Five Early Warning Signs a Moat Is Eroding

Competitive advantages don't announce their departure. The structural damage is usually done — months, sometimes years — before it shows up in an analyst's note or a consensus rating change. A wide-moat business becomes a narrow-moat one through a sequence of small concessions, each of which has a plausible innocent explanation in isolation. That's what makes moat erosion genuinely difficult to detect: the individual signals are ambiguous. Only the pattern condemns.

The traditional forensic approaches — looking for erosion in declining return on invested capital or in shrinking market share — are useful but late. ROIC declines show up in annual filings, which already lag the underlying competitive reality by definition. Market share data is often quarterly, industry-reported, and backward-looking by construction. By the time those metrics move, the structural position has already shifted. An investor acting on confirmed ROIC decline has typically arrived somewhere between eighteen and thirty-six months after the moat began to erode.

What follows are five leading indicators that consistently appear ahead of acknowledged moat deterioration. They surface in earnings transcripts, financial footnotes, management behavior, and attrition patterns — not in the headline numbers. None of them is dispositive on its own. One or two in a single quarter is noise. Three or more, sustained across reporting periods, is something else entirely.

1. Margin Compression That Doesn't Bounce Back

Margin compression is the most quoted moat-erosion signal and the most frequently misread. A single-year decline in gross or operating margins can mean almost anything: raw material inflation, a product transition year, an investment cycle, expansion into a new geography with front-loaded costs. These explanations are often legitimate. The structural tell is different — it's margin compression that persists across three or more consecutive years, continues even as revenue grows, and cannot be traced to a reversible cost driver.

The distinction matters enormously. A business absorbing temporary cost pressure will show contraction followed by recovery as the pressure passes. A business losing competitive position shows contraction that flattens but never rebounds — because the recovery requires pricing power the moat once provided but no longer can.

Intel's trajectory through fiscal 2021 to fiscal 2023 illustrates the pattern precisely. In their fiscal 2021 10-K, Intel reported gross margins of approximately 55.4%. By their fiscal 2023 annual filing, that figure had fallen to around 40.2% — not because a commodity cycle turned against them, but because TSMC's manufacturing advantages and AMD's competitive product cadence had structurally eroded Intel's ability to command premium pricing on its core architectures. Not a cyclical problem. A structural one. And the decline was visible across three consecutive 10-Ks before competitive dynamics became the dominant narrative in the investment press.

Kodak told the same story a decade earlier. Gross margins that had run above 40% in 2000 declined steadily through the mid-2000s as digital photography commoditized the film business. Each annual report offered a new explanation — digital transition costs, restructuring charges, raw material headwinds. The pattern across years was the signal.

The practical screen: when gross margins have declined in each of the past three annual filings — or operating margins have fallen cumulatively by more than three percentage points — without a clear reversible cause, the business deserves a reassessment of its competitive position, not just its cost structure. A moat with genuine structural depth should eventually convert revenue growth back into margin expansion. Persistent compression without recovery is one of the cleaner tells that it can't.

2. Pricing Concessions to Keep the Business

A business with genuine pricing power raises prices without losing volume. That's the operational definition. When management begins describing deals closed at 'competitive pricing,' 'flexible terms,' or 'customer-friendly structures,' they're describing the absence of pricing power without quite naming it.

Blockbuster's elimination of late fees in January 2005 is the canonical example — a pricing concession so explicit and consequential that it removed approximately $400 million in annual revenue. Management at the time described the decision as a customer experience improvement. What it actually was: a direct capitulation to Netflix's no-late-fee subscription model. The ability to charge for a service Blockbuster had monetized since their founding had evaporated because a competitor had demonstrated that customers would switch for the alternative. By the time the late fees disappeared, the durable competitive advantage was already gone. The concession was its public acknowledgment.

For software and enterprise service businesses, pricing concessions are harder to read but equally diagnostic. Watch for gross margin declining while revenue grows — this pattern, sustained over two or three years, often means that revenue growth is being purchased through discounting rather than earned through product leadership. Watch for contract duration shortening in subscription businesses. Watch for 'retention-focused pricing' and 'win-back programs' appearing in investor presentations. Each is a polite variant of the same fact: the switching costs are diminishing and the pricing power is leaving with them.

3. Customer Concentration Shifting Toward Fewer, Larger Buyers

High customer concentration is a recognized risk factor. Less discussed — and more diagnostic — is the trend in concentration over time. A business whose top-five customers represented 20% of revenue five years ago but now represent 35% has experienced a meaningful shift in bargaining power. And that shift almost always advantages the buyer, not the seller.

When revenue concentrates toward fewer, larger customers, those customers gain negotiating leverage. The first sign is often quiet: pricing holds but terms loosen. Payment periods extend. Service level agreements become more demanding. The customer asks for exclusivities, volume discounts, or integration commitments that individually seem manageable — but collectively signal that the relationship balance has reversed.

Nokia's dependence on major carrier contracts sharpened in the years before the iPhone changed the handset market. As Nokia's smartphone product differentiation weakened against early Android devices, carrier relationships began shifting. Carriers reduced minimum order commitments, began qualifying multiple handset suppliers for previously exclusive contract terms, and negotiated average selling prices more aggressively than they had when Nokia's market position was unambiguous. Nokia's revenue remained large. The concentration metrics in their filings didn't signal emergency. But the direction was clear — and it preceded the broadly acknowledged market share decline by roughly two years.

