search drives SMB but CAC is degrading.'
Log every experiment result to institutional memory so the team never re-runs a test that already has an answer.
A growth engineer or PM posts experiment results in #growth-experiments.
@cortex log this experiment: [headline test on pricing page, variant B won, +12% conversion, n=2,400, p<0.05, ran Apr 1–14]
The specific experiment parameters and the team member who reported results.
The experiment result stored permanently: 'Pricing page headline test, variant B +12% conversion, Apr 2026, n=2400, p<0.05.' The next time someone proposes a similar test, the agent surfaces this result first.
Trace which blog posts and content pieces actually drive pipeline — not just traffic.
Weekly recipe. Correlates content publish dates with inbound deal creation.
Workflow recipe: @cortex identify which content pieces in the last 30 days are correlated with inbound deal creation — show conversion rate from first touch to meeting booked
The 30-day window reviewed and any anomalies in the data this run.
Content → pipeline correlation by topic — accumulated over months. After 8 weeks, patterns surface: 'Competitor comparison posts drive 3x the demo bookings of thought leadership content for this ICP.'
Alert on paid channel CAC anomalies before the budget is burned on a broken campaign.
Daily recipe. Pages growth lead if any channel drifts more than 15% from rolling average.
Workflow recipe (daily): scan paid channels for CAC drift > 15% from 7-day average — alert if threshold crossed
Today's scan results and any channels that hit the threshold.
Rolling CAC baseline per channel, updated daily. Anomaly context builds over time — 'LinkedIn CAC spiked 3 times in Q1, each time coinciding with a bid strategy change.' Surfaces on the next alert.
Weekly cohort retention digest — where users drop off and which acquisition channels produce the stickiest cohorts.
Scheduled Friday recipe. Delivered to #growth before the week closes.
Workflow recipe: @cortex retention digest for cohorts acquired in the past 4 weeks — D7, D30 by acquisition channel
This week's cohort windows and the specific channels included.
Cohort retention curves by acquisition channel — accumulated over quarters. Feeds into attribution and content-to-pipeline analysis automatically: the growth stack starts connecting its own dots.
Working in every job above.
What happened in this session — the Slack thread you referenced, the customer you named, the draft you iterated on. Persists through the conversation and short-term window. Volatile by design: fades unless it matters.
Facts that proved their value — accessed repeatedly, confirmed accurate, promoted by the system. “Drift raised their Pro price to $400/seat.” “Acme Corp runs on Salesforce CPQ.” Written once by accident, available forever to every session.
Stop re-running experiments. Start compounding what you know.
Deploy in 10 minutes. No DevOps required.