OpenClaw Q1 2026 OKR Cascade

Period: January 1 - March 31, 2026 (with 6-week mid-quarter checkpoint recommended) Owner: toli (sole founder) Scoring: 0.7 achievement = success (Google/Intel standard). 1.0 = target was too easy. Rule: Every KR is passively measurable. No human checking. Data source and query specified for each.

Company Level (toli owns)#

Objective C-1: Prove the Core Four system earns its cost through autonomous operation#

The org must demonstrate that it produces more value than it consumes, and that toli's daily involvement shrinks to investor-level oversight.

KR C-1.1: Autonomous task completion rate from TBD (baseline week 1) to 85%

• Measured by: agent_events WHERE escalated = false / total_events, sampled weekly
• Data source: OpenClaw agent event logs (~/.openclaw/logs/)
• Baseline: To be measured week 1 (current estimate: <50% based on all heartbeats disabled)

KR C-1.2: Founder daily active involvement from ~2-4 hours to <30 minutes (morning brief + acknowledgment only)

• Measured by: count of toli-initiated Telegram messages to agent bots per day, 7-day rolling average
• Data source: Telegram bot API getUpdates filtered by from.id = 1613502218, daily cron query
• Baseline: To be measured week 1

KR C-1.3: API cost per completed autonomous task from TBD to <$1.50/task

• Measured by: daily Anthropic/OpenRouter API spend (from billing dashboard or usage logs) divided by count of tasks marked completed in agent logs
• Data source: ~/.openclaw/logs/usage.log cross-referenced with Anthropic billing API
• Baseline: To be measured week 1 (current spend: $30-42/day with near-zero autonomous output)

Objective C-2: Generate first meaningful revenue from souls.zip#

Revenue is existential. $80K back taxes + $130K loan with $0 current revenue. souls.zip is the primary product. This objective validates that the agent org can produce revenue, not just activity.

KR C-2.1: Monthly recurring revenue from $0 to $2,000 MRR by end of Q1

• Measured by: Stripe MRR dashboard query (stripe.subscriptions.list(status=active), sum of plan amounts)
• Data source: Stripe API, queried weekly by Cherry or Larry
• Baseline: $0

KR C-2.2: Paying customers from 0 to 20

• Measured by: stripe.customers.list with at least one active subscription, deduplicated by email
• Data source: Stripe API
• Baseline: 0

KR C-2.3: Trial-to-paid conversion rate from 0% to 25%

• Measured by: customers who converted to paid / customers who started trial, cohorted monthly
• Data source: Stripe subscription events (customer.subscription.created with status=trialing -> status=active)
• Baseline: 0% (no trial flow exists yet; infrastructure is a prerequisite)

Objective C-3: Establish OpenClaw as the reference implementation for solo-founder AI orgs#

If the Core Four model works, documenting and sharing it is both a growth channel for souls.zip and an audience-building strategy. The org itself becomes the proof.

KR C-3.1: Published content pieces about the OpenClaw model from 0 to 12 (1/week for 12 weeks)

• Measured by: count of published posts on blog/social with tag openclaw-model or equivalent
• Data source: CMS publish log or shared-knowledge/decisions/completed/content-shipped/ directory file count
• Baseline: 0

KR C-3.2: Organic inbound traffic to souls.zip from 0 to 500 unique visitors/month

• Measured by: Google Analytics or Plausible unique_visitors for souls.zip domain, monthly
• Data source: Analytics dashboard API query
• Baseline: ~0 (site exists but no traffic funnel)

KR C-3.3: Email list subscribers from 0 to 200

• Measured by: subscriber count in email platform (ConvertKit, Resend, or equivalent)
• Data source: Email platform API subscribers.list(status=active).count
• Baseline: 0

Domain Orchestrator Level#

Content Team (Jerry COO + Mary CMO) --> supports C-3, C-2#

Note: Mary (CMO) co-owns content strategy and execution oversight alongside Jerry. Jerry leads operational pipeline execution; Mary owns content strategy direction and brand narrative alignment.

Objective CT-1: Build and operate a self-sustaining content pipeline that publishes without founder involvement

KR CT-1.1: Content pieces published autonomously (no toli edits required) from 0 to 8/month

• Measured by: files in shared-knowledge/decisions/completed/content-shipped/ where toli_edited = false
• Data source: Decision log metadata per content piece
• Baseline: 0 (no autonomous content pipeline exists)

KR CT-1.2: Content pipeline cycle time (idea to published) from undefined to <72 hours average

• Measured by: timestamp delta between content-draft-created and content-shipped events in decision logs
• Data source: shared-knowledge/decisions/ file timestamps
• Baseline: To be measured week 1 (current: 44+ hours just for approval, infinite for execution)

