RRI Technical Execution Roadmap

Overview

Executive Summary

This document is the definitive technical execution plan for Robbins Research International. It covers 17 initiatives across three phases — DERISK (remove what can kill you), UNCLOG (remove what slows you down), SCALE (grow) — sequenced across four waves from March 2026 through Q1 2027. One initiative (U5, HubSpot Bypass) is already complete.

The plan addresses an engineering organization that is structurally broken: no sprint has closed in three weeks, the team burned out after a month of 10-hour days, two engineers are bus factor 1 on revenue-critical systems serving $180M in annual revenue, and a Heroku contract deadline in September 2026 has no migration started. Simultaneously, RRI has an extraordinary AI capability — 38+ agents and tools deployed across 11 departments — that is ungoverned and under-resourced.

Two resource scenarios frame the investment decision. Scenario A (current team only) delivers 11 of 17 initiatives, defers 6 SCALE products to 2027, and puts $4.3M-$12.8M of revenue at risk — including almost certainly missing the September Heroku deadline. Scenario B (fully resourced at $1.2-1.6M incremental across two quarters) delivers all 17 initiatives on schedule, unlocks $4.9M-$20.8M in new and protected revenue, and produces a 3:1 to 14:1 ROI on the investment. The checkout pre-fill initiative (U2) alone pays for Jay Lane’s entire full-time salary in two months.

UPW readiness: Wave 0 (5 emergency initiatives) must complete before UPW. The critical gate is D4 — token caching + guest checkout fallback must handle peak event load. A Go/No-Go check is scheduled the day before the event.

View Team Structure & Process Design →

Pre-UPW — Wave 0

Pre-UPW Emergency Actions

Six initiatives must complete before UPW. One immovable deadline.

Immediate Actions

Action	Owner	Time	Why Urgent
Audit Redis `maxmemory-policy` on Linode 45.79.132.111	Zach Hardesty	30 min	If not `noeviction`, BullMQ jobs are silently disappearing
Contact Blackthorn support to drain 400K webhook queue	Spork	1 hour	18+ days to drain at max throughput. Must start immediately.
Call Obv.io support: batch attendee endpoint? Per-event rate limit?	Spork	30 min	If batch endpoint exists, U9 becomes trivial
Verify `SELECT FOR UPDATE SKIP LOCKED` in event-api	Spork	1 hour	Must verify before scaling dynos — duplicate attendee risk

Wave 0 Initiative Summary

ID	Initiative	Priority	Owner	Status	Target
U5	HubSpot Bypass for Transactional Emails	COMPLETE	Spork	COMPLETE	—
D4	Auth Load Capacity & Cognito Hardening	P0	Johnny + Zach	IN PROGRESS	Pre-UPW
D3	Secrets Audit — Phase 0 (Cognito rotation only)	P0	Johnny + Zach	IN PROGRESS	Pre-UPW
U6	Blackthorn Transaction Matching Window	P1	Johnny + Spork	STARTING	Pre-UPW
D6	Load Testing — Phase 1 (Grafana dashboard only)	P1	Zach	STARTING	Pre-UPW
U9	Obv.io Sync Acceleration — Phase 1	P1	Spork	STARTING	Pre-UPW

Go/No-Go Checklist

Before UPW proceeds, the following must be confirmed:

D4: Token caching active + guest checkout fallback tested
D4: Guest checkout fallback deployed (3-second timeout → Stripe guest mode)
D4: Chatot pre-warm cron configured (100-200 bot sessions 45min before each pitch window)
D3: Cognito client IDs from rri-stress-test rotated
U6: Blackthorn matching window extended from 5 min to exponential backoff
U6: Blackthorn support contacted, High Volume Transaction Job enabled
D6: Grafana dashboard showing live order-ingestion metrics
D6: Checkly checkout monitoring active from 12 global regions
U9: Obv.io sync interval reduced from 30s to 5s
U9: SKIP LOCKED verified before dyno scaling

If D4 fails the Go/No-Go: The entire UPW checkout flow is at risk. Layer 4 (guest checkout fallback) is the safety net — it bypasses Cognito entirely for checkout. Revenue at risk per 30-minute auth failure window: $371K (750 failed logins x $495 average).

DERISK — Remove What Can Kill You

D1. Bus Factor Elimination Program

Three engineers each represent a single point of failure on systems that generate $180M/year. Zach Hardesty is the only person who can operate the K8s/GitOps/observability/data lake infrastructure. Johnny Yarlott holds 97% of commits on the core payments API. Nick Jensen owns TonyRobbins.com and is INACTIVE in Atlassian. If any one of them is hit by a bus — or just quits — RRI’s revenue infrastructure stops.

Technical Approach

6-week pairing sprint program: 2 weeks per critical engineer with a designated backup
Driver-navigator model during live operations — backup learns by doing, not watching
CODEOWNERS enforcement on 5 critical repos to distribute code review
Migrate @alphonso77/rri-lifeforce to @rri npm org scope (Day 3, zero cost)
Nick Jensen 1:1 with Lior required Week 1 before any sprint commitment

Description

Three engineers are each a single point of failure on systems generating id="d1">

DERISK — Remove What Can Kill You

D1. Bus Factor Elimination Program

Three engineers each represent a single point of failure on systems that generate $180M/year. Zach Hardesty is the only person who can operate the K8s/GitOps/observability/data lake infrastructure. Johnny Yarlott holds 97% of commits on the core payments API. Nick Jensen owns TonyRobbins.com and is INACTIVE in Atlassian. If any one of them is hit by a bus — or just quits — RRI’s revenue infrastructure stops.

Technical Approach

6-week pairing sprint program: 2 weeks per critical engineer with a designated backup
Driver-navigator model during live operations — backup learns by doing, not watching
CODEOWNERS enforcement on 5 critical repos to distribute code review
Migrate @alphonso77/rri-lifeforce to @rri npm org scope (Day 3, zero cost)
Nick Jensen 1:1 with Lior required Week 1 before any sprint commitment

80M/year. If any one quits or is unavailable, revenue infrastructure stops.

Ideal Outcome

Every critical system has a designated backup who can operate it independently. Commit distribution shifts from >90% single-author to <60%.

Best Result

Pairing sprints complete, backups validated, Nick Jensen re-engaged. Knowledge transfer creates organizational resilience.

Worst Result (If We Don’t Do This)

Zach, Johnny, or Nick departs with no backup. Revenue-critical systems become unmaintainable. K8s migration (S2) cannot proceed.

Resources Required

People

Johnny + Zach + Nick + designated backups

Timeline

6 weeks

With Claude Code

5 weeks ⚡ accelerated

Cost

$7K-$19K (without hire)

DERISK — Remove What Can Kill You

D1. Bus Factor Elimination Program

Three engineers each represent a single point of failure on systems that generate $180M/year. Zach Hardesty is the only person who can operate the K8s/GitOps/observability/data lake infrastructure. Johnny Yarlott holds 97% of commits on the core payments API. Nick Jensen owns TonyRobbins.com and is INACTIVE in Atlassian. If any one of them is hit by a bus — or just quits — RRI’s revenue infrastructure stops.

