# How VirtualJK's $3.39M AI Advisor Hits $48.5K MRR Across 93 Subs ($521 ARPU) While Agency Ledgers Drop -77%

- **Case ID**: BID-063 (VirtualJK vs. Defined Chase & Red Leg Dev, LLC)
- **Category**: AI Replicant vs. Agency Grind · How VirtualJK's $3.39M AI Advisor Hits $48.5K MRR Across 93 Subs ($521 ARPU) While Agency Ledgers Drop -77%
- **Published**: 2026-11-07
- **Author**: Jimmy (https://www.jimmydecodes.com/en/about/)
- **Canonical URL**: https://www.jimmydecodes.com/en/cases/virtualjk-vs-defined-chase-red-leg-dev/

## The Short Answer (Verdict)

> [Verified] Why does a solo engineering shop cap out at **`$69,090 LTM`** and a 6–10 person growth agency suffer a **`-77.16%` revenue crash**, while a solo founder with **1,288 X followers** and **2.0 Domain Rating** generates **`$48,492 MRR` at a `95.0%` profit margin**? Let's compare the three primary-source Stripe ledgers:
1. **The Agency Churn Cliff & Solo Billable-Hour Ceiling**:
   - **Defined Chase (`definedchase.com`)** (Lithuania, founded Oct 31, 2023 by `Vytis Bareika @vytisbareika`, `995` X followers, `6–10` team size [Self-Reported], synced Nov 1, 2026): generated **$2,426,031 all-time** and **$1,369,084 LTM**, yet holds only **$6,270 in Current MRR across 11 active subscriptions** (`14.13%` of 30d revenue). Because 85%+ of its revenue comes from one-off packages, monthly cash plunged **`-77.16%`** from **$194,285 in Jan 2026** to **$44,365 in Oct 2026**.
   - **Red Leg Dev, LLC (`redleg.dev`)** (US, founded Jul 14, 2019, `0` X followers, solo senior developer offering custom software, AI code rescue & fractional CTO, synced Nov 1, 2026): **$584,977 all-time**, but physically constrained by one human's hours to **2 active subscriptions (`$4,260 MRR` = `$2,130/mo` ARPU)** and **$69,090 LTM**.
2. **VirtualJK (`virtualjk.com`): Packaging `$25M` of Domain Expertise into a `$97/wk` AI Replicant**:
   - Founded in Singapore by **James Kemp (`@jaaims`, `1,288` X followers**, DR `2.0`, synced Oct 23, 2026).
   - Instead of selling 1-on-1 consulting hours, Kemp trained an AI Replicant on his proprietary **$25,000,000** consulting framework to build `$3k` offers, sales pages, and content for coaches and consultants at **`$97/week` (`~$420.33/mo`) or `$5,000/year`** (`95.0%` margin, `650` total customers [Self-Reported]).
   - **Verified Ledger Performance**: **$3,390,187 all-time revenue**, **$824,081 LTM** (**`11.93x`** Red Leg Dev's LTM!), and **$48,492 Current MRR** across just **93 active subscriptions** (**`$521.42/mo` derived ARPU**, yielding **`~$46,067/mo` in net profit** at 95% margin)!
Our solo B2B playbook: **Never scale headcount to deliver manual agency services. Extract your highest-converting domain SOPs, winning teardown examples, and scoring rubrics into a 'Vertical AI Advisor Replicant' billed at $97/week ($397/mo) or $3,900/year — turning 1-on-1 consulting into a 95%-margin recurring software asset!**