Objective CT-2: Make Barry the reliable public-facing content execution arm

KR CT-2.1: Barry uptime from 0% (offline since Feb 17) to 95%+ during business hours

• Measured by: Barry heartbeat cron success rate (cron/jobs.json Barry heartbeat consecutiveErrors / total runs)
• Data source: Cron execution log on Barry's Docker instance
• Baseline: 0% (Barry offline)

KR CT-2.2: Jerry-to-Barry content handoff success rate from 0% to 90%

• Measured by: bridge messages sent by Jerry that resulted in confirmed publication by Barry / total bridge messages sent
• Data source: Bridge tools HTTP response log (extensions/bridge-tools/ logs)
• Baseline: 0% (pipeline broken)

Revenue Team (Cherry, Revenue Lead) --> supports C-2#

Objective RT-1: Build and execute a revenue pipeline that converts outreach into paying souls.zip customers

KR RT-1.1: Personalized outreach messages sent from 0 to 100/week by end of Q1

• Measured by: emails sent via AgentMail with template tag outreach, weekly count
• Data source: AgentMail send log or SMTP relay log for gary@souls.zip
• Baseline: 0 (Cherry has no email execution primitive currently -- MUST wire AgentMail first)

KR RT-1.2: Outreach reply rate from 0% to 8%

• Measured by: inbound replies to outreach emails / outreach emails sent, weekly cohort
• Data source: AgentMail inbox filtered by in_reply_to header matching outreach message IDs
• Baseline: 0%

KR RT-1.3: Sales-qualified leads (replied + expressed interest) from 0 to 15 by end of Q1

• Measured by: count of unique contacts in Cherry's QUEUE.md with status qualified
• Data source: workspace-cherry/QUEUE.md entries with status: qualified
• Baseline: 0

Engineering Team (Gary, CTO) --> supports C-1, C-2#

Objective ET-1: Ship souls.zip to revenue-ready state and stabilize agent infrastructure

KR ET-1.1: souls.zip core features shipped (auth, agent marketplace, Stripe billing) from 0/3 to 3/3

• Measured by: feature branches merged to main with passing CI for each milestone
• Data source: GitHub API pulls.list(state=closed, base=main) on souls.zip repo, filtered by milestone labels
• Baseline: 0/3 (Phase 2 comms sat unexecuted for 44+ hours)

KR ET-1.2: Agent infrastructure escalation rate from TBD to <15%

• Measured by: agent_events WHERE escalated = true / total_events, 7-day rolling
• Data source: OpenClaw agent event logs
• Baseline: To be measured week 1 (estimate: >50% given all autonomous ops disabled)

KR ET-1.3: Compound loop completion rate from 0% to 90% (loops that run successfully / loops scheduled)

• Measured by: cron job success count / cron job scheduled count for compound loop jobs
• Data source: cron/jobs.json execution history, consecutiveErrors field
• Baseline: 0% (all compound loops disabled)

Specialist Agent Level#

SEO (Ori) --> supports C-3, CT-1#

KR Ori-1: SEO-optimized pages on souls.zip from 0 to 10

• Measured by: pages with meta title, meta description, H1, and target keyword present
• Data source: automated site crawl (Screaming Frog export or custom script) weekly
• Baseline: 0

KR Ori-2: Indexed pages in Google from TBD to 15+

• Measured by: Google Search Console sitemaps + index coverage API
• Data source: Google Search Console API, weekly query
• Baseline: To be measured week 1

Copywriter (Carrie) --> supports CT-1, C-3#

KR Carrie-1: Content drafts produced from 0 to 12/month (feeding Jerry's pipeline)

• Measured by: files created in workspace-jerry/content-drafts/ or equivalent staging directory
• Data source: Directory file count with creation date filtering
• Baseline: 0

KR Carrie-2: First-pass approval rate (drafts accepted by Jerry without revision requests) from TBD to 70%

• Measured by: drafts approved / drafts submitted, tracked in content decision log
• Data source: shared-knowledge/decisions/completed/ content review entries
• Baseline: To be measured week 1

Design (Dory / Rory) --> supports CT-1, C-2#

KR Dory-1: Brand assets produced for souls.zip (landing page, social templates, email templates) from 0 to 5 core assets

• Measured by: asset files in workspace-dory/deliverables/ or design system directory
• Data source: Directory listing with file type filter (.fig, .svg, .png)
• Baseline: 0

KR Rory-1: Social/blog images generated from 0 to 3/week (matching Jerry's content calendar)

• Measured by: image files created in content staging directory, weekly count
• Data source: workspace-rory/deliverables/ file count with date filter
• Baseline: 0

Frontend (Perry) --> supports ET-1#

KR Perry-1: souls.zip frontend pages shipped from TBD to all pages in Phase 2 spec