Technical Approach

6-week pairing sprint program: 2 weeks per critical engineer with a designated backup
Driver-navigator model during live operations — backup learns by doing, not watching
CODEOWNERS enforcement on 5 critical repos to distribute code review
Migrate @alphonso77/rri-lifeforce to @rri npm org scope (Day 3, zero cost)
Nick Jensen 1:1 with Lior required Week 1 before any sprint commitment

9K (without hire)

DERISK

D2. Members Portal Migration (Node 11 → Node 22 LTS)

The Members Portal runs on Node 11, which has been end-of-life since April 2019. It uses abandoned dependencies (kue — last updated 2017, request — deprecated 2020) and has committed .env credentials in git history. This is the system where 200K+ customers manage their event tickets, coaching sessions, and product access.

Technical Approach

Target Node 22 LTS (Node 20 enters Maintenance Oct 2025)
Phase 0 credential rotation before any code changes
Replace kue with BullMQ (standardizes with order-ingestion)
Replace request with native fetch() (stable in Node 22)
Upgrade Stripe SDK v6 → v12 → v17 incrementally (v12 is the API version pinning boundary)
Kue → BullMQ: parallel queues with feature flag. Never cold-cutover a job queue.

Description

Members Portal runs on Node 11 (EOL since 2019) with abandoned dependencies and committed credentials in git history. 200K+ customers use this system.

Ideal Outcome

Portal running on Node 22 LTS with modern dependencies, no committed secrets, and a named backup for the sole contractor maintainer.

Best Result

Migration complete, Stripe SDK upgraded, BullMQ standardized across the stack. Portal becomes maintainable by multiple developers.

Worst Result (If We Don’t Do This)

Node 11 runtime vulnerability exploited. Committed credentials used for unauthorized access. Federico leaves and portal is orphaned.

Resources Required

People

Federico Del Rio + Johnny (review)

Timeline

4-6 weeks

With Claude Code

2-3 weeks ⚡ accelerated

Cost

$4.5K-$15K

DERISK

D2. Members Portal Migration (Node 11 → Node 22 LTS)

The Members Portal runs on Node 11, which has been end-of-life since April 2019. It uses abandoned dependencies (kue — last updated 2017, request — deprecated 2020) and has committed .env credentials in git history. This is the system where 200K+ customers manage their event tickets, coaching sessions, and product access.

Technical Approach

Target Node 22 LTS (Node 20 enters Maintenance Oct 2025)
Phase 0 credential rotation before any code changes
Replace kue with BullMQ (standardizes with order-ingestion)
Replace request with native fetch() (stable in Node 22)
Upgrade Stripe SDK v6 → v12 → v17 incrementally (v12 is the API version pinning boundary)
Kue → BullMQ: parallel queues with feature flag. Never cold-cutover a job queue.

5K

DERISK

D3. Secrets Audit & Credential Rotation

Six repos have committed secrets in git history. The Members Portal .env contents are unknown until git history is inspected. The rri-stress-test repo contains Cognito client IDs that compound with D4’s auth vulnerability. Credentials must be rotated first, then git history purged.

Technical Approach

Two-layer scanning: TruffleHog (free, 800+ detectors) + GitHub Secret Protection ($19/committer/month)
Rotate FIRST, purge SECOND — rotation closes the attack vector immediately
git-filter-repo for history purge (Git project’s recommended tool)
TruffleHog pre-commit hooks deployed globally

Description

Six repos have committed secrets in git history. Attack surface is unknown until audited. Credentials must be rotated before any other code changes.

Ideal Outcome

Zero verified secrets in any repo. TruffleHog CI gate blocks future commits. GitHub Secret Protection active on push.

Best Result

All credentials rotated, git history purged, pre-commit hooks deployed globally. Security posture transforms from unknown to verified.

Worst Result (If We Don’t Do This)

Committed credentials are discovered by an attacker. Data breach on a system processing millions in payments. Regulatory and reputational exposure.

Resources Required

People

Johnny + Zach + Lior

Timeline

2 days

With Claude Code

2 days

Cost

$9.8K/year

DERISK

D4. Auth Load Capacity & Cognito Hardening

Cognito has a default rate limit of 120 RPS shared across ALL user pools. At scale this can bottleneck — but for UPW March 12 (virtual, ~20,000 participants, ~1,500 buyers across 4 sales moments over 4 days), peak concurrent auth load during the biggest pitch window is likely 1-3 RPS at checkout, 10-50 RPS including page loads. Well within the 120 RPS default limit.

Three-Layer Defense for UPW

Layer	What	Timeline	Cost
1	Token caching via ElastiCache Redis	3-4 days	$50-80/month
2	Guest checkout fallback (3-sec timeout → Stripe guest)	2-3 days	$0
3	CloudWatch monitoring + Chatot alerts	1-2 days	$0

At virtual UPW, participants join via Zoom/Obv.io (not Cognito). Only buyers hit the checkout auth path. The biggest pitch window might drive 400-800 checkout attempts over 5-10 minutes. Token caching + guest checkout fallback handle this comfortably.

Description

Cognito auth must handle peak event load reliably. Three-layer defense: token caching, guest checkout fallback, and CloudWatch monitoring.

Ideal Outcome

Token caching reduces direct Cognito calls. Guest checkout fallback activates automatically if auth is slow. Real-time monitoring with alerts.

Best Result

Zero auth failures during events. Guest checkout fallback tested and ready. Pre-warm automation replaces manual process.

Worst Result (If We Don’t Do This)

Auth failure during peak sales window. 750 failed logins in 30 minutes = $371K revenue at risk per incident.

Resources Required

People

Johnny + Zach + Spork

Timeline

7 days

With Claude Code

4-5 days ⚡ accelerated

Cost

$50-80/mo + $8-12K one-time

DERISK

D5. Redis Single Point of Failure

All BullMQ job queues — including the order-ingestion pipeline that processes every payment — run on a single unmanaged Redis instance on Linode. No redundancy. No failover. No monitoring. If that single VM goes down, every payment webhook stops processing.

Technical Approach

Pre-UPW (quick fix): Linode Redis Sentinel — master + replica + 3 sentinels, 15-30s failover
Post-UPW (permanent fix): Migrate to Upstash Redis Fixed 1GB plan ($15/month) — managed, Heroku-compatible, K8s-portable
Circuit breaker: enableOfflineQueue: false on BullMQ. Redis down = 503 to Stripe. Stripe retries for 72 hours.
Must use Upstash Fixed plan, NOT Pay-as-You-Go (BullMQ polls aggressively)

Description

All BullMQ job queues run on a single unmanaged Redis VM on Linode. No redundancy, no failover, no monitoring. If it goes down, every payment stops processing.

Ideal Outcome

Managed Redis (Upstash Fixed) with automatic failover. BullMQ circuit breaker returns 503 to Stripe on Redis failure — Stripe retries for 72 hours, zero orders lost.

Best Result

Redis migrated to Upstash, circuit breaker tested, Linode decommissioned. Infrastructure becomes K8s-portable for S2.

Worst Result (If We Don’t Do This)

Linode VM crashes during a live event. Every payment webhook from Stripe, ClickFunnels, Shopify, and CopeCart stops processing. Complete revenue halt.

