How to Value Technology Companies
Technology companies have broken the traditional valuation playbook. When Amazon traded at 300x trailing earnings in the early 2000s, most value investors dismissed it as absurd. Two decades and a 50x return later, the problem was not with Amazon but with valuation frameworks that could not account for the economics of a platform business investing heavily in long-term dominance. When Snowflake went public in 2020 at roughly 175x forward revenue, the valuation seemed disconnected from any fundamental anchor. Yet the company's hypergrowth trajectory, best-in-class net retention, and consumption-based revenue model placed it in a category where revenue multiples, not earnings multiples, were the appropriate lens.
Valuing technology companies requires adapting traditional frameworks to businesses that often have negative current earnings, massive intangible assets, and growth trajectories that can make even generous projections look conservative in hindsight. The key is not to abandon fundamental analysis but to apply it with the right metrics for the right type of technology business.
Why Standard Metrics Fall Short
Traditional valuation metrics were designed for industrial-era businesses: companies with physical assets, stable margins, and predictable growth. Technology companies violate many of these assumptions.
Earnings are deliberately suppressed. Many technology companies choose to reinvest aggressively rather than show profits. Amazon's operating margins were below 5% for most of its history, not because the business was unprofitable, but because management chose to invest every marginal dollar into new warehouses, AWS infrastructure, and market expansion. Using P/E ratios for companies that are intentionally minimizing earnings produces misleadingly high multiples or no multiple at all.
Assets are intangible. A software company's most valuable assets, its code, its algorithms, its customer relationships, its brand, do not appear on the balance sheet under GAAP. Microsoft's balance sheet shows roughly $120 billion in goodwill and intangible assets from acquisitions, but the value of Windows, Office 365, Azure, and LinkedIn far exceeds any accounting figure. Book value and asset-based valuation are largely irrelevant.
Revenue quality varies enormously. A dollar of recurring subscription revenue from a SaaS company is worth far more than a dollar of one-time hardware revenue. The subscription dollar will likely repeat next year (and grow), while the hardware dollar requires a new sale to replace it. Valuation multiples must account for these quality differences.
SaaS and Cloud Companies
Software-as-a-Service companies have become the most prominent subsector of technology, and the market has developed specialized metrics for valuing them.
Annual Recurring Revenue (ARR). The annualized value of all active subscriptions. ARR is the baseline for SaaS valuation because it represents the revenue the company would generate over the next twelve months if it added no new customers, lost no existing ones, and made no pricing changes. ARR growth rate is the single most important metric for high-growth SaaS companies.
Net Revenue Retention (NRR). Measures the dollar value of revenue from existing customers compared to the prior year, including upsells and churn. An NRR above 120% means the company is growing its revenue from existing customers by 20% annually before adding any new business. Snowflake's NRR peaked above 170%. CrowdStrike consistently maintained NRR above 120%. High NRR indicates strong product-market fit and pricing power.
Rule of 40. A widely used heuristic that combines revenue growth rate and profit margin (typically free cash flow margin). If the sum exceeds 40%, the company is considered healthy. A company growing at 30% with a 15% FCF margin scores 45, which is solid. A company growing at 10% with a 5% FCF margin scores 15, which is concerning. The Rule of 40 captures the tradeoff between growth and profitability, recognizing that high-growth companies may justifiably sacrifice margins.
EV/Revenue multiples. For SaaS companies without positive earnings, enterprise value to revenue (often forward revenue) is the primary valuation metric. As of early 2025, public SaaS companies traded across a wide range: 5-10x for slower-growth names, 15-25x for mid-growth with strong NRR, and 30x+ for hypergrowth leaders. The multiple correlates most strongly with revenue growth rate, followed by NRR and free cash flow margin.
LTV/CAC ratio. Customer lifetime value (LTV) divided by customer acquisition cost (CAC). This unit economics metric indicates whether the company's sales and marketing spending is generating adequate returns. A ratio above 3.0 is generally considered healthy. Below 1.0, the company is spending more to acquire customers than those customers will generate in revenue over their lifetime.
Platform and Marketplace Companies
Platform businesses like Alphabet, Meta, Amazon's marketplace, and Airbnb have different economic characteristics than pure SaaS companies. Their value derives from network effects: the platform becomes more valuable to each user as more users join.
Gross Merchandise Value (GMV) or Total Payment Volume (TPV). For marketplaces, the total transaction volume flowing through the platform is a key metric. Shopify, for instance, reports GMV to show the scale of commerce enabled by its platform. The take rate (revenue as a percentage of GMV) and its trend over time indicate the platform's pricing power.
Revenue per user and engagement metrics. For advertising-based platforms, revenue per average user (ARPU), daily active users (DAUs), and engagement time drive valuation. Meta's ARPU in North America exceeded $60 per quarter by 2024, compared to roughly $5 in Asia-Pacific. Understanding geographic ARPU differences is important for modeling growth.
