Value Investing in Tech Stocks: How to Apply Fundamental Analysis to Software and Semiconductors
The moment many investors encounter a software company trading at 40x earnings or a chip designer with minimal book value, they conclude that traditional value investing does not apply to technology. That conclusion is partly right and mostly wrong.
You are correct that applying Ben Graham's cigar-butt framework -- buying companies near net asset value with modest P/E ratios -- rarely works in technology. The economic structure of software businesses and semiconductor companies is simply different from capital-intensive industrials. But the core principle of value investing -- buying a business for less than its intrinsic worth -- applies just as powerfully to technology. The task is using the right tools.
This article is for educational purposes only and does not constitute financial advice.
Why Traditional P/E and P/B Fail for Technology Companies
The P/E problem. Price-to-earnings ratios assume that reported earnings reflect economic reality. For software companies, this breaks down in several ways. GAAP requires the immediate expensing of research and development costs, even though R&D creates valuable assets (future products, platform improvements, data advantages) that will generate returns for years. A software company spending 30% of revenue on R&D shows depressed earnings -- not because the business is uneconomic, but because accounting forces it to expense future-oriented investments today.
Additionally, stock-based compensation (SBC) -- a massive expense at most tech companies -- is a real cost to shareholders in the form of dilution, but it does not consume cash. Companies that add SBC back to reach "non-GAAP earnings" are showing investors a rosier picture that excludes a genuine shareholder cost.
The P/B problem. Price-to-book ratios measure a company's price relative to the accounting value of its assets. For a manufacturer with factories and equipment, book value is meaningful. For a software company, the most valuable assets are the code, the customer relationships, the brand, and the talent -- none of which appear on the balance sheet at economic value. A SaaS company with $1 billion in ARR might have a book value of $100 million because most of its value is in intangible human and intellectual capital. Screening for low P/B in technology almost always screens out the best businesses.
High ROIC as the alternative lens. The better framework for technology is to recognize that what makes a great tech business is not asset cheapness but return on invested capital. A software company that earns $100 million in FCF while employing only $200 million in capital has a 50% ROIC. That is not a company you want to screen out with a P/B filter. It is a company you want to own at the right price.
Better Metrics for Technology Valuation
EV/FCF: The Foundational Multiple
Enterprise value to free cash flow is the most useful headline valuation multiple for technology companies. It avoids the earnings manipulation possible through non-cash accounting choices and focuses on actual cash generation. Unlike EV/EBITDA, it accounts for the CapEx requirement -- relevant for semiconductor companies with significant wafer fabrication investments.
A technology company with EV/FCF below 20 is typically cheap relative to its history unless the business is genuinely in secular decline. Above 40, you are paying for a substantial amount of future growth that has not yet materialized. The specific threshold depends on the growth rate: a company growing FCF at 25% annually can justify a much higher multiple than one growing at 5%.
Rule of 40: The SaaS Health Check
For Software-as-a-Service businesses specifically, the Rule of 40 provides a quick gauge of whether a company is balancing growth and profitability appropriately:
Rule of 40 Score = Revenue Growth Rate (%) + FCF Margin (%)
A score above 40 indicates a healthy SaaS business. A company growing revenue at 30% with a 15% FCF margin scores 45 -- passing the test. A company growing at 10% with a 5% FCF margin scores 15 -- well below threshold and potentially at risk.
The Rule of 40 is particularly useful because it accommodates the trade-off between growth and profitability at different stages of a SaaS company's lifecycle. A high-growth company might sacrifice margin for customer acquisition; a mature SaaS business should be generating substantial FCF margins if it is no longer investing aggressively.
FCF Margin as a Long-Term Quality Signal
FCF margin (FCF / Revenue) is the ultimate long-term efficiency measure for software. Mature, well-run SaaS businesses regularly achieve FCF margins of 25-40%. This happens because software economics improve with scale: the marginal cost of serving an additional customer is near zero once the platform is built, while the revenue from that customer is highly recurring.
Software companies with FCF margins above 20% and growing have demonstrated that their unit economics work. Below 10% in a mature business warrants investigation into whether the cost structure is sustainable.
Net Revenue Retention (NRR): The Moat Proxy
For SaaS companies, Net Revenue Retention (also called Net Dollar Retention) is one of the most powerful metrics for assessing competitive moat:
NRR = (Starting ARR + Expansion Revenue − Contraction − Churn) / Starting ARR
An NRR above 120% means that even if the company never signed a single new customer, existing customers would still grow the business by 20% per year through upsells, expanded seats, and added modules. That is the compounding effect of a deeply embedded product.