Resources Required

People

Zach + Johnny + Spork

Timeline

3-4 weeks

With Claude Code

2 weeks ⚡ accelerated

Cost

$15/mo (Upstash)

DERISK

D5. Redis Single Point of Failure

All BullMQ job queues — including the order-ingestion pipeline that processes every payment — run on a single unmanaged Redis instance on Linode. No redundancy. No failover. No monitoring. If that single VM goes down, every payment webhook stops processing.

Technical Approach

Pre-UPW (quick fix): Linode Redis Sentinel — master + replica + 3 sentinels, 15-30s failover
Post-UPW (permanent fix): Migrate to Upstash Redis Fixed 1GB plan ($15/month) — managed, Heroku-compatible, K8s-portable
Circuit breaker: enableOfflineQueue: false on BullMQ. Redis down = 503 to Stripe. Stripe retries for 72 hours.
Must use Upstash Fixed plan, NOT Pay-as-You-Go (BullMQ polls aggressively)

5/mo (Upstash)

DERISK

D6. Pre-Event Load Testing & Monitoring

RRI runs 9 major revenue events per year with no load testing, no synthetic monitoring, and no real-time operational dashboard. Leadership makes decisions during events on gut and panic because data is 3-4 hours delayed.

Technical Approach

Load testing: k6 OSS (free) — 90% HTTP VUs + 10% browser VUs on Zach’s K8s
Synthetic monitoring: Checkly Team ($64/month, 22 global locations, Playwright-based)
Dashboard: Grafana (existing LGTM stack) — Business Metrics / Pipeline Health / Infrastructure

Critical warning: NEVER load test against production Salesforce (governor limits) or production Redis (job corruption). Requires staging environment.

Description

RRI runs 9 major revenue events per year with no load testing, no synthetic monitoring, and no real-time operational dashboard. Leadership flies blind during events.

Ideal Outcome

Grafana dashboard showing live orders/minute during events. k6 load tests validate capacity before every event. Checkly monitors checkout from 12 global regions.

Best Result

Data-driven event operations. Leadership watches revenue per minute from iPads. Capacity issues caught 48 hours before events, not during them.

Worst Result (If We Don’t Do This)

Continued blind operations during highest-revenue moments. Failures discovered by customers, not monitoring. No capacity validation before events.

Resources Required

People

Zach + Spork + Johnny

Timeline

3 weeks

With Claude Code

1.5 weeks ⚡ accelerated

Cost

$64-388/mo + $8K setup

DERISK

D7. CI/CD on Revenue-Critical Systems

The four Heroku apps that process every dollar of RRI revenue have no CI/CD pipeline. Engineers push directly to master. rri-order-ingestion has zero automated tests blocking deployment. One bad push during an event could halt revenue.

Technical Approach

CI: lint → test with coverage threshold → TruffleHog secret scan → block merge if any fails
CD: deploy to Heroku staging → smoke test /health → GitHub Environment approval gate → promote
Reusable workflows: rri-dev/.github org repo with shared node-ci.yml and heroku-deploy.yml
Branch protection: GitHub Rulesets at org level with Evaluate mode rollout

Description

The four Heroku apps processing every dollar of revenue have no CI/CD. Engineers push directly to master. One bad push during an event halts revenue.

Ideal Outcome

All 5 repos have CI gates blocking merge. Staging deployment with smoke tests before production. Zero direct pushes to master.

Best Result

Automated testing catches bugs before deployment. Staging environment validates changes. Reusable CI workflows across all repos.

Worst Result (If We Don’t Do This)

Bad code pushed to order-ingestion during a live event. Full revenue pipeline halt. No rollback mechanism, manual recovery required.

Resources Required

People

Johnny (Week 1) + Zach (Week 2) + Federico (Week 3)

Timeline

3 weeks

With Claude Code

1-1.5 weeks ⚡ accelerated

Cost

$28-50/mo (staging dynos)

DERISK

D8. Contractor Continuity Plan

Three contractors each sole-maintain critical systems with no documentation, no backup, and unknown contract notice periods. Federico Del Rio has 374 commits on the Members Portal — the only person who can operate it. Jonathan Perez sole-maintains the RPM Planner. The Nearsure contract notice period is unknown to the CTO.

Technical Approach

4-week documentation sprint: 6 artifacts per system (README, ADRs, Context Diagram, Env Manifest, Runbook, Dependency Map)
Named backups: Josh Fuller (Members Portal), coordinate with Justin/Alex (RPM Planner), Jay + Johnny (AI tools)
Doppler ($50/month) replaces .env files — architectural fix, not process fix

Description

Three contractors sole-maintain critical systems with no documentation, no backup, and unknown contract notice periods.

Ideal Outcome

6 artifacts per system (README, ADRs, diagrams, env manifest, runbook, dependency map). Named backups validated. Contract terms documented.

Best Result

Any contractor can give notice and the system remains operational. Doppler replaces .env files. Knowledge is organizational, not personal.

Worst Result (If We Don’t Do This)

Federico or Jonathan gives 2-week notice. Members Portal or RPM Planner becomes unmaintainable. No documentation exists to onboard a replacement.

Resources Required

People

Lior + Federico + Jonathan + Josh Fuller

Timeline

4 weeks

With Claude Code

2-3 weeks ⚡ accelerated

Cost

$10K-17.5K + $50/mo (Doppler)

DERISK

D8. Contractor Continuity Plan

Three contractors each sole-maintain critical systems with no documentation, no backup, and unknown contract notice periods. Federico Del Rio has 374 commits on the Members Portal — the only person who can operate it. Jonathan Perez sole-maintains the RPM Planner. The Nearsure contract notice period is unknown to the CTO.

Technical Approach

4-week documentation sprint: 6 artifacts per system (README, ADRs, Context Diagram, Env Manifest, Runbook, Dependency Map)
Named backups: Josh Fuller (Members Portal), coordinate with Justin/Alex (RPM Planner), Jay + Johnny (AI tools)
Doppler ($50/month) replaces .env files — architectural fix, not process fix

0K-17.5K + $50/mo (Doppler)

Organizational Restructuring

Team Structure & Process Design

The engineering organization needs a structural overhaul before any roadmap initiative can execute reliably. The current state: everyone does everything, no sprints close, engineers are burned out, and 5 people are bus factor 1 on revenue-critical systems.

Current Team (12 engineers/contractors)

Team	Lead	People	Key Risk
Run (Ops)	Spork	Sean, Dean, Johnny (BF1), Zach (BF1), Josh, Tim	Spork in 6+ meetings/day, no triage layer
Build (Products)	Justin	Nick (BF1), Ken, Esmee, Caitlin, Alex, Pam, Jay (half-time)	No PM layer, no sprint velocity
Contractors	—	Federico (BF1), Jonathan (BF1), Freddy, awilmort, OnBuild	Unknown notice periods, no backups

Proposed Structure: Build vs. Run Separation

Run Team (Kanban) — Spork + new Run Team Lead + DevOps + Integration Engineer + Event Ops contractors. WIP limits, P1/P2/P3 tiers, OpsGenie on-call.
Build Team (Scrum) — Justin as VP/Head of Product + dedicated POs + PM/Scrum Master + Nick + Ken + Esmee + new Data Engineer. 2-week sprints, 20% interrupt buffer, 70%+ velocity target.
AI Team — Jay Lane (full-time) + Freddy + Daniel (under governance). BA Agent creates Jira tickets from meeting transcripts.