• Measured by: routes with passing integration tests in souls.zip frontend repo
• Data source: CI test results filtered by test/integration/ path
• Baseline: To be measured week 1

Backend (Harry) --> supports ET-1#

KR Harry-1: souls.zip API endpoints shipped from TBD to all endpoints in Phase 2 spec

• Measured by: API routes with passing contract tests
• Data source: CI test results filtered by test/api/ path
• Baseline: To be measured week 1

Finance (Larry, CFO) --> supports C-1#

KR Larry-1: Weekly financial health report delivered to Lacie from 0 to 1/week (every Monday)

• Measured by: files in shared-knowledge/reports/financial/ with weekly cadence
• Data source: Directory file count with date gap analysis
• Baseline: 0

KR Larry-2: API cost tracking accuracy (reported vs actual billing) within 5% variance

• Measured by: |reported_cost - actual_billing| / actual_billing, monthly
• Data source: Larry's weekly report figures vs Anthropic/OpenRouter billing dashboard exports
• Baseline: To be measured month 1

Improvement Loop OKRs (Soul Engineer owns)#

Objective IL-1: Build and operate the agent self-improvement system so the org compounds in capability weekly#

KR IL-1.1: Error recurrence rate declining month-over-month

• Measured by: count of errors with reason matching a previously-seen reason / total errors, computed monthly
• Data source: agent_errors log grouped by reason field, monthly cohort comparison
• Baseline: To be measured month 1 (current: unmeasured -- error logging exists but recurrence tracking does not)
• Target trajectory: Month 1 baseline -> Month 2 at least 20% lower -> Month 3 at least 35% lower than baseline

KR IL-1.2: Escalation rate drops from current baseline to <15%

• Measured by: agent_events WHERE escalated = true / total_events, 7-day rolling average
• Data source: OpenClaw agent event logs
• Baseline: To be measured week 1 (estimate: >40% given current manual-only operation)

KR IL-1.3: Learnings promoted to soul patches from 0 to 3+/month

• Measured by: new files in .learnings/soul-patches/ directory per month
• Data source: ls -la .learnings/soul-patches/ | wc -l delta month-over-month (cron job, weekly)
• Baseline: 0 (soul patches directory does not exist yet; compound loops disabled)

Objective IL-2: Establish the observability layer so org health is always visible without human inspection#

KR IL-2.1: Automated weekly org health report produced from 0 to 1/week

• Measured by: files in shared-knowledge/reports/org-health/ with weekly cadence
• Data source: Directory file count, expected 1 new file per Sunday (Soul Engineer weekly audit output)
• Baseline: 0

KR IL-2.2: Agent STATUS.json adoption from 0/6 core agents to 6/6 core agents (Lacie, Jerry, Gary, Cherry, Larry, Soul Engineer) by end of Q1 — measured by count of agents with STATUS.json present and updated in last 24h

• Measured by: count of agents with a valid STATUS.json file updated within last 24 hours
• Data source: find ~/.openclaw/agents/*/STATUS.json -mtime -1 | wc -l (cron check)
• Baseline: 0 (STATUS.json schema not yet implemented)

Review Cadence#

KR Formula Reference#

Every KR in this document follows this formula:

[metric name] from [current baseline] to [target]
measured by [data source + exact query/path]
by [date or cadence]

Scoring guide:

• 0.0 - 0.3: Failed to make meaningful progress
• 0.4 - 0.6: Made progress but fell short
• 0.7: Success (this is the target, not 1.0)
• 0.8 - 1.0: Exceptional (or target was too conservative)

Baselines marked "To be measured week 1" must be captured by January 7, 2026. Soul Engineer is responsible for ensuring all baseline measurements are taken and recorded in this document during the first week of the quarter. Without baselines, KRs are unmeasurable and the OKR system is decorative.

Dependencies and Prerequisites#

The following must be true before these OKRs can be meaningfully pursued:

Notes#

• 6-week cycle option: Given the early stage and pace of change in agent capabilities, consider treating weeks 1-6 as Cycle A and weeks 7-12 as Cycle B. Score and rebase at the midpoint. This matches the research recommendation for year-1 AI orgs.
• Cost constraint: All autonomous operations must stay within $8-12/day (Phase 1) scaling to $25-35/day only after demonstrated value delivery. Larry tracks this weekly. If cost exceeds $25/day without corresponding revenue, Lacie triggers a cost review.
• Financial context: $80K back taxes + $130K loan + $0 revenue. Every KR that does not connect to revenue generation or cost reduction within 2 hops should be questioned. The org exists to generate revenue, not to be an interesting experiment.
• Agent team design note: The org has consolidated from 17 agents to Core Four (Lacie, Gary, Cherry, Jerry) + Barry, with 12 others being archived. Specialist KRs below transfer to the Core Four agent that spawns the relevant on-demand specialist.