NRR above 130% characterizes the best enterprise software companies -- the ones where customers grow into larger and larger contracts over time. NRR below 100% means the company is losing more from churn and downgrades than it is gaining from expansion -- a dangerous trajectory.
ROIC: The Capital Efficiency Test
Return on invested capital is as important for technology as for any other sector. The best technology companies -- Microsoft, Apple, Visa -- have achieved exceptional long-term returns precisely because they earn very high ROICs on relatively modest capital bases.
For screening purposes, a ROIC above 20% for a technology company indicates genuine competitive advantage. Combined with an EV/FCF multiple that does not fully price in that advantage, it is the core of a compelling technology value opportunity.
How to Value SaaS Companies
SaaS companies are often discussed in terms of ARR (Annual Recurring Revenue) multiples -- the ratio of enterprise value to forward ARR. This metric exists because for a pre-profitability or low-profitability SaaS company, ARR is a reasonable proxy for the ultimate earning power of the business.
The appropriate ARR multiple depends on:
- Growth rate: faster-growing businesses warrant higher multiples
- NRR: companies with NRR above 120% have built-in compounding that justifies premium valuations
- Gross margins: SaaS gross margins typically range from 65-85%; higher gross margins mean more of each dollar of revenue flows to the bottom line
- Rule of 40 score: a benchmark of operational health
A quick-and-dirty framework: a SaaS business growing at 30%+ with NRR above 120% and Rule of 40 score above 50 might justify 10-15x ARR. A business growing at 10% with NRR near 100% and Rule of 40 score of 20 might trade at 4-6x ARR. These ranges shift with interest rates and market sentiment, but the relative ranking tends to be stable.
Semiconductor Cyclicality: How to Avoid Catching a Falling Knife
Semiconductors are among the most cyclical businesses in technology. Demand from end markets (PCs, smartphones, data centers, automotive) moves in waves, and the semiconductor industry's long production lead times mean it regularly oscillates between shortage and oversupply.
The mistake investors make is buying semiconductor stocks on the strength of peak earnings, then being surprised when earnings fall 40-60% in the next down cycle. The better framework:
Buy on trough earnings, not peak earnings. When a semiconductor company is posting its lowest earnings in five years and analysts are writing it off, the stock may be cheap on normalized (mid-cycle) earnings even though it looks expensive on current earnings. Conversely, when everything is booming and the company is posting record EPS, that EPS is inflated and the P/E ratio is deceptively low.
Use mid-cycle earnings or FCF. Average the last five or seven years of FCF to get a normalized earning power estimate. Apply a reasonable multiple to that figure. This smooths the cyclical peaks and troughs.
Watch inventory cycles. Semiconductor inventory builds at OEM customers (the companies buying chips) are an early warning of demand softness. Rising days inventory outstanding at major customers historically precedes order cancellations and semiconductor inventory corrections.
Applying Piotroski F-Score to Technology
The Piotroski F-Score is a nine-point fundamental quality scoring system originally designed for industrial and value stocks. Not all nine criteria apply equally well to technology companies, but several are highly relevant:
Most relevant for tech:
- Return on assets trend: is profitability improving?
- Operating cash flow: is OCF positive?
- Change in current ratio: is short-term liquidity improving?
- Change in gross margin: are economics improving or deteriorating at the gross margin level?
- Change in asset turnover: is the business generating more revenue per dollar of assets?
Less relevant for high-growth tech:
- Long-term debt ratio: many great tech companies legitimately add debt as they mature; de-levering is not always the right decision
- Shares outstanding: many tech companies issue shares for acquisitions and compensation; the absolute criterion needs context
A Piotroski score of 7+ for a technology company suggests improving fundamentals across multiple dimensions simultaneously -- typically a positive signal even if the absolute valuation level is not yet Ben Graham territory.
Screening Technology on ValueMarkers: EV/FCF < 25 + ROIC > 20%
A practical starting screen for finding technology value opportunities combines two filters:
EV/FCF < 25 -- ensures you are not overpaying dramatically relative to actual cash generation
ROIC > 20% -- ensures the business has genuine competitive advantages and is not just accidentally cheap
This combination is relatively rare because high-ROIC technology businesses usually trade at premium multiples. When you find one that falls into the screen -- typically after a sector selloff, a short-term earnings miss, or a macro-driven market correction -- you have a candidate worth deep-diving.
Pair this screen with the Piotroski F-Score to filter for businesses where fundamentals are improving, and you have a rigorous, quantitative entry point for technology value investing that respects the specific economics of the sector.
All financial metrics mentioned are for educational illustration. Past performance of any metric as a predictive tool does not guarantee future results.