New Hires & Resources

Role	Team	Salary	Priority
Jay Lane (FT conversion)	AI	$175K ($87.5K incr.)	#0
Run Team Lead	Run	$140-160K	#1 BLOCKING
DevOps Engineer	Run	$130-150K	#2
Integration Engineer	Run	$120-140K	#3
Data Engineer	Build	$130-150K	#4
PM / Scrum Master	Cross-team	$120-140K	#5 HIGH
Tony AI Product Owner	Build	$130-160K	#6
TR Experience Product Owner	Build	$130-160K	#7
Event Ops Contractors (x2)	Run	$50-80K each	Immediate
MSP for Help Desk	IT	$60-180K	Phase 2
Fractional Compliance	Security	$120-240K	Phase 2

The Core Problem: Justin & Spork in Meetings

Justin and Spork spend their days in requirements meetings and stakeholder updates instead of leading their teams. The PM / Scrum Master sits in those meetings so they don’t have to. Dedicated Product Owners for Tony AI and TR Experience mean Alex isn’t spread across everything. An AI Business Analyst Agent creates Jira tickets from meeting transcripts with acceptance criteria.

Process & Ceremonies

Build Team: Sprint Planning (biweekly, 2hr) → Daily Standup (15min) → Sprint Review (biweekly, 1hr) → Retrospective (biweekly, 45min) + weekly Backlog Refinement.
Run Team: Daily Ops Sync (15min) → Weekly Ops Review (30min) → Post-Mortems within 48hr of P1.
Cross-Team: Quarterly Roadmap Planning (half day) + Monthly Product Council (90min, SVPG model) + Monthly AI Governance + Monthly DORA Metrics Review.

Product Ownership: Justin Kahn as VP/Head of Product. Dedicated POs for Tony AI ($23M ARR) and TR Experience (portal + Mastery Path + Event Passport). PCR process with 30-day lead time. PAD (U4) enforces freeze windows in code, not policy.

Full details: Current vs. proposed org charts, role descriptions, incident severity tiers (P1/P2/P3 with SLAs), sprint ceremony cards, restructuring timeline, and process tooling are all in the Team Structure & Process Design page.

UNCLOG — Remove What Slows You Down

U1. Build vs. Run Team Restructuring

RRI’s engineering team is structurally broken: everyone does everything, which means nothing gets done reliably. No sprint has closed in 3 weeks. Spork attends 6+ standing meetings daily, functioning as a human router. This is a systems failure, not a people failure. The fix: separate Run (Kanban, reactive, Spork) from Build (Scrum, planned, Justin).

Blocking Hire: Run Team Lead ($140-160K)

45-55 days post-to-offer. Must post within 5 days of announcement.

Critical action: Kill Spork’s 6+ daily meetings BEFORE March 12. Route status through Kingler (Chief of Staff AI). Erik Logan must cancel cross-department meetings.

Description

Engineering is structurally broken: everyone does everything, no sprint closes, team is burned out. Spork attends 6+ meetings daily as a human router.

Ideal Outcome

Two structured teams: Run (Kanban, ops) and Build (Scrum, features). 70%+ sprint velocity. No engineer working 10-hour days for 3+ consecutive days.

Best Result

Sustainable engineering cadence. Sprints complete on time. P1 incidents resolved in 30 minutes. Engineers do engineering, not meetings.

Worst Result (If We Don’t Do This)

Continued burnout. Key engineers quit. No sprint completes. No features ship. Revenue growth impossible.

Resources Required

People

Lior + Spork + Erik + Yogesh + new hires

Timeline

12 weeks

With Claude Code

12 weeks

Cost

$570-680K/yr headcount

UNCLOG

U2. Checkout Pre-fill & One-Click Purchase

During live events, upsell links send customers to a blank checkout form despite having portal accounts with stored Stripe cards. 30%+ checkout abandonment. Stripe Payment Element upgrade + CustomerSession API + signed JWT identity bridge.

Description

During events, upsell links send customers to a blank checkout form despite having portal accounts with stored Stripe cards. 30%+ checkout abandonment.

Ideal Outcome

Returning customers see saved cards at checkout. Apple Pay/Google Pay enabled. Signed JWT bridges identity from Obv.io to checkout.

Best Result

Checkout abandonment drops from 30% to below 15%. Stripe Link provides 14% conversion uplift. $600K-$900K annual revenue recovered.

Worst Result (If We Don’t Do This)

$600K-$900K/year in lost revenue continues. Every event loses 30%+ of upsell buyers to checkout friction.

Resources Required

People

Nick + Johnny + Federico

Timeline

4 weeks

With Claude Code

2 weeks ⚡ accelerated

Cost

$0 infrastructure, 78 eng hours

UNCLOG

U3. SSO Unification Across Systems

Customers interact with 11+ separate login systems. Each event purchase, portal visit, and content access requires a separate authentication. SSO is the foundation for Mastery Path (S3), Event Passport (S4), and Portal Unification (S5).

5-Phase Implementation

Phase 1: Cognito OIDC configuration (2 weeks)
Phase 2: Members Portal → Cognito integration (2 weeks, parallel with D2)
Phase 3: Tony AI + RPM → Cognito (2 weeks)
Phase 4: Obv.io SSO Bridge Service (3 weeks)
Phase 5: Experience API native OIDC (1 week design)

Description

Customers interact with 11+ separate login systems. Every event purchase requires re-authentication. No identity bridge between Obv.io and checkout.

Ideal Outcome

Single login across TonyRobbins.com, Members Portal, Tony AI, and Obv.io. Zero forced re-registrations during events.

Best Result

Foundation for Mastery Path (S3), Event Passport (S4), and Portal Unification (S5). Customer experience transforms from fragmented to seamless.

Worst Result (If We Don’t Do This)

SCALE products (S3, S4, S5) cannot launch — all require SSO as foundation. $5.4M+ in downstream revenue blocked. Portal unification impossible.

Resources Required

People

Johnny + Nick + Federico + Spork

Timeline

6-8 weeks

With Claude Code

3-4 weeks ⚡ accelerated

Cost

$41K-70K + $495-700/mo

UNCLOG

U4. Product Admin Dashboard

Every new product SKU, pricing phase change, and event configuration requires an engineer. Marketing changes things 5 minutes before go-live. A code freeze was attempted — it lasted 2 weeks. Without self-service tooling, code freeze is pure policy theater.

Key Unlock: BullMQ Phase Scheduler

Marketing schedules price transitions with dates. Scheduler fires at activation time, updates Stripe + Sanity + SF automatically. Zero engineering involvement. PAD “freeze mode” flag transforms policy into system-enforced guardrail.

Description

Every new SKU, pricing change, and event config requires an engineer to manually update Stripe, Salesforce, Sanity, and order-ingestion. Marketing changes things 5 minutes before go-live.

Ideal Outcome

Marketing creates and schedules SKUs without engineering tickets. BullMQ phase scheduler automates pricing transitions. Freeze mode enforced by system.

Best Result

Engineering freed from SKU management. Code freeze holds for a full event cycle. 5-10 hours per event cycle recovered.

Worst Result (If We Don’t Do This)

Engineering continues as SKU monkeys. Code freeze attempts fail (again). Last-minute changes cascade through every system. Errors at scale.

