Analyzing a Company's Competitive Moat Using Only Public Filings

Warren Buffett popularized the term, and Morningstar later turned it into a formal research framework, but the idea behind a competitive moat is simple: some businesses keep earning high returns for decades because something structural stops competitors from taking those returns away. The hard part is telling a real moat from a well-written story about one. Every investor deck claims a durable competitive advantage, and most of them are describing a head start that will erode within a few years.

You don't need expert networks, channel checks, or a paid data terminal to test these claims. The 10-K, the proxy statement, and a few years of financial statements, all free on EDGAR, contain most of the evidence. Here's the process I use: what each moat type looks like in the numbers, where the qualitative evidence hides in the filings, and how to check that the two agree.

The Five Moat Types and Their Fingerprints

Morningstar's taxonomy is the standard one, and I like it because each moat type leaves a different fingerprint in the financial data. If a company claims one of these, you can go look for the corresponding trace.

Cost advantages. The company can produce or deliver at lower cost than anyone else in its market. Walmart is the classic case. In the filings, this shows up as gross or operating margins that sit above peer levels year after year even while the company prices competitively. Check cost of revenue as a percentage of sales against two or three direct competitors.

Switching costs. Customers would face real cost, hassle, or risk if they moved to a competitor. Enterprise software is the textbook example. Look for disclosed retention rates, long average customer relationships, and steadily building deferred revenue. When churn isn't disclosed directly, deferred revenue trends give you a rough read on it.

Network effects. The product gets more valuable as more people use it. Payment networks and marketplaces are the usual examples, and I'll vouch for this one from the inside, having spent years at American Express, which runs a closed-loop network connecting card members and merchants. In filings, network effects show up as user growth that keeps compounding while sales and marketing spend shrinks as a percentage of revenue.

Intangible assets. Patents, brands, and regulatory licenses that block competition outright or make it irrational. Pharma companies with patent protection and consumer brands with pricing power are the standard cases. The fingerprint is premium pricing sustained over many years, visible as high gross margins that refuse to compress.

Efficient scale. The market only supports a small number of players, so entering it would wreck returns for everyone, and rational competitors stay out. Pipelines, railroads, and cell towers fit here. Look for stable market share in an industry where nobody has added meaningful new capacity in a long time.

Where the Evidence Hides in the 10-K

Three sections of the 10-K do most of the work.

Item 1, the business description. Read it for specificity. A company that says it has a large and loyal customer base is telling you nothing. A company that says its average customer relationship runs more than a decade, with retention above ninety percent, is making a claim you can track from one filing to the next. Real moats tend to get described in concrete, checkable terms, because management knows they are the core of the story.

Item 1A, the risk factors. This section is more revealing than it looks, because companies have to disclose what could genuinely hurt them. When patent expiration sits near the top of the risk list, patents are the moat. When the company spends two paragraphs on customer retention risk, switching costs are what keeps the model working. I often read 1A before Item 1, since it tells you what management is actually worried about losing.

The competition discussion. Most 10-Ks describe the competitive landscape somewhere in Item 1. Count how many competitors get named and note how the company says it differentiates. Narrow-moat companies tend to list a crowd of rivals competing on price and features. Wide-moat companies often name only a few, or describe a combination of capabilities that no single competitor matches.

Gross Margin Stability as a Pricing Power Test

Gross margin is the cleanest financial read on pricing power. A company that holds or expands its gross margin through inflation cycles, new entrants, and demand swings is demonstrating, every quarter, that customers will pay its price and competitors can't profitably undercut it.

Pull at least five years of gross margin data, ten if you can get it. Stability matters more than the level. Say one company swings between 40 and 60 percent depending on the year while another holds steady between 62 and 65. The second business has far more pricing power even though the averages look similar, because the swings in the first tell you the market sets its prices, and the steadiness of the second tells you the company does.

Then compare against the industry. A company running well above its industry's typical gross margin has a cost advantage, a pricing advantage, or both, and if the gap has persisted for years the advantage is probably structural. If the gap opened up recently, stay skeptical, since supply shortages and product cycles can produce the same picture temporarily.