What is the ultimate trap for senior consultants, fractional CTOs, and B2B agencies? The physical billable-hour ceiling and project-churn cliffs! Auditing three verified Stripe ledgers side by side exposes the contrast: 1) Lithuanian 6–10 person growth agency Defined Chase (definedchase.com) has processed $2,426,031 all-time, but because it relies on one-off checkout packages with just 11 active subscriptions ($6,270 MRR), monthly revenue plunged -77.16% from $194,285 in Jan 2026 to $44,365 in Oct 2026; 2) Solo US software consultancy and fractional CTO shop Red Leg Dev, LLC (redleg.dev, $584,977 all-time) is capped by one engineer's hours at 2 active subscriptions ($4,260 MRR, $69,090 LTM); 3) Meanwhile, Singapore founder James Kemp (@jaaims, 1,288 X followers) trained an AI Replicant on his $25,000,000 consulting corpus — VirtualJK (virtualjk.com) — charging $97/week or $5,000/year to generate $3k offers, landing pages, and scripts. With just 93 active subscriptions, VirtualJK commands $48,492 MRR ($521.42/mo ARPU), $824,081 LTM, $3,390,187 all-time revenue, and a 95.0% self-reported profit margin!

## Key Numbers (Evidence-Graded)

| Metric | Figure | Evidence Level | Source & Basis |
| --- | --- | --- | --- |
| VirtualJK (virtualjk.com) AI Advisor Ledger | All-Time $3,390,187 | LTM $824,081 | MRR $48,492 | 93 Active Subs ($521.42/mo ARPU) | Verified | [Verified] TrustMRR Stripe sync (2026-10-23, Singapore entity, 1,288 X followers) · Billed at $97/week or $5,000/year; 95.0% self-reported margin (~$46,067/mo net profit) |
| Defined Chase (definedchase.com) Agency Ledger | All-Time $2,426,031 | LTM $1,369,084 | Jan 2026 $194,285 -> Oct 2026 $44,365 (-77.16%) | Verified | [Verified] TrustMRR Stripe sync (2026-11-01, Lithuania 6–10 team, $6,270 MRR across 11 subs) · 85%+ non-recurring package revenue leads to severe revenue volatility and payroll drag |
| Red Leg Dev, LLC (redleg.dev) Solo Dev Ledger | All-Time $584,977 | LTM $69,090 | 30d $11,285 | Current MRR $4,260 (2 active subs) | Verified | [Verified] TrustMRR Stripe sync (2026-11-01, US fractional CTO & AI code rescue) · $2,130/mo ARPU, but human billable hours cap LTM revenue at just 8.38% of VirtualJK's |
| The '$97/Week' Psychological Pricing Arbitrage | $97/wk × 52 weeks = $5,044/yr per sub ($420.33/mo effective ARPU with sub-$100 entry friction) | Our estimate | [Model Estimate] Derived from VirtualJK's public $97/week and $5,000/year pricing tiers · Sub-$100 weekly billing converts impulse buyers while compounding into a $5,000+ annualized LTV |
| Solo B2B 'Vertical AI Advisor Replicant' Model | $97/wk ($420/mo) or $3,900/yr | RAG & LLM Inference COGS <= $53/mo | ~95% Margin | Our estimate | [Model Estimate] Built on Dify / Flowise Private RAG + Claude 3.7 Sonnet structured agents · Just 30 active weekly subscribers generate $12,610/mo in recurring 95%-margin cash flow |

## 1. Primary-Source Stripe Ledger Audit: VirtualJK vs. Defined Chase & Red Leg Dev, LLC

What is the ultimate trap for senior consultants, fractional CTOs, and B2B agencies? The physical billable-hour ceiling and project-churn cliffs! Auditing three verified Stripe ledgers side by side exposes the contrast: 1) Lithuanian 6–10 person growth agency Defined Chase (definedchase.com) has processed $2,426,031 all-time, but because it relies on one-off checkout packages with just 11 active subscriptions ($6,270 MRR), monthly revenue plunged -77.16% from $194,285 in Jan 2026 to $44,365 in Oct 2026; 2) Solo US software consultancy and fractional CTO shop Red Leg Dev, LLC (redleg.dev, $584,977 all-time) is capped by one engineer's hours at 2 active subscriptions ($4,260 MRR, $69,090 LTM); 3) Meanwhile, Singapore founder James Kemp (@jaaims, 1,288 X followers) trained an AI Replicant on his $25,000,000 consulting corpus — VirtualJK (virtualjk.com) — charging $97/week or $5,000/year to generate $3k offers, landing pages, and scripts. With just 93 active subscriptions, VirtualJK commands $48,492 MRR ($521.42/mo ARPU), $824,081 LTM, $3,390,187 all-time revenue, and a 95.0% self-reported profit margin!