Resources Required

People

Nick + Zach + Tim Hooker

Timeline

8 weeks

With Claude Code

4-5 weeks ⚡ accelerated

Cost

$90K-160K build + $275-500/mo

UNCLOG — COMPLETE

U5. HubSpot Bypass for Transactional Emails

Status: COMPLETE as of March 4, 2026. SMTP2Go now sends confirmation emails directly from order-ingestion, bypassing HubSpot rate limits that were delaying post-purchase email confirmations by hours.

Owner: Spork · Cost: Minimal (SMTP2Go operational) · Unblocks: U6

UNCLOG

U6. Blackthorn Transaction Matching Window

After order-ingestion creates a Salesforce Opportunity, the link-opp-to-transaction job waits for Blackthorn’s record to appear. Current matching window: 5 minutes. During high-volume events, delays beyond 5 minutes cause “orphaned transactions.” There are currently 400K webhook events in the Blackthorn queue.

Technical Approach

Extend matching window from 5 min to exponential backoff: 2min, 5min, 10min, 20min, 30min, 60min, 2hr
Separate BullMQ reconciliation queue for orphan recovery (every 15min during events)
Embed Opportunity external_id in PaymentIntent metadata for deterministic matching
Deploy Spork’s async SF close-won trigger PR post-UPW

Description

After order-ingestion creates a Salesforce Opportunity, the matching job waits for Blackthorn sync. 5-minute window causes orphaned transactions during high-volume events. 400K webhooks queued.

Ideal Outcome

Exponential backoff matching (2min to 2hr). Reconciliation queue catches stragglers. Deterministic matching via PaymentIntent metadata.

Best Result

Zero orphaned transactions during events. Automated reconciliation links stragglers within 2 hours. Manual cleanup eliminated.

Worst Result (If We Don’t Do This)

$33K-$75K+ in orphaned transactions per event requiring manual cleanup. Accounting reconciliation becomes a multi-day effort after every event.

Resources Required

People

Johnny + Spork

Timeline

2-3 weeks

With Claude Code

1-1.5 weeks ⚡ accelerated

Cost

$6K-12K

UNCLOG

U7. Jay Lane Full-Time + AI ROI Framework

Jay Lane has deployed 30+ AI tools across 11 departments in 180 days — while working half-time. Proven impact: $670K+/year. Pipeline: $5M+/year. Despite this, Yogesh has throttled him to 4 projects/month because there’s no financial ROI framework. The business case: $670K proven / $175K full-time cost = 283% ROI with 3.1-month payback.

Framework

Three-dimension ROI scoring: Cost Efficiency (35%) + Time Efficiency (35%) + Growth Impact (30%)
Four-tier risk classification: Low / Medium / High / Unacceptable
Agent Registry in SharePoint: 38+ agents catalogued with risk tier, ROI score, owner
Monthly AI Governance Committee: Jay + Justin + Spork + Lior
Selene generates monthly portfolio report for Yogesh

Description

Jay deployed 30+ AI tools across 11 departments in 180 days at half-time. $670K+ proven impact. Yogesh throttles to 4 projects/month because there is no ROI framework.

Ideal Outcome

Jay full-time with ROI scoring framework. All 38+ agents catalogued with risk tiers. Monthly portfolio report for Yogesh. Daniel’s program included in governance.

Best Result

AI investment unlocked — Yogesh approves based on data. Three ungoverned AI programs unified. $5M+ pipeline activated at full capacity.

Worst Result (If We Don’t Do This)

Jay stays half-time. $5M+ pipeline remains uncaptured. Three AI programs continue ungoverned. Yogesh blocks investment without data.

Resources Required

People

Lior + Jay + Justin

Timeline

4 weeks

With Claude Code

2-3 weeks ⚡ accelerated

Cost

$75K-110K/yr incr.

UNCLOG

U8. Product Ownership Unification

Product ownership is fragmented across 6+ people. No one owns the full customer experience. This creates conflicting priorities, duplicated efforts, and SKU chaos that cascades through every system.

Approach

Appoint Justin Kahn as unified product owner (VP/Head of Product)
SVPG Product Council: quarterly strategy review + monthly operating review
Product Change Request (PCR) process: 30-day lead time, engineering impact assessment
North Star: Mastery Path progression — every product decision evaluated against it

Description

Product ownership fragmented across 6+ people. No one owns the full customer experience. Conflicting priorities, duplicated efforts, SKU chaos.

Ideal Outcome

Justin as unified VP/Head of Product. SVPG Product Council. PCR process with 30-day lead time. Dedicated POs for Tony AI and TR Experience.

Best Result

First event with clean code freeze. No last-minute unreviewed SKU changes. Every product decision evaluated against Mastery Path North Star.

Worst Result (If We Don’t Do This)

SKU chaos continues cascading through every system. Code freeze attempts fail. SCALE products (S3, S4, S5) launch without unified direction.

Resources Required

People

Justin + Erik + Lior + new POs

Timeline

8 weeks

With Claude Code

8 weeks

Cost

$25-55K/yr adj. + PO hires

UNCLOG

U9. Obv.io Sync Acceleration

The rri-event-api worker runs in a 30-second polling loop with 1-second rate limits, using 1-2 workers. At peak, customers wait 45+ minutes for their magic link. Late buyers during Day 4’s final pitch don’t get links before the event ends.

Two Phases

Phase 1 (pre-UPW): Reduce interval 30s → 5s, add Promise.all parallelism (10-20 concurrent), scale to 3-4 dynos
Phase 2 (post-UPW): Replace polling with Postgres LISTEN/NOTIFY. Sub-second dispatch. HireFire autoscaling.

Throughput: Baseline: 25 min for 1,500 buyers. Phase 1: under 2 min. Phase 2: under 60 seconds.

Description

Event-api polls in 30-second loop with 1-second rate limit. At peak sales, customers wait 45+ minutes for magic links. Late buyers miss the event entirely.

Ideal Outcome

Magic link generation within 5 seconds of purchase. Event-driven via LISTEN/NOTIFY instead of polling. HireFire autoscaling.

Best Result

Sub-5-second magic link delivery at any volume. Zero customer complaints about event access delays.

Worst Result (If We Don’t Do This)

Continued 45+ minute delays during peak sales. Refund requests from customers who paid but couldn’t access the event. Revenue leakage and brand damage.

Resources Required

People

Spork + Zach

Timeline

3 weeks

With Claude Code

1.5-2 weeks ⚡ accelerated

Cost

$10K-18K + $85-120/mo

UNCLOG

U9. Obv.io Sync Acceleration

The rri-event-api worker runs in a 30-second polling loop with 1-second rate limits, using 1-2 workers. At peak, customers wait 45+ minutes for their magic link. Late buyers during Day 4’s final pitch don’t get links before the event ends.

Two Phases

Phase 1 (pre-UPW): Reduce interval 30s → 5s, add Promise.all parallelism (10-20 concurrent), scale to 3-4 dynos
Phase 2 (post-UPW): Replace polling with Postgres LISTEN/NOTIFY. Sub-second dispatch. HireFire autoscaling.

Throughput: Baseline: 25 min for 1,500 buyers. Phase 1: under 2 min. Phase 2: under 60 seconds.

