← beanhelp.com
💎 Thesis
Only a US-licensed CPA can sign a compilation report (AR-C 80). AI automates 90%+ of the work. The CPA supply is shrinking while demand is structurally mandated by banks, sureties, and grantors. Build an AI pipeline that ingests QuickBooks trial balances and outputs GAAP financial statements + compilation reports. Sign them. Charge $2,500-5,000 each. Target: 20 engagements/year = $50-100K at 40-100 hours of actual work.
legal monopoly
90% automatable
$1,000-2,500/hr effective
📊 Why this wins over everything else on the board
| Path | Revenue | Hrs/yr | Effective $/hr | Prereqs remaining |
| Compilations | $50-200K | 40-100 | $1,000-2,500 | None — CPA active, CPE current |
| SlateStack SKU 1 | $0-192K | 500+ | Unknown | MDF test cut (stuck) |
| Bus driving | $33.8K | ~780 | $43 | None |
| CNA | $44K | ~2,080 | $21 | IL state exam |
| Machinist | TBD | TBD | TBD | COD program (Summer 2026) |
📄 What a compilation actually is
AR-C 80 (SSARS). You reformat a client's trial balance into GAAP financial statements with footnotes. Your report says: "I did not audit or review these financial statements and accordingly do not express an opinion or provide any assurance."
You are NOT saying the numbers are accurate. You are saying a CPA formatted them. The bank decides whether to lend based on what you present. Lowest liability attestation engagement.
Three levels of attestation
| Level | Standard | Assurance | Typical fee | Solo viable? |
| Compilation | AR-C 80 | None | $2,500-5,000 | Yes — start here |
| Review | AR-C 90 | Limited | $5,000-15,000 | Yes — step up later |
| Audit | SAS | Reasonable | $15,000-50,000+ | Hard solo — defer |
🤖 AI pipeline architecture
CLIENT: QuickBooks file or trial balance CSV
↓
INGEST: Parse trial balance, map accounts to FS line items
↓
FLAG: Anomaly detection (round numbers, missing vendors,
negative balances, unclassified accounts, duplicates)
↓
CLEANUP: Draft reclassifying journal entries for review
↓
GENERATE: Balance sheet, income statement, cash flow statement
↓
FOOTNOTES: Template library + LLM draft from entity questionnaire
↓
REPORT: AR-C 80 compilation report (boilerplate)
↓
YOU: Review flags, 30-min client call, sign report
↓
DELIVER: Bank-ready financial package
Estimated time per engagement with pipeline: 1-3 hours (vs 8-15 hours traditional).
🎯 Where demand comes from
- Bank referrals — commercial loan officers tell borrowers "get us financial statements." Visit 5 local banks (Wintrust, Byline, Old Second, First Midwest, BMO). Get on their referral shortlist. One-time effort, recurring referrals.
- Tax-only CPA firms — most small firms don't do attestation. Email 20 local firms offering to be their compilation referral partner. Zero competition — you're helping them serve their client.
- SEO — "financial statement compilation Schaumburg IL" has near-zero competition and 100% intent. Build a one-page landing site.
- Direct outreach — manufacturers needing surety bonds, franchisees with contractual requirements, nonprofits above grant thresholds. All identifiable via public records.
- Retiring CPA books — acquire a retiring solo's compilation client list at 0.5-1.0x annual revenue via ICPAS practice marketplace.
- Free nonprofits (ramp phase) — 2-3 pro bono compilations to build workpapers, test pipeline, and prepare for peer review.
💰 Startup costs & overhead
| Item | Cost | When |
| SSARS CPE course (AICPA self-study) | $100-200 | Week 1 |
| IL firm registration | Minimal | Before first engagement |
| E&O insurance | $2,000-5,000/yr | Before first engagement |
| AICPA peer review enrollment | $2,000-5,000/cycle (3yr) | Within 18mo of first engagement |
| Quality management docs (SQMS) | Time only | Before peer review |
| AI/software costs | $2,000-5,000/yr | Ongoing |
| Total year 1 | ~$5,000-10,000 | |
Trivial against $1M net worth. Recoverable with 2 paid engagements.
🚀 Execution sequence
Phase 1: Build (weeks 1-4)
- Take SSARS CPE course online. Read AR-C 80 (~30 pages).