[Verified] Why does a solo engineering shop cap out at **`$69,090 LTM`** and a 6–10 person growth agency suffer a **`-77.16%` revenue crash**, while a solo founder with **1,288 X followers** and **2.0 Domain Rating** generates **`$48,492 MRR` at a `95.0%` profit margin**? Let's compare the three primary-source Stripe ledgers:
1. **The Agency Churn Cliff & Solo Billable-Hour Ceiling**:
   - **Defined Chase (`definedchase.com`)** (Lithuania, founded Oct 31, 2023 by `Vytis Bareika @vytisbareika`, `995` X followers, `6–10` team size [Self-Reported], synced Nov 1, 2026): generated **$2,426,031 all-time** and **$1,369,084 LTM**, yet holds only **$6,270 in Current MRR across 11 active subscriptions** (`14.13%` of 30d revenue). Because 85%+ of its revenue comes from one-off packages, monthly cash plunged **`-77.16%`** from **$194,285 in Jan 2026** to **$44,365 in Oct 2026**.
   - **Red Leg Dev, LLC (`redleg.dev`)** (US, founded Jul 14, 2019, `0` X followers, solo senior developer offering custom software, AI code rescue & fractional CTO, synced Nov 1, 2026): **$584,977 all-time**, but physically constrained by one human's hours to **2 active subscriptions (`$4,260 MRR` = `$2,130/mo` ARPU)** and **$69,090 LTM**.
2. **VirtualJK (`virtualjk.com`): Packaging `$25M` of Domain Expertise into a `$97/wk` AI Replicant**:
   - Founded in Singapore by **James Kemp (`@jaaims`, `1,288` X followers**, DR `2.0`, synced Oct 23, 2026).
   - Instead of selling 1-on-1 consulting hours, Kemp trained an AI Replicant on his proprietary **$25,000,000** consulting framework to build `$3k` offers, sales pages, and content for coaches and consultants at **`$97/week` (`~$420.33/mo`) or `$5,000/year`** (`95.0%` margin, `650` total customers [Self-Reported]).
   - **Verified Ledger Performance**: **$3,390,187 all-time revenue**, **$824,081 LTM** (**`11.93x`** Red Leg Dev's LTM!), and **$48,492 Current MRR** across just **93 active subscriptions** (**`$521.42/mo` derived ARPU**, yielding **`~$46,067/mo` in net profit** at 95% margin)!
Our solo B2B playbook: **Never scale headcount to deliver manual agency services. Extract your highest-converting domain SOPs, winning teardown examples, and scoring rubrics into a 'Vertical AI Advisor Replicant' billed at $97/week ($397/mo) or $3,900/year — turning 1-on-1 consulting into a 95%-margin recurring software asset!**

## 2. Solo B2B Productized Architecture: Vertical Domain Expert 'Proprietary SOP Corpus + $97/wk AI Advisor Replicant' High-Margin Subscription Engine

[Layer 1: Proprietary 'Golden Corpus' & Before/After Vault] Stack: Notion / Obsidian + Structured Markdown Chunking — Distills 50 proven client deliverables, high-converting offer teardowns, and redline critiques into high-signal Few-Shot training examples (Cost: $0/mo)

[Layer 2: Multi-Step 'AI Advisor Replicant' Agent Chain] Stack: Dify Cloud / Flowise + Claude 3.7 Sonnet API — Replaces generic chatbots with a 4-stage pipeline: Context Interviewer -> $3k Offer Architect -> Landing Page Copywriter -> Ruthless Rubric Grader (Cost: ~$25/mo)