0K-18K + $85-120/mo

SCALE — Now Grow

S1. AI Agent Fleet Expansion & Governance

Unify Jay’s 30+ tools and Justin’s 8-agent bot fleet under a single governance framework. Deploy fleet registry, ROI tracking dashboard, and prioritized agent roadmap. Include Daniel’s parallel AI program from Day 1.

Description

Unify Jay’s 30+ tools and Justin’s 8-agent bot fleet under a single governance framework with registry, ROI tracking, and deployment pipeline.

Ideal Outcome

Fleet registry covering all programs including Daniel’s. ROI dashboard for Yogesh. Agent deployment pipeline with approval gates.

Best Result

Fleet expansion to 40+ agents enables $5M+ in AI-driven savings and revenue. All programs governed, measured, and prioritized.

Worst Result (If We Don’t Do This)

Three ungoverned AI programs create redundancy, security gaps, and conflicting data access. Yogesh shuts down AI investment entirely.

Resources Required

People

Jay + Justin + Spork + Lior

Timeline

4+ weeks

With Claude Code

3 weeks ⚡ accelerated

Cost

$200-500/mo compute

SCALE

S2. Heroku to K8s Migration

Migrate all production services from Heroku to K8s (Talos Linux + ArgoCD + Cilium) before the September 2026 contract end. 3 of 27+ repos already run on K8s. 16-week phased migration.

Description

Migrate all production services from Heroku to K8s before the September 2026 contract end. 3 of 27+ repos already on K8s.

Ideal Outcome

All services on K8s (Talos Linux + ArgoCD + Cilium). Heroku contract not renewed. $96K-id="s2">
SCALE

S2. Heroku to K8s Migration

Migrate all production services from Heroku to K8s (Talos Linux + ArgoCD + Cilium) before the September 2026 contract end. 3 of 27+ repos already run on K8s. 16-week phased migration.
80K/year infrastructure savings.

Best Result

Complete migration by August 2026. Infrastructure costs drop 40-60%. Team owns the full stack. Platform ready for SCALE products.

Worst Result (If We Don’t Do This)

Miss September deadline. id="s2">

SCALE

S2. Heroku to K8s Migration

Migrate all production services from Heroku to K8s (Talos Linux + ArgoCD + Cilium) before the September 2026 contract end. 3 of 27+ repos already run on K8s. 16-week phased migration.

50K+ contract extension. Unmigrated services on end-of-life platform. Infrastructure team consumed by dual maintenance.

Resources Required

People

Zach + new DevOps hire

Timeline

16 weeks

With Claude Code

10-12 weeks ⚡ accelerated

Cost

Net savings $96K-180K/yr

SCALE

S3. Mastery Path — Unified Subscription Progression

Structured post-event progression: Foundation ($149/mo) → Growth ($249/mo) → Mastery ($497/mo) → Leadership. Converts event-dependent revenue into compounding monthly subscriptions.

Description

Structured post-event progression: Foundation (id="s3">

SCALE

S3. Mastery Path — Unified Subscription Progression

Structured post-event progression: Foundation ($149/mo) → Growth ($249/mo) → Mastery ($497/mo) → Leadership. Converts event-dependent revenue into compounding monthly subscriptions.

49/mo) → Growth ($249/mo) → Mastery ($497/mo). Converts event revenue into compounding monthly subscriptions.

Ideal Outcome

Subscription tiers live. Post-event conversion flow operational. Tony AI recommending personalized next steps. Subscription dashboard showing MRR.

Best Result

$5.4M ARR from first event cohort at 20% conversion. Events become the acquisition engine, subscriptions become the product.

Worst Result (If We Don’t Do This)

RRI remains event-dependent. id="s3">

SCALE

S3. Mastery Path — Unified Subscription Progression

Structured post-event progression: Foundation ($149/mo) → Growth ($249/mo) → Mastery ($497/mo) → Leadership. Converts event-dependent revenue into compounding monthly subscriptions.

80M revenue requires constant re-acquisition. No compounding. No platform story for Dreamforce.

Resources Required

People

Justin + Build team + Jesse

Timeline

12-16 weeks

With Claude Code

8-10 weeks ⚡ accelerated

Cost

Build team capacity

SCALE

S4. Event Passport — Predictable Annual Revenue

Annual event subscription bundles: Base ($1,997/yr), + Mastery Path ($2,997/yr), VIP with 1 in-person ($7,997/yr). Revenue is predictable. Marketing sells once, not 5 separate events.

Description

Annual event subscription bundles: Base (id="s4">

SCALE

S4. Event Passport — Predictable Annual Revenue

Annual event subscription bundles: Base ($1,997/yr), + Mastery Path ($2,997/yr), VIP with 1 in-person ($7,997/yr). Revenue is predictable. Marketing sells once, not 5 separate events.

,997/yr), + Mastery Path ($2,997/yr), VIP with 1 in-person ($7,997/yr). Marketing sells once, not 5 separate events.

Ideal Outcome

Passport tiers purchasable. Event Credits auto-allocated per event. Upgrade/downgrade working. Predictable annual revenue.

Best Result

id="s4">

SCALE

S4. Event Passport — Predictable Annual Revenue

Annual event subscription bundles: Base ($1,997/yr), + Mastery Path ($2,997/yr), VIP with 1 in-person ($7,997/yr). Revenue is predictable. Marketing sells once, not 5 separate events.

0M predictable annual revenue at 5,000 passport buyers. Revenue predictability transforms company valuation.

Worst Result (If We Don’t Do This)

Events sold individually. Revenue unpredictable quarter to quarter. Marketing runs 5+ separate campaigns instead of one.

Resources Required

People

Justin + Jesse + Tim Hooker

Timeline

8 weeks

With Claude Code

5-6 weeks ⚡ accelerated

Cost

Build team capacity

SCALE

S5. Portal Unification — 11 Portals → 1

Replace 11 disconnected customer-facing portals with a single Tony Robbins portal built on Nick Jensen’s Experience API (Sanity + Next.js). One login, one home for all purchases, events, coaching, content, and Tony AI.

Description

Replace 11 disconnected customer-facing portals with a single Tony Robbins portal. One login, one home for all purchases, events, coaching, content, and Tony AI.

Ideal Outcome

Single portal on Experience API (Sanity + Next.js). Cross-sell and upsell natural. Support costs drop. Platform for Mastery Path.

Best Result

Unified customer experience. Platform story complete for Dreamforce demo. Mobile app via React Native.

Worst Result (If We Don’t Do This)

11 portals continue diverging. Customer experience remains fragmented. Cross-sell impossible. Support costs compound.

Resources Required

People

Nick + Build team

Timeline

16-20 weeks

With Claude Code

10-14 weeks ⚡ accelerated

Cost

Build team capacity

SCALE

S6. Post-Coaching Call AI Phase 2

AI records coaching Zoom calls, transcribes through Tony’s coaching standards, emails branded recap to client — written as if from Tony. Every coaching session becomes a branded touchpoint; Tony AI accumulates real coaching data.

Description

AI records coaching Zoom calls, transcribes through Tony’s coaching standards, emails branded recap to client. Every session becomes a branded touchpoint.

Ideal Outcome

Coaching calls automatically transcribed and summarized. Branded recaps sent within 1 hour. Coach dashboard live. Tony AI accumulates real coaching data.

