The Investment Thesis
A data-driven case for why this market is underserved, why we're the team to capture it, and the evidence from 1,885 real client interactions that proves product-market fit.
The Opportunity
There are 3.6 million+ Indian-origin professionals in the US. They work at Google, Meta, Amazon, Microsoft, Apple. They earn $200K-$500K total comp with RSUs, ESPP, and multiple income streams. They have bank accounts in India, rental properties, and complex visa-dependent tax situations.
Nobody serves them properly. TurboTax can't handle FBAR, PFIC, or multi-state RSU reporting. Generic CPAs charge $300 for a filing that misses $3,000+ in RSU cost basis errors. H&R Block barely registers (4 mentions in 1,885 interactions).
From our Client Intelligence Report: 66% of clients have foreign assets, but only 14.6% are on H1B. That means 77% of the addressable market is invisible to anyone marketing "H1B tax specialist." The real market is every immigrant professional with cross-border complexity. It's 5-10x larger than current targeting suggests.
Market Sizing
| Segment | Population | ARPU | Market Value |
|---|---|---|---|
| H1B visa holders (active) | 800,000+ | $400-800 | $320M-640M |
| Green card holders (India-origin) | ~1,000,000 | $300-600 | $300M-600M |
| NRIs with US tax obligations | ~500,000 | $400-800 | $200M-400M |
| F1/OPT students | ~300,000 | $200-300 | $60M-90M |
| Tech contractors (1099) | ~200,000 | $300-500 | $60M-100M |
| Total Addressable Market | 2.8M+ | $940M-$1.8B |
Focusing on tech workers with RSU/ESPP + foreign assets + $150K+ income: ~500K households, $400 avg ARPU = $200M SAM. This is the segment where we have proven product-market fit through 1,885 real interactions.
Product-Market Fit Signals
These aren't projections. These are from analyzing 1,885 real interactions with paying and prospective clients. → Full Client Intelligence
The "Show Me One Thing" Lever
The single most powerful conversion insight from 705 consultation calls:
"Based on your W-2, you're likely missing [specific deduction] worth $[amount] — that alone pays for our fee."
Emotional Co-occurrence = Conversion Predictor
| Client State | % of Clients | Conversion |
|---|---|---|
| Anxious + Confused + Price-sensitive + Trust-seeking Complete Overwhelm | 15.6% | Highest — when one emotion breaks, others fall |
| Anxious + Price-sensitive + Trust-seeking | 10.9% | High — needs one value demonstration |
| Price-sensitive + Trust-seeking only | 10.5% | Medium — rational evaluator |
| Anxious + Trust-seeking (not price-sensitive) | 7.4% | Very high — scared but committed |
Unit Economics
Pricing Tiers (Validated)
| Tier | Price | Client Profile | Frequency |
|---|---|---|---|
| Basic | $199 | Simple W-2, standard deductions | ~15% of clients |
| Core | $399 | RSU, multiple income sources | ~40% of clients |
| Pro | $799 | Business, rental, foreign income | ~30% of clients |
| Complex | $999 | Multi-entity, PFIC, 6+ properties | ~15% of clients |
49% of consultation calls (304 of 637) involve questions AI can handle as first-line support — pricing, timelines, filing status. This means the same team can handle 2x the clients without hiring. AI handles the repetitive; experts handle the complex. → See AI bot data
Blended ARPU Estimate
Weighted by tier distribution: ~$520/client. With platform expansion (LLC, insurance), ARPU grows to $800-1,200 through cross-sell without additional acquisition cost.
Growth Mechanics
Predictable Seasonal Ramp
4x surge from January to February across all three channels. This is predictable — tax season demand. The business model benefits from concentrated demand: scale up seasonally, retain through year-round intelligence.
Distribution Channels
| Channel | Why It Works | Status |
|---|---|---|
| Referral (community) | 48 referral mentions in data. Indian tech community is deeply connected via WhatsApp, Blind, FAANG Slack | Active |
| Viral tax tools | "Your Real Salary" card, RSU Tax Gap Calculator — shareable, Spotify-Wrapped style | Research |
| FAANG FIRE partnership | 26K subscribers, exact same audience. Co-branded RSU tax guide. | Planned |
| Content (LinkedIn/Reddit) | Each research finding is a viral-ready hook: "We analyzed 1,885 interactions..." | Ready |
Temporal Patterns
The Moat
We tested this against 1,885 real interactions. Not theory — observation. → Full competitive analysis
"No competitor — not TurboTax, not a generic CPA, not H&R Block — understands the intersection of immigration, equity compensation, and cross-border tax. 74% of our clients touch at least one of these. A generalist cannot serve them."
The Intersection Nobody Else Serves
Competitor Landscape (From Real Data)
| Competitor | Client Mentions | Threat | Our Edge |
|---|---|---|---|
| Traditional CPA | 689 | Medium | $300 generic filing misses NRI-specific deductions worth $3K+ |
| TurboTax | 190 | Low | Can't handle FBAR, PFIC, multi-state RSU, rental depreciation |
| TaxGPT | — | Low | Document Q&A only. No aggregated intelligence, no proactive insights |
| Mezzi | — | Low | Requires Plaid linking. We work from docs you already have |
| H&R Block | 4 | None | Walk-in retail. Different universe from digital-native audience |
Platform Expansion
The tax product is Customer Zero. The underlying engine — AI-powered document extraction, structured form-filling, agentic workflows — is a horizontal capability. → See platform strategy
| Vertical | Same Engine, New Domain | Market |
|---|---|---|
| Tax Filing (current) | Ingest W-2/1099 → Fill organizer → AI advisory | $940M-1.8B |
| LLC Formation | Ingest ID → Navigate state portals → File articles | $500M+ |
| Insurance Quoting | Ingest loss runs → Fill carrier applications | $1B+ |
| Benefits Admin | Ingest census data → Automate HRIS enrollment | $2B+ |
Every vertical we enter improves the core engine. Document extraction accuracy improves across all verticals simultaneously. The 5th customer gets a better product than the 1st — automatically.
Why This Team
The evidence is this page. And the Client Intelligence Report. And the Thesis & Technology page. And the fact that all of this exists.
"We don't guess. We mine data, decode patterns, and build every decision on a foundation of research. This intelligence library is living proof — it grows with every insight we uncover."