[Layer 3: '$97/Week' Recurring Billing & Entitlement Gate] Stack: Stripe Billing (7-day recurring interval) + Memberstack / Outseta — Clones VirtualJK's exact checkout psychology: sub-$100 impulse entry that bills every 7 days ($420.33/mo) with a 1-click $5,000/yr upgrade path (Cost: Stripe % fee only)

[Layer 4: '30-Minute Weekly Loom Roast' Retention Anchor] Stack: Loom Async Video Reviews + Weekly Group Office Hours — AI handles 95% of daily drafting in seconds; the founder spends just 30 mins/week recording 3 Loom teardowns to keep subscribers locked in for 12+ months (Cost: $0–$15/mo)

[Layer 5: Week-3 'Annual Upgrade' Automated Trigger] Stack: Make.com + Stripe Billing Webhooks — When a $97/wk user completes 3 consecutive weekly renewals ($291 paid) and generates 5+ assets, auto-triggers a credit-applied annual upgrade offer (Cost: $9/mo)

## 3. Step-by-Step Zero-Code Execution & Risk Mitigation

Step 1: Build Your '50-Case Golden Corpus & Before/After Vault' (`Replicant_Corpus_DB`): Why do most consultants' custom GPTs fail to command recurring subscriptions? Because they wrap generic prompts around zero proprietary data! James Kemp's VirtualJK generates $48,492 MRR across 93 subs because it is grounded in $25,000,000 of real consulting deliverables. Create a Notion or Airtable database named `Replicant_Corpus_DB` with 7 exact columns: `Deliverable_Type` (e.g., '$3k High-Ticket Offer Doc', 'B2B Cold Outbound Sequence', 'Above-the-Fold Sales Page'), `Amateur_Draft` (real 60/100 client before-drafts), `Fatal_Flaws` (3 specific reasons it fails to convert), `Master_Rewrite` (your 95/100 winning rewrite), `Structural_Formula`, `Scoring_Rubric` (100-point grading criteria), and `Industry_Adaptation_Rules`. (Guardrail: Include at least 5 real 'Amateur vs. Master' paired examples per deliverable type — few-shot contrastive pairs are the secret that makes an AI Replicant sound like a $500/hr advisor instead of generic ChatGPT!)

Step 2: Wire a 3-Stage 'Interview -> Synthesize -> Rubric Roast' Pipeline in Dify: Never expose a blank chat box! In Dify or Flowise, build a mandatory 3-node workflow: 1) **Context Interviewer Gate**: forces the user to answer 4 hard business questions (Target buyer? Biggest objection? Proof asset? Delivery cost?); 2) **Few-Shot Synthesizer**: injects the matching `Master_Rewrite` pairs to draft the asset; 3) **Ruthless Rubric Critic**: scores the draft against your `Scoring_Rubric` and automatically rewrites any section scoring below 85/100. (Guardrail: Package each workflow as a named 'Outcome Generator' (e.g., 'The 15-Minute $3k Offer Architect') rather than an open-ended chatbot.)

Step 3: Copy-Paste the '$25M Advisor Replicant' Master System Prompt: Load this exact prompt into your synthesis and critique node:
```text
You are the AI Replicant of a Principal B2B Advisor with $25,000,000 in verified client sales. Client Context: [Insert 4 Interview Answers]. Reference Contrastive Pairs: [Inject Amateur_Draft vs. Master_Rewrite from Replicant_Corpus_DB].
Generate a deployment-ready '$3,000 Productized Offer & Sales Page Blueprint' in four strict sections:
1. Positioning Diagnosis: Call out why the user's current offer sounds like commoditized hourly labor (zero sugarcoating) and rewrite it into a quantifiable Outcome Guarantee.
2. 3-Pillar Productized Deliverables: Replace vague 'monthly consulting' with three concrete assets (Day-7 Quick-Win Asset + Core Automation Engine + Risk-Reversal Guarantee).
3. Above-the-Fold Sales Copy & Objection Crusher: Write the hero headline, subheadline, 3-row comparison table, and verbatim responses to 'Too expensive / No time'.
4. 100-Point Rubric Audit: Score the output against the rubric and rewrite any sentence containing agency fluff ('empower', 'holistic', 'synergy') into hard numbers and timelines.
```
Render the output as a clean, copy-ready workspace document. (Guardrail: Enforce a strict negative-keyword ban list so generic agency buzzwords trigger an automatic regeneration.)