Best Result

$364K/year savings vs Salesforce Einstein. Coaching quality improves through consistent follow-up. Tony AI gets real coaching data.

Worst Result (If We Don’t Do This)

Coaching follow-up remains manual and inconsistent. $364K/year continues going to Einstein. Tony AI has no real coaching data to learn from.

Resources Required

People

Jay + 1 engineer + Chris Schenke

Timeline

6 weeks

With Claude Code

3-4 weeks ⚡ accelerated

Cost

API usage + model inference

SCALE

S7. Lead-to-Speed AI — Sub-60-Second Response

Replace the current 7-8 minute lead routing pipeline with sub-60-second AI response via Tony AI. Every lead gets personalized, qualifying interaction before a human rep touches them.

Description

Replace 7-8 minute lead routing pipeline with sub-60-second AI response via Tony AI. Every lead gets personalized qualifying interaction before a human rep.

Ideal Outcome

New leads get AI response within 60 seconds. Warm handoff to human rep with full context. Rep dashboard showing AI-qualified leads.

Best Result

$3.5M potential at 1% conversion improvement on 700K summit leads. Industry data: responding within 1 min vs 5 min increases conversion 3-5x.

Worst Result (If We Don’t Do This)

Leads continue waiting 7-8 minutes. Cold by the time a rep responds. Conversion rates stay flat while competitors adopt AI-first.

Resources Required

People

Jay + backend engineer

Timeline

6-8 weeks

With Claude Code

3-4 weeks ⚡ accelerated

Cost

Jay + 1 engineer

SCALE

S8. Real-Time Event Intelligence Dashboard

Single dashboard during events: live orders/minute, revenue by SKU/geography/time, conversion funnel, auth success rate, pipeline health, magic link delivery rate. Selene delivers executive summaries. Builds on D6’s Grafana infrastructure.

Description

Single dashboard during events: live orders/minute, revenue by SKU/geography, conversion funnel, auth success rate, pipeline health, magic link delivery.

Ideal Outcome

Leadership watches revenue per minute from iPads. Selene delivers AI executive summaries. Primus shows rep performance. Data-driven mid-event decisions.

Best Result

First event with full visibility. Mid-event decisions based on data, not gut. Problems detected in seconds, not hours.

Worst Result (If We Don’t Do This)

Leadership continues flying blind during highest-revenue moments. Decisions based on panic and 3-4 hour delayed data.

Resources Required

People

Zach + Caitlin Noble

Timeline

4 weeks

With Claude Code

2-3 weeks ⚡ accelerated

Cost

Grafana (existing) + connections

Planning

Dependencies & Critical Path

Critical Paths

Critical Path	Sequence	Total Weeks	Target
Primary (SSO → Portal)	D3 → D4 → D2 → U3 → S5	20 weeks	September 2026
Revenue (SSO → Mastery Path)	D4 → U2 → U3 → S3	21 weeks	October 2026
Infrastructure	D1 → D7 → S2	25 weeks	September 2026 (Heroku)

Hard Dependencies

From	To	Type	Explanation
D3	D2	HARD	Must rotate credentials before any code changes
D3	D7	HARD	TruffleHog gate requires rotated secrets
D3	D8	HARD	Documentation needs sanitized env manifests
D4	D6	HARD	Cognito hardened before auth surge test
D4	U2	HARD	Cognito reliable before checkout pre-fill
D4 + D2	U3	HARD	Cognito + Node 22 before SSO integration
U4	U8	HARD	PCR requires self-service dashboard
U3	S3, S4, S5	HARD	All three SCALE products require SSO

Planning

Resource Scenarios

Scenario A: Current Team — No New Hires

All 17 initiatives sequenced against 12 available engineers/contractors. No approved hires. Burnout already active. Team at 120-130% of sustainable capacity.

Total revenue at risk (Scenario A): $4.3M-$12.8M. Delivers 11 of 17 initiatives. Defers 6 SCALE products to 2027. Almost certainly misses September Heroku deadline.

Scenario B: Fully Resourced

All new hires approved. Jay Lane converted full-time. MSP and compliance engaged. All 17 initiatives execute on schedule across two quarters (Q2-Q3 2026).

Engineering Hires

Role	Salary	Start Date	First Productive
Jay Lane (FT conversion)	$175K ($87.5K incr.)	April 1	Immediate
Event Ops Contractor #1	$65K annualized	April 14	April 28
Event Ops Contractor #2	$65K annualized	April 21	May 5
Run Team Lead	$150K + $27K recruiting	May 15	June 15
DevOps Engineer	$140K + $9.6K recruiting	May 19	June 19
Integration Engineer	$130K + $9.6K recruiting	May 21	June 21
Data Engineer	$140K + $9.6K recruiting	June 3	July 3

PM, Product Ownership & Operations

Resource	Annual Cost	What It Unlocks
PM / Scrum Master	$120-140K	Sits in requirements meetings so Justin & Spork don’t. Runs sprint ceremonies. Updates stakeholders.
Tony AI Product Owner	$130-160K	Dedicated ownership of $23M ARR product. Growth strategy, retention, feature prioritization.
TR Experience Product Owner	$130-160K	Owns portal unification, Mastery Path, Event Passport — the platform play.
MSP for Help Desk	$60-180K	Takes IT tickets off Sean. 24/7 coverage. Sean focuses on security & infrastructure.
Fractional Compliance	$120-240K	Pen testing, incident response plan, access audits, legal discovery. Table stakes at $180M.
QA (AI Agents)	$2.4-6K	Automated testing on revenue-critical paths. 95% cheaper than QA hire. See Team & Process.

Revenue Capture (Scenario B)

Initiative	Timing	Annual Revenue Impact	Confidence
U2 Checkout Pre-fill	Ships May 2026	$600K-$900K uplift	HIGH
S3 Mastery Path	October 2026	$1.5M-$5.4M ARR	MEDIUM
S4 Event Passport	September 2026	$2M-$10M ARR	MEDIUM
S6 Coaching AI	Q3 2026	$364K savings	HIGH
S7 Lead-to-Speed AI	Q4 2026	$0-$3.5M potential	LOW-MED
S2 Cost Avoidance	September 2026	$96K-$180K savings	HIGH
Engineer Retention	Ongoing	$300K-$500K avoided	HIGH
Total		$4.9M-$20.8M

ROI Calculation

Metric	Conservative	High
Incremental investment	$1.2-1.6M	$1.2-1.6M
Revenue captured / protected	$5.0M	$20.8M
ROI	3:1	14:1
Payback period	6 months	3 months

The narrative for Yogesh: We are asking for $1.2-1.6M in team and infrastructure investment across two quarters. U2 (checkout pre-fill) alone generates $600K-$900K per year. S2 (K8s migration) saves $96-180K/year. The PM and PO hires unlock the SCALE products that drive the $145M ARR product portfolio. The alternative (Scenario A) is not “save $1.2M” — it is “save $1.2M and lose $5M-$10M in revenue from products that never ship.”

Planning

Quarterly Milestones

Q2 2026 (April - June): Wave 1-2 Delivery

Theme: Stabilize the foundation. Ship the highest-revenue initiative (U2). Build the safety nets.