ROIC, the Closest Thing to a Single Moat Metric

If I had to test a moat with one number, it would be return on invested capital. Basic economics says high returns attract competition, and competition pushes returns back down toward the cost of capital. So when a company keeps earning well above its cost of capital year after year, something is blocking that process, and the blockage is the moat.

The standard calculation is net operating profit after tax divided by invested capital. Say a company reports 800 million dollars of operating income and pays around a 21 percent effective tax rate. NOPAT comes out to roughly 630 million. If invested capital, meaning total assets minus non-interest-bearing current liabilities, is 4 billion, ROIC lands just under 16 percent, comfortably above the cost of capital for most businesses.

One good year proves very little. Plenty of companies post a few strong years after a product launch or during an industry upcycle. What you want is persistence, meaning returns above the cost of capital across five to ten years, ideally through at least one downturn. The longer high returns hold, the more confident you can be that they come from structure rather than timing.

R&D Spending: Level Versus Efficiency

R&D relative to revenue tells you whether a company is investing to defend its position, but the raw ratio is easy to misread. Spending more doesn't automatically build more moat; what matters is what the spending produces.

Compare R&D-to-revenue ratios across direct competitors, then look at the output. Say one company spends 15 percent of revenue on R&D and ships a steady stream of successful products while a rival spends 20 percent and keeps falling behind. The first has an innovation engine, and the second has an expensive habit. Patent grants, product launch cadence, and revenue from recently introduced products (some companies disclose this in the 10-K) are all usable proxies for R&D productivity.

Also watch the trend. A company cutting R&D as a percentage of revenue may be harvesting its position instead of defending it. That flatters earnings for a few years while the moat quietly narrows, and by the time the erosion shows up in revenue it's expensive to reverse.

Customer Concentration

SEC rules require companies to disclose any customer that accounts for 10 percent or more of revenue. You'll find this in Item 1 or in the revenue footnotes, and it's worth reading every time, because concentration usually undermines a moat. A company with two customers at a quarter of revenue each has limited pricing power with either of them, no matter how sticky the product is, and losing one would be an existential event.

The profile you want is the reverse, thousands of customers with none above a few percent of revenue. Diversification means no single relationship can force concessions, and it means the retention and pricing numbers in the filing describe a broad base rather than a couple of make-or-break accounts.

The Qualitative Tells

Beyond the numbers, a handful of narrative details in filings correlate strongly with durable moats. I scan for these:

• Long-term contracts. Multi-year customer agreements point to switching costs and revenue visibility. Contract lengths and renewal rates, where disclosed, tell you how durable that is.
• Hard-won licenses and approvals. Regulatory clearances that took years to obtain are barriers a new entrant would have to climb from a standing start.
• Proprietary data. Data sets built over many years compound in value, and a newcomer can't buy its way to parity quickly.
• Deep integration. When the product is embedded in customers' workflows or technology stacks, switching costs stay high even if the product itself could be replicated.

The proxy statement is worth a pass here too. If executive compensation is tied to retention, returns on capital, or relative market share, management is being paid to defend the specific advantage the 10-K describes, which is a better sign than bonuses tied purely to short-term revenue growth.

None of these guarantees a moat on its own, but each one is a concrete claim you can test against the financial evidence from the earlier sections.

Putting It Together

A full assessment pairs the quantitative record (gross margin stability, ROIC persistence, customer concentration, R&D productivity) with the qualitative story in Item 1 and Item 1A (barriers to entry, switching costs, network dynamics, licenses). Neither side is enough on its own, and the failure modes differ. Strong financials can come from a temporarily favorable position that simply hasn't been attacked yet, while a compelling competitive narrative that never shows up in returns on capital is marketing, however well written.

What convinces me is agreement between the two. When management describes switching costs and deferred revenue keeps building, or the risk factors fret about patent cliffs while gross margins have held for a decade, the numbers and the story are pointing at the same underlying structure. In practice, start with five years of ROIC and gross margins, form a view on whether the returns look defended, then read Item 1 and 1A to figure out what is doing the defending. One careful afternoon on EDGAR will get you further than most of the commentary you'll find about the same company.