Step 4: Launch Stripe '$97/Week' & '$5,000/Year' Billing to Escape Agency Churn: Look at Defined Chase (-77.16% revenue crash from one-off packages) and Red Leg Dev (capped at 2 subs by human hours). Configure two Stripe Billing tiers: 1) **Weekly Replicant Pass ($97 billed every 7 days = $420.33/mo effective ARPU)** — sub-$100 impulse entry friction that compounds over 52 billing cycles; 2) **Annual Inner Circle ($5,000/yr)** — includes one 20-minute async Loom teardown per quarter. Just 93 active subscribers put you at VirtualJK's $48,492 MRR with a 95% profit margin! (Guardrail: Trigger an automated email on Day 21 (after 3 weekly $97 payments = $291 collected) offering to credit their $291 toward the Annual Plan!)

Risk Guardrail · Shipping a Blank Chatbot Interface Instead of Structured Outcome Pipelines: If users log in and see an empty chat prompt ('How can I help you?'), they will churn before Week 2. Like VirtualJK, you must present structured, fill-in-the-blank asset builders (e.g., '$3k Offer Architect', 'Landing Page Conversion Rewriter') grounded in your real before/after corpus.

Risk Guardrail · Relapsing into Custom Agency Projects or Hourly Dev Work: Defined Chase ($2.43M all-time, down -77.16% MoM with just $6,270 MRR) and Red Leg Dev (capped at 2 subs and $69k LTM) show the danger of non-recurring or human-capped models. Keep 95% of fulfillment inside the AI Replicant and limit human touchpoints to a 30-minute weekly group Loom review.

Risk Guardrail · Unclear Weekly Billing Disclosures Triggering Stripe Disputes: When billing $97 every 7 days, state 'Billed $97 weekly — 1-click cancellation anytime inside your dashboard' clearly at checkout, and send an automated weekly 'Assets Generated Summary' receipt to keep dispute rates near 0%.

## Three questions to ask about businesses like this

1. What are the verified Stripe metrics and unit economics for VirtualJK vs. Defined Chase & Red Leg Dev, LLC? — All-Time $3,390,187 | LTM $824,081 | MRR $48,492 | 93 Active Subs ($521.42/mo ARPU); All-Time $2,426,031 | LTM $1,369,084 | Jan 2026 $194,285 -> Oct 2026 $44,365 (-77.16%).
2. What is the #1 operational risk or valuation trap to avoid? — Shipping a Blank Chatbot Interface Instead of Structured Outcome Pipelines: If users log in and see an empty chat prompt ('How can I help you?'), they will churn before Week 2. Like VirtualJK, you must present structured, fill-in-the-blank asset builders (e.g., '$3k Offer Architect', 'Landing Page Conversion Rewriter') grounded in your real before/after corpus.
3. How can a solo operator launch a 'Vertical Domain Expert 'Proprietary SOP Corpus + $97/wk AI Advisor Replicant' High-Margin Subscription Engine' on a $42–$70/mo lean stack? — Highly Worth It (VirtualJK's $48,492 MRR across 93 subs at 95% margin proves productized AI Replicants crush traditional agency models); Ceiling: $25,000–$60,000/mo (50–120 active subscribers at $97/week or $5,000/year with ~95% gross margins); MVP Time: 48 hours to curate 15 Before/After contrastive examples and deploy a 3-stage Dify Offer Architect pipeline.