Milestone	Target	Initiative	Verification
Secrets audit complete	April 15	D3	Zero verified secrets in any repo HEAD
CI/CD live on all 5 repos	April 15	D7	GitHub Actions passing, branch protection active
Contractor docs complete	April 15	D8	Named backup can deploy independently
Jay Lane full-time	April 1	U7	Yogesh approved, contract signed
One-click checkout shipped	May 31	U2	Checkout abandonment below 15%
Redis migrated to Upstash	May 15	D5	All BullMQ jobs on Upstash
Bus factor sprints complete	May 31	D1	Backups can operate each system
K8s migration started	April 15	S2	First non-critical service migrated

Q3 2026 (July - September): Wave 3 + Heroku + Dreamforce

Theme: Ship SCALE products. Complete Heroku migration. Prepare for Dreamforce demo.

Milestone	Target	Initiative	Verification
SSO live across all systems	June 30	U3	Single login works across all properties
Build vs. Run operational	June 30	U1	Run Team Lead independent, 70%+ velocity
AI governance live	July 31	S1	Agent registry covers all programs
Coaching AI Phase 2	August 31	S6	AI summaries across coaching fleet
Real-time dashboard at BizMas	August 15	S8	Leadership watching revenue per minute
K8s migration complete	August 31	S2	All services off Heroku

Q4 2026 (October - December): Wave 4 SCALE Products

Theme: Launch revenue products. Harvest the infrastructure investment.

Milestone	Target	Initiative	Verification
Mastery Path MVP	October 31	S3	First subscription cohort
Event Passport in market	September 30	S4	$2.5M+ ARR run rate
Portal Unification 70%	December 31	S5	Experience API serving events, coaching, Tony AI
AI fleet at 40+ agents	December 31	S1	Full governance, ROI tracked
40-hour weeks	December 31	U1	No 10-hour days for 3+ consecutive days

Planning

Risk Register

ID	Risk	Severity	Probability	Mitigation	Owner
RF1	Nick Jensen disengagement / departure	CRITICAL	MEDIUM	Lior 1:1 Week 1 post-UPW. If Nick leaves, U2, U4, S3, S5 all at risk.	Lior
RF2	Federico contractor departure (Nearsure)	HIGH	UNKNOWN	D8 captures knowledge. Josh Fuller as backup. Review Nearsure notice period THIS WEEK.	Lior + Spork
RF4	Blackthorn 400K queue won’t drain before UPW	HIGH	MEDIUM	Contact support immediately. Extended matching window (U6) is the fallback.	Johnny + Spork
RF5	S2 Heroku migration misses September deadline	HIGH	MEDIUM	Must start April 15. New DevOps hire essential. Fallback: $150K+ extension.	Zach + Lior
RF6	Daniel’s AI program creates ungoverned third track	MEDIUM	HIGH	U7 framework includes Daniel from Day 1. Frame as “visibility” not “control.”	Lior + Jay
RF7	Spork overload in Wave 0 (4 initiatives in 9 days)	MEDIUM	HIGH	Kill daily meetings before March 12. Route through Kingler. Erik must enforce.	Erik Logan
RF8	SKIP LOCKED not implemented — duplicate attendees	HIGH	MEDIUM	Verify before scaling dynos in U9. Must verify before March 12.	Spork

Appendix

Tool Recommendations & Cost Summary

Security & Secrets

Tool	Purpose	Cost	Initiative
TruffleHog	Secret scanning (CI + pre-commit)	Free (OSS)	D3, D7
GitHub Secret Protection	Push protection	$19/committer/month	D3
git-filter-repo	Git history purge	Free (OSS)	D3
Doppler	Secrets management	$50/month	D8

Monitoring & Load Testing

Tool	Purpose	Cost	Initiative
k6 OSS	Load testing	Free (OSS)	D6
Checkly Team	Synthetic monitoring (22 global locations)	$64/month	D6
Grafana (existing)	Dashboards + LGTM stack	Already deployed	D6, S8

Infrastructure

Tool	Purpose	Cost	Initiative
Upstash Redis (Fixed 1GB)	Managed Redis for BullMQ	$15/month	D5
HireFire	Heroku dyno autoscaling	$9-19/month	U9

Team Operations

Tool	Purpose	Cost	Initiative
OpsGenie	On-call rotation & alerting	$9/user/month	U1
Swarmia	DORA metrics & sprint velocity	~$30/user/month	U1
Retool	Phase 1 UI for Product Admin	$50/user/month	U4
Swimm	Documentation with CI stale-doc checks	$8/user/month	D1

Total Tooling Cost

Scenario	Monthly	Annual
Essential only (OpsGenie + Doppler + Upstash + Checkly)	~$237/month	~$2,844/year
Full recommended stack	~$727/month	~$8,724/year

Full Cost Summary

Category	Scenario A (Current Team)	Scenario B (Fully Resourced)
Existing payroll (est.)	$1,847,500	$1,847,500
Existing contractors	$390,000	$390,000
New engineering hires (2026 pro-rated)	$0	~$625,000
PM / Scrum Master + Product Owners	$0	$380-460K
MSP + Fractional Compliance	$0	$180-420K
Recruiting fees	$0	$55,800
Tooling	~$2,844	~$8,724
Infrastructure (D4 Cognito)	$600-960/year (token caching)	$600-960/year (token caching)
Incremental ask	$0	~$1.2M-$1.6M
Revenue at risk / captured	$4.3M-$12.8M at risk	$4.9M-$20.8M captured

This document is a living artifact. It will be updated after UPW March 12 with actuals from Wave 0 execution and refined quarterly as the DERISK/UNCLOG/SCALE cycle repeats. The framework is a lens, not a checklist.

Next review: March 13, 2026 (Post-UPW retrospective) · Document owner: Lior Weinstein, Fractional CTO

Confidential Document

Executive Summary

Pre-UPW Emergency Actions

Immediate Actions

Wave 0 Initiative Summary

Go/No-Go Checklist

Technical Approach

Technical Approach

Technical Approach

Technical Approach

Technical Approach

Technical Approach

Three-Layer Defense for UPW

Technical Approach

Technical Approach

Technical Approach

Technical Approach

Technical Approach

Technical Approach

Current Team (12 engineers/contractors)

Proposed Structure: Build vs. Run Separation

New Hires & Resources

The Core Problem: Justin & Spork in Meetings

Process & Ceremonies

Blocking Hire: Run Team Lead ($140-160K)

5-Phase Implementation

Key Unlock: BullMQ Phase Scheduler

Technical Approach

Framework

Approach

Two Phases

Two Phases

Dependencies & Critical Path

Critical Paths

Hard Dependencies

Resource Scenarios

Scenario A: Current Team — No New Hires

Scenario B: Fully Resourced

Engineering Hires

PM, Product Ownership & Operations

Revenue Capture (Scenario B)

ROI Calculation

Quarterly Milestones

Q2 2026 (April - June): Wave 1-2 Delivery

Q3 2026 (July - September): Wave 3 + Heroku + Dreamforce

Q4 2026 (October - December): Wave 4 SCALE Products

Risk Register

Tool Recommendations & Cost Summary

Security & Secrets

Monitoring & Load Testing

Infrastructure

Team Operations

Total Tooling Cost

Full Cost Summary