Check the 10-K's Risk Factors section for changes in customer concentration disclosures year over year. Watch for the appearance — or geographic expansion — of language describing revenue dependence on a small number of customers that wasn't present in prior filings. That disclosure isn't boilerplate. It's management communicating, accurately and without fanfare, where the negotiating leverage now lives.

4. R&D Spending Accelerating Without Revenue Acceleration

Rising research and development investment is routinely framed as a sign of confidence in the future. Sometimes it is. But there's a specific pattern worth separating out: R&D-to-revenue ratios that increase over three or more consecutive years without a commensurate acceleration in revenue growth or in commercially successful product output. This pattern typically doesn't indicate strategic confidence. It indicates structural defensiveness.

A business with a genuine competitive advantage generates its structural position at relatively stable R&D intensity, because the advantage doesn't require constant reinvention to sustain itself. When R&D intensity accelerates without corresponding commercial output — particularly in a business that once had stable competitive positioning — it often means the existing advantages are deteriorating faster than new ones can be built. The company is running to stand still, spending more to defend what the moat once maintained automatically.

Nokia's R&D-to-revenue ratio increased from approximately 9% in fiscal 2007 to nearly 15% by fiscal 2011, per their annual reports during that period. This is not how a business with a widening competitive advantage deploys capital. A company whose structural position was expanding would be converting more revenue into earnings, not consuming more of it in defense. Nokia was spending more and losing faster — a pattern visible in the financials two to three years before it was broadly described as a strategic failure.

This dynamic is worth watching carefully in any business whose competitive position depends on proprietary data, workflow integration, or accumulated customer knowledge — areas where well-resourced competitors are now applying AI-powered alternatives. A company that responds by accelerating R&D intensity without demonstrating commercial output from that investment is displaying the signal, not defusing it. Our post on evaluating R&D spending covers the specific metrics that distinguish forward investment from defensive consumption.

5. Key-Talent Exits From the Moat-Building Functions

This is the most difficult signal to quantify and the most forward-looking of the five. Key-talent exits don't appear in any financial filing — they surface in LinkedIn updates, in conference speaker rosters, in the footnotes of competitors' hiring announcements. But they're worth systematic attention, particularly in businesses where the competitive advantage lives in human capital, proprietary process knowledge, or relationship depth.

I'll be direct about the limits here: the data is genuinely thin, and the interpretation is more subjective than I'd like. I've seen businesses lose important people and recover fully within eighteen months, because the institutional knowledge turned out to be more distributed than the individual departures suggested. This signal only becomes diagnostic — rather than merely concerning — when departures cluster in the specific functions that sustain the competitive advantage, and when the destinations are direct competitors or well-funded startups targeting the same customers.

Kodak's digital imaging division experienced sustained talent departures through the late 1990s and early 2000s, as engineers who had built early digital camera prototypes moved toward consumer electronics companies and startups with more committed digital roadmaps. These weren't departures from accounting or procurement. They were from the functions where Kodak's future competitive position would be decided — and the institutional knowledge that left was precisely the institutional knowledge that mattered.

For asset-light businesses where the competitive advantage is embedded in human relationships and expertise rather than in physical capital, this signal is structurally more significant. The moat can walk out the door. A data science lead departing an intangible asset-heavy business, a key account manager leaving a switching-cost-dependent software company, a principal engineer exiting a platform building its next-generation infrastructure — these departures are more diagnostic than a CFO transition. And unlike every other signal on this list, there's no annual filing to catch you up. By the time attrition reaches a proxy statement's turnover disclosures, the institutional knowledge has already migrated.

Key Takeaways

• Margin compression sustained across three or more years — especially compression that persists as revenue grows — is structural, not cyclical. Intel's gross margin declined from approximately 55% in fiscal year 2021 to approximately 40% by fiscal year 2023, visible in consecutive annual filings before the competitive dynamics became the center of mainstream discussion.
• Pricing concessions are often the clearest sign: Blockbuster eliminated late fees in January 2005, removing approximately $400 million in annual revenue as a direct response to Netflix's no-late-fee model. Watch for gross margin declining while revenue is maintained — that combination usually means discounting, not product leadership.
• Customer concentration trends matter more than concentration levels. When top customers grow as a share of revenue over multiple years, bargaining power is shifting toward the buyer — and that shift typically precedes margin compression by 12 to 18 months. Annual 10-K Risk Factors disclosures often reveal this trend before it's visible in the headline financials.
• Rising R&D intensity without commercial acceleration indicates defensive spending, not forward confidence. Nokia's R&D-to-revenue ratio expanded from approximately 9% in fiscal 2007 to approximately 15% by fiscal 2011 as its smartphone competitive position deteriorated — the capital was chasing a moat that was already leaving.
• Key-talent exits from the moat-building functions — not general attrition but departures from the people who directly sustain the competitive advantage — are the most forward-looking signal and the most difficult to quantify. The pattern only becomes diagnostic when departures cluster in the relevant functions and the destinations are competitive threats.