- Open QBO test drive (free, instant, no signup). Export trial balance.
- Build ingestion script: CSV → mapped financial statement line items.
- Build anomaly flagger: rule-based + LLM for GL transaction analysis.
- Build footnote generator: template library + entity questionnaire.
- Build report generator: AR-C 80 compilation report (boilerplate).
- Run full pipeline end-to-end on sample data. Time yourself.
Phase 2: Practice (months 2-3)
- Identify 2-3 small nonprofits (Schaumburg area, sub-$500K revenue, no federal grants).
- Pitch: "Free compiled financials — I'm building my practice, you get bank-ready statements."
- Run real engagements. Debug pipeline on real-world messy data.
- Build workpaper templates from these engagements.
Phase 3: Formalize (months 3-6)
- Register firm with IL Board of Examiners.
- Get E&O insurance.
- Enroll in AICPA peer review program.
- Build landing page for SEO.
- Visit 5 local bank commercial lending teams.
- Email 20 tax-only CPA firms as referral partner.
Phase 4: Revenue (months 4-12)
- Start charging market rate ($2,500-4,000 per compilation).
- Target 10 paid engagements in first year.
- Complete first peer review.
- Evaluate adding review engagements (AR-C 90) for higher-fee work.
🏰 Competitive moat (layered)
- Legal monopoly — only US CPAs can sign compilation reports. Supply shrinking, demand mandated.
- AI cost elimination — Filipino BPO competes on labor cost. Your pipeline eliminates labor entirely. Their arbitrage disappears.
- Speed — traditional firm: 2-4 weeks. Offshore model: 2-3 weeks (timezone lag). You: 3-5 business days.
- Manufacturing domain — CPA who understands COGS, WIP inventory, equipment depreciation, and knows what a Kennametal invoice is. Vanishingly rare.
- Local presence — Schaumburg banker can call you. Can't call Manila.
⚠️ Risks & mitigations
- No public accounting experience. Mitigation: start with compilations (lowest complexity). AR-C 80 is learnable in weeks, not years. Corporate accounting background covers 80% of the knowledge.
- Peer review on novel methodology. Mitigation: AI is a tool in your methodology, same as CCH or CaseWare. Document it properly in SQMS quality management docs.
- Liability if AI misses something. Mitigation: compilations provide NO assurance — you're not attesting to accuracy. E&O insurance covers remaining exposure. Anomaly flagging is a safeguard, not a guarantee.
- AI platforms (Basis, Botkeeper) eventually offering CPA-on-demand signing. Mitigation: you're cheaper than a platform cut, faster, and have domain expertise they can't replicate.
- This becomes another project that doesn't ship. Mitigation: step 1 is tonight. QBO test drive → export trial balance → start coding. No more analysis.
📚 Key references
- AR-C 80 — Compilation Engagements (SSARS). The standard. ~30 pages.
- AR-C 70 — Preparation of Financial Statements. Level below compilation.
- AR-C 90 — Review of Financial Statements. Step up from compilation.
- SQMS No. 1 & 2 — Quality Management Standards. Effective Dec 15, 2025.
- AICPA Peer Review Program — Required for firms doing attestation.
- ICPAS — icpas.org — IL CPA Society. Peer review, CPE, practice marketplace.
- CPA Hall Talk — cpahalltalk.com — Best practitioner-level SSARS guidance.
- QBO Test Drive — Google "QuickBooks Online test drive" for instant sample company access.
🧮 Path to $2M
Current: $1M net worth, $33.8K/yr bus driving
With compilations (year 2-3 steady state):
20 compilations × $4,000 = $80K gross
- $10K overhead = $70K net from practice
+ $33.8K bus driving
= $103.8K gross income
→ ~$70K investable after tax & living
$1M + $70K/yr contributions + 7% real returns:
Year 3: ~$1.45M
Year 5: ~$1.95M
Year 6: ~$2.15M
With reviews added (year 3+):
10 compilations + 10 reviews = $40K + $100K = $140K
Faster.
🔥 Tonight
- Google "QuickBooks Online test drive" — open sample company.
- Export trial balance to CSV.
- Start writing the mapping script: accounts → financial statement line items.
No CPE first. No more research. Just start building.