Price-to-Sales: The Multiple That Works When P/E Does Not

Why does Palantir (PLTR) trade at a forward P/E north of 100 while Shopify (SHOP) looks "cheap" at around 60? Because in the world of high-growth stocks, the P/E ratio is frequently broken — and the Price-to-Sales (P/S) ratio is the metric that actually lets you compare.

The P/S ratio is what seasoned growth investors reach for when the P/E ratio stops making sense. That happens more often than you would think — especially in software, biotech, and early-stage platforms. Once you know what "normal" looks like per industry, this one ratio cuts through a lot of noise.

What is the Price-to-Sales ratio?

It is the simplest valuation multiple in the book. You take a company's market capitalization and divide by its trailing-twelve-months revenue. Some investors use the per-share version (share price divided by revenue per share), which gives the same answer.

If NVIDIA (NVDA) has a market cap of roughly $3 trillion and trailing revenue of about $130 billion, its P/S is approximately 23x. That single number tells you the market is paying about $23 for every $1 of NVDA revenue.

Why does that matter? Because earnings — especially GAAP earnings — can be massaged, deferred, or wiped out by one-time charges. Revenue is harder to fake. For growth companies that reinvest everything they earn into R&D, sales, and infrastructure, the P/S ratio is often the cleanest valuation signal available.

When should you use P/S instead of P/E?

Use it in three situations.

First, when earnings are negative or near zero. A loss-making company has no meaningful P/E — the denominator is either negative (meaningless) or tiny (mathematically huge). PLTR at various points in its history, Uber in its early years, and most biotechs fall into this bucket.

Second, when earnings are volatile or distorted. Cyclical companies — think semiconductors, oil, or autos — swing from massive losses to massive profits across a cycle. P/E expands and contracts wildly. P/S smooths this out because revenue is less cyclical than margin.

Third, when companies are early in their monetization curve. A subscription platform with 90% gross margins but heavy sales-and-marketing spend will report low operating income by choice. P/S captures what the asset is worth at scale; P/E captures the ugly in-between.

How do you calculate a sensible P/S benchmark?

Compare three things: industry median, the company's own five-year average, and peer group.

All figures are rough, rounded, and meant as illustrations — not investment advice. Actual multiples change daily.

See how PLTR jumps off the page at roughly 79x? That is the P/S ratio doing its job. Without it, you might be tempted to call PLTR "expensive but fair" by some P/E comparison; with it, you see that the market is making a far more extreme bet on future revenue growth than anywhere else in the table.

Why do SaaS companies trade at huge P/S multiples?

Because software revenue is structurally different. A SaaS dollar of revenue comes with around 75–85% gross margin, minimal incremental cost to serve the next customer, and (ideally) a multi-year retention curve. A dollar of Walmart (WMT) revenue comes with approximately 25% gross margin and no recurring-revenue economics.

So paying 15x sales for a SaaS company that will eventually convert that revenue into 30% operating margins at scale is very different from paying 15x sales for a retailer that will never get there. The P/S multiple is a proxy for the eventual profit margin the market expects.

That is also why P/S multiples compress when gross margins compress. If AI infrastructure costs eat into SaaS gross margins over the next 3–5 years, the premium multiples tech software has enjoyed will have to mean-revert. Watch the gross margin first — the P/S follows.

What are the biggest mistakes with P/S?

Four, in descending order of damage:

1. Ignoring profitability entirely. A P/S of 2x looks cheap until you notice the company has negative gross margins. Some subscription box startups and loss-leader retailers run this playbook.
2. Cross-industry comparisons. A P/S of 5x is the bargain bin for cloud software and an absolute warning sign for a trucking company. Always compare to industry peers, not the S&P 500 average.
3. Confusing revenue with revenue quality. Ad-driven revenue (volatile) is not the same as subscription revenue (recurring). Transaction take rates (contracted) are different from one-time license sales. Adjust your P/S benchmark accordingly.
4. Forgetting net debt. P/S uses market cap, which ignores the balance sheet. A company with 2x net debt / equity looks better on P/S than one with net cash — but the equity holder absorbs the leverage risk.

The fix for mistakes 1 and 4 is to use EV/Revenue instead, which adjusts for both debt and cash. The fix for 2 and 3 is discipline: only compare apples to apples.

How do value investors actually use P/S?

They use it as a sanity check, not a primary screen. Kenneth Fisher popularized P/S in the 1980s via his book Super Stocks, arguing that a P/S below 0.75x for a non-cyclical company was a high-probability setup.

Modern value investors tend to weight P/S against ROIC, gross margin, and revenue growth. A cheap P/S combined with expanding gross margin is a powerful combination; a cheap P/S combined with collapsing gross margin is a value trap.

For deeper reading on how the legends think about multiples, see our super investors library and compare the frameworks.

When should you NOT use P/S?

Three situations — learn them before you get burned:

1. Financials. Banks and insurers do not have "revenue" in the conventional sense. Use P/BV (price-to-book) or P/TBV instead for JPMorgan (JPM) or Bank of America (BAC).
2. Mature, highly levered companies. A utility with consistent earnings and predictable dividends is better analyzed on free cash flow yield or EV/EBITDA.
3. Companies with significant non-revenue intrinsic value. Holding companies, conglomerates, and real-estate-heavy businesses need asset-based valuation — not revenue multiples.

Counter-argument: does a cheap P/S actually predict returns?

Academic research shows P/S has been a mid-tier factor. On its own, it beats market cap weighting but loses to quality-adjusted value factors over long samples. The 1996 James O'Shaughnessy book What Works on Wall Street tested this extensively and found P/S combined with relative strength outperformed.

The bearish counter: P/S is easier to game than earnings-based multiples. Aggressive channel stuffing, sweetheart customer deals, and unprofitable customer acquisition can all inflate revenue and mask the fact that every additional dollar of sales is destroying value.

The practical takeaway: P/S is a diagnostic tool, not a verdict. Combine it with gross margin, cash conversion, and revenue retention to triangulate whether a "cheap" P/S is genuinely cheap or a trap.

Ready to analyze these stocks yourself? Search any ticker on MainRatios to see valuations from 6 legendary investors - free.