Methodology
How the monitor works
The monitor uses reported quarterly financial statements to show whether AI infrastructure investment is producing cash returns.
What this tracks
Haukro AI Infra Monitor tracks capital spending, cash efficiency, and company financial health across the AI infrastructure value chain. Coverage is organized into five categories.
- Capital — MSFT, AMZN, GOOGL, META, ORCL
- AI Compute — NVDA, AMD, TSM, MU, AVGO
- Data Center — VRT, CSCO, APH, DELL, JCI
- Cybersecurity / AI Operations — PANW, FTNT, CRWD, ZS, NOW
- Power & Grid — CEG, VST, PWR, ETN, HUBB
Data sources and update frequency
Financial-statement data is sourced primarily from Financial Modeling Prep (FMP). For symbols in the configured fallback set, Alpha Vantage and SEC CompanyFacts may be used when FMP cannot supply usable fundamentals.
- Company results are cached server-side through Next.js with a 43,200-second (12-hour) revalidation window.
- Eligible seed-backed companies can use a Last verified snapshot during a temporary FMP failure.
- Each result includes a last-updated timestamp. A fresh provider result replaces a seed snapshot when it becomes available.
Metrics explained
| Metric | Code-defined calculation |
|---|---|
| CROIC % | TTM free cash flow ÷ average invested capital × 100. Average invested capital is (beginning invested capital + ending invested capital) ÷ 2. |
| Invested capital | Total debt + total stockholders’ equity − cash and cash equivalents − short-term investments. |
| FCF Conversion % | TTM free cash flow ÷ TTM operating cash flow × 100. |
| CAPEX/OCF % | Absolute TTM capital expenditure ÷ TTM operating cash flow × 100. |
| IC Growth % | (Ending invested capital − beginning invested capital) ÷ beginning invested capital × 100. |
| TTM FCF | The sum of the latest four quarterly free-cash-flow values. Table values are normalized to USD for comparability. |
Status classification
Status summarizes a transparent financial-stress score. High invested-capital growth does not earn points by itself; it matters only when cash returns or cash conversion are weak.
| Component | Stress points |
|---|---|
| CROIC | ≥20%: 0 · 10–<20%: 1 · 5–<10%: 2 · 0–<5%: 3 · <0%: 4 |
| FCF Conversion | ≥60%: 0 · 30–<60%: 1 · 0–<30%: 2 · <0%: 3 |
| CAPEX/OCF | ≤40%: 0 · >40–70%: 1 · >70–100%: 2 · >100%: 3 |
| IC Growth | No points for growth alone. +1 above 50% when CROIC <15% or FCF Conversion <40%; +2 above 70% when CROIC <10% or FCF Conversion <20%. |
| Status | 0–2 Strong · 3–4 Healthy · 5–6 Watch · 7–8 Weak · 9+ Stressed |
Risk Rank
Risk Rank is not a separate model. It orders covered companies by financial-stress score, with the highest stress ranked #1. Ties are resolved by lower CROIC, then lower FCF Conversion, then ticker alphabetically.
Bottleneck Risk and company financial health
Status and Risk Rank describe an individual company’s financial health using returns, cash quality, investment pressure, and conditional invested-capital growth. Bottleneck Risk is different: it describes external sector and infrastructure constraints such as chip supply, power and grid capacity, or permitting. It can affect an entire category regardless of a company’s financial condition.
Limitations and independence
- The monitor requires at least eight matching quarterly cash-flow and balance-sheet dates with all required fields before it calculates CROIC.
- Financial data can be delayed, incomplete, corrected by issuers, or unavailable for a company. Data pending is shown rather than fabricated metrics.
- Haukro AI Infra Monitor is independent and is not affiliated with, endorsed by, or sponsored by eToro or any covered company.
- The service is provided for informational and educational purposes only.