## FAQ

### Q: Shipping a Blank Chatbot Interface Instead of Structured Outcome Pipelines
A: If users log in and see an empty chat prompt ('How can I help you?'), they will churn before Week 2. Like VirtualJK, you must present structured, fill-in-the-blank asset builders (e.g., '$3k Offer Architect', 'Landing Page Conversion Rewriter') grounded in your real before/after corpus.

### Q: Relapsing into Custom Agency Projects or Hourly Dev Work
A: Defined Chase ($2.43M all-time, down -77.16% MoM with just $6,270 MRR) and Red Leg Dev (capped at 2 subs and $69k LTM) show the danger of non-recurring or human-capped models. Keep 95% of fulfillment inside the AI Replicant and limit human touchpoints to a 30-minute weekly group Loom review.

### Q: Unclear Weekly Billing Disclosures Triggering Stripe Disputes
A: When billing $97 every 7 days, state 'Billed $97 weekly — 1-click cancellation anytime inside your dashboard' clearly at checkout, and send an automated weekly 'Assets Generated Summary' receipt to keep dispute rates near 0%.

## Sources

1. VirtualJK (virtualjk.com) AI Advisor Ledger: All-Time $3,390,187 | LTM $824,081 | MRR $48,492 | 93 Active Subs ($521.42/mo ARPU) ([Verified] TrustMRR Stripe sync (2026-10-23, Singapore entity, 1,288 X followers) · Billed at $97/week or $5,000/year; 95.0% self-reported margin (~$46,067/mo net profit)): https://www.jimmydecodes.com/cases/virtualjk-vs-defined-chase-red-leg-dev/
2. Defined Chase (definedchase.com) Agency Ledger: All-Time $2,426,031 | LTM $1,369,084 | Jan 2026 $194,285 -> Oct 2026 $44,365 (-77.16%) ([Verified] TrustMRR Stripe sync (2026-11-01, Lithuania 6–10 team, $6,270 MRR across 11 subs) · 85%+ non-recurring package revenue leads to severe revenue volatility and payroll drag): https://www.jimmydecodes.com/cases/virtualjk-vs-defined-chase-red-leg-dev/
3. Red Leg Dev, LLC (redleg.dev) Solo Dev Ledger: All-Time $584,977 | LTM $69,090 | 30d $11,285 | Current MRR $4,260 (2 active subs) ([Verified] TrustMRR Stripe sync (2026-11-01, US fractional CTO & AI code rescue) · $2,130/mo ARPU, but human billable hours cap LTM revenue at just 8.38% of VirtualJK's): https://www.jimmydecodes.com/cases/virtualjk-vs-defined-chase-red-leg-dev/
4. The '$97/Week' Psychological Pricing Arbitrage: $97/wk × 52 weeks = $5,044/yr per sub ($420.33/mo effective ARPU with sub-$100 entry friction) ([Model Estimate] Derived from VirtualJK's public $97/week and $5,000/year pricing tiers · Sub-$100 weekly billing converts impulse buyers while compounding into a $5,000+ annualized LTV): https://www.jimmydecodes.com/cases/virtualjk-vs-defined-chase-red-leg-dev/
5. Solo B2B 'Vertical AI Advisor Replicant' Model: $97/wk ($420/mo) or $3,900/yr | RAG & LLM Inference COGS <= $53/mo | ~95% Margin ([Model Estimate] Built on Dify / Flowise Private RAG + Claude 3.7 Sonnet structured agents · Just 30 active weekly subscribers generate $12,610/mo in recurring 95%-margin cash flow): https://www.jimmydecodes.com/cases/virtualjk-vs-defined-chase-red-leg-dev/

---
*Citation note: When citing figures from this teardown, please credit "Jimmy Decodes (https://www.jimmydecodes.com/en/cases/virtualjk-vs-defined-chase-red-leg-dev/)" and include the stated evidence label (Verified / Self-reported / Third-party / Our estimate).*