EV/EBITDA and P/E for profitable platforms. Unlike pre-profit SaaS companies, mature platforms like Alphabet and Meta are highly profitable and can be valued on traditional earnings-based multiples. Alphabet traded at roughly 20-22x forward earnings in early 2025, a premium to the S&P 500 but a discount to its historical average, reflecting concerns about AI competition and regulatory risk.
Semiconductor Companies
Semiconductor companies are cyclical technology businesses with massive capital requirements and volatile demand. They require a different valuation approach than software companies.
EV/EBITDA adjusted for cycle position. Semiconductor companies look cheapest at the peak of the cycle (high earnings, low multiple) and most expensive at the trough (depressed earnings, high multiple). Using mid-cycle earnings, an average of earnings across a full demand cycle, produces a more accurate valuation than using current-year figures.
Design wins and market share. For fabless semiconductor companies like Nvidia, AMD, and Qualcomm, future revenue depends on winning design sockets in next-generation products. A design win at a major customer (Apple, Samsung, Tesla) represents years of future revenue. Analysts track design win announcements and customer qualification progress as leading indicators.
Capital intensity. Integrated device manufacturers like Intel and Samsung, and foundries like TSMC, require enormous capital expenditures. TSMC's annual capex exceeded $30 billion in recent years. Valuation must account for the capex required to maintain technology leadership, which reduces free cash flow relative to EBITDA.
Hardware and Device Companies
Traditional hardware companies, including PC manufacturers (Dell, HP), networking equipment (Cisco), and consumer electronics (Apple), are valued more conventionally.
P/E and EV/EBITDA. These companies generally have positive, stable earnings and can be valued on standard multiples. Dell trades at a single-digit P/E reflecting its commodity hardware business. Apple trades at 30x+ reflecting its ecosystem, services revenue, and brand loyalty.
Services mix. The transition from one-time hardware sales to recurring services revenue has become the single most important valuation driver for many hardware companies. Apple's Services segment (App Store, iCloud, Apple Music, Apple TV+, AppleCare) generates higher margins and more predictable revenue than hardware. As the services mix increases, the overall valuation multiple expands.
Product cycle dynamics. Hardware companies are subject to product cycles. A major product launch (new iPhone, new gaming console) drives above-trend revenue that normalizes in subsequent quarters. Valuation should be based on normalized revenue across a product cycle rather than on peak or trough quarters.
Growth-Adjusted Valuation
For technology companies of all types, growth is the dominant valuation variable. Two companies with identical current revenue will trade at vastly different multiples if one is growing at 30% and the other at 5%.
The PEG ratio (P/E divided by earnings growth rate) and the growth-adjusted EV/Revenue ratio (EV/Revenue divided by revenue growth rate) attempt to normalize for growth differences. A company trading at 20x revenue but growing at 50% has a growth-adjusted multiple of 0.4x, which is cheaper on a growth-adjusted basis than a company trading at 8x revenue but growing at 10% (growth-adjusted multiple of 0.8x).
These ratios are crude but useful for initial screening. A full DCF model that projects revenue growth, margin expansion, and capital requirements over a 10-year horizon provides a more rigorous assessment. For high-growth technology companies, the DCF's terminal value will dominate, which means the assumptions about when growth decelerates and where margins stabilize matter more than near-term projections.
Common Valuation Mistakes in Tech
Extrapolating hypergrowth indefinitely. A company growing revenue at 60% annually will not do so for a decade. The base effect alone makes it mathematically difficult. A company with $1 billion in revenue growing at 60% reaches $110 billion in 10 years, which would make it one of the largest technology companies in the world. Growth deceleration is not a possibility but a certainty; the question is when and how sharply.
Ignoring stock-based compensation. Technology companies distribute large amounts of equity to employees. If stock-based compensation represents 15% of revenue, ignoring it inflates free cash flow and understates the true cost of operating the business. Treat SBC as a real expense that dilutes existing shareholders.
Confusing revenue with value. Revenue multiples are a shortcut, not a valuation method. A dollar of 80%-margin SaaS revenue is worth far more than a dollar of 20%-margin hardware revenue. EV/Revenue comparisons across companies with different margin profiles are misleading without margin adjustment.
Valuing optionality at infinity. Technology companies often have "option value" from potential new products, markets, or business models. Alphabet has Waymo; Amazon has healthcare initiatives; Microsoft has gaming. These options have value, but assigning unlimited value to possibilities that may never materialize is speculation, not analysis. Value what exists and treat optionality as upside potential with a probability-weighted estimate.
Put these principles into practice. Track fundamentals, build portfolios, and analyze stocks with AI-powered insights.
Start Free on GridOasis →