This week's earnings scan wasn't about who surprised—it was about who keeps doing it. Running a broad pull through the FMP Earnings Surprises Bulk API surfaced a short list of companies where beats are no longer episodic but sequential. In a tape where leadership is narrowing and estimates are being revised more defensively, persistence itself is starting to read as a signal.

Across the dataset, five names stood out for extending earnings-beat streaks well beyond what short-term positioning or one-off operational wins would explain. This article breaks down how those streaks were identified using the FMP Earnings Surprises Bulk API, why repeatability matters more than magnitude in the current environment, and how to think about earnings consistency as a screening input rather than a post-hoc narrative.

Five Companies With Long Earnings Beat Streaks

NVDA NVIDIA Corporation

Beat Streak: 12 quarters.
Next quarterly report: Feb. 25 — EPS: $1.52; Revenue: $65.55B (consensus).

Twelve straight earnings beats at NVIDIA now sit less in the realm of surprise and more in the category of structural execution. At this point in the cycle, the signal is not that demand for accelerated computing exists—that is well understood—but that consensus continues to underestimate the speed and scale at which NVIDIA converts that demand into reported results. The streak reflects a persistent gap between external modeling and internal visibility, particularly around data center revenue mix, platform attach rates, and customer deployment timelines.

What makes the pattern notable is that it has held through multiple estimate-reset phases, including periods when expectations were aggressively revised upward. That persistence suggests analysts may still be struggling to model second-order effects—such as networking, software, and system-level revenue—rather than just headline GPU shipments. To contextualize whether the streak remains informative, datasets like segment-level revenue disclosures from the income statement, combined with backlog or deferred revenue trends, are more useful than headline growth rates alone. Monitoring estimate dispersion and revision velocity alongside those fundamentals helps frame whether this is still a forecasting issue or simply expectations converging.

META Meta Platforms, Inc.

Beat Streak: 12 quarters.
Next quarterly report: April 29 — EPS: $6.71; Revenue: $54.36B (consensus).

Meta's twelve-quarter beat streak reflects a different dynamic: not explosive top-line surprises, but steady operating discipline layered on top of a resilient advertising base. The signal here is consistency in margin delivery rather than revenue volatility. Even as engagement trends, ad load, and pricing dynamics fluctuate across platforms, Meta has repeatedly reported results that land above expectations built on conservative cost assumptions.

This pattern highlights the importance of tracking operating leverage rather than focusing exclusively on ad growth narratives. Analysts appear to systematically underestimate the company's ability to flex expenses—particularly in infrastructure, headcount, and content spend—relative to revenue trajectories. To assess whether this streak remains meaningful, operating margin trends from the income statement, paired with capex disclosures and headcount data, provide more insight than user metrics alone. Analyst estimate ranges and cost-side revisions are especially relevant inputs to watch as expectations attempt to normalize.

GH Guardant Health, Inc.

Beat Streak: 7 quarters.
Next quarterly report: Feb. 19 — EPS: -$0.42; Revenue: $271.2M (consensus).

Guardant Health's seven-quarter beat streak stands out because it is occurring despite continued losses. The signal here is not profitability, but consistency in execution relative to expectations in a development-heavy diagnostic business. Repeated beats in a negative-EPS context often point to revenue traction, expense control, or timing advantages that are not fully captured in consensus models.

In Guardant's case, the pattern suggests analysts may be lagging adoption curves across oncology testing and screening products, while also misjudging the cadence of operating expense growth. Because absolute profitability is not yet the anchor, the more relevant datasets are revenue growth by test category, gross margin progression, and cash flow burn from the cash flow statement. Pairing those with insider transaction data and R&D expense trends can help determine whether the streak reflects sustainable operational progress or simply conservative forecasting in an inherently uncertain space.

WSM Williams-Sonoma, Inc.

Beat Streak: 6 quarters.
Next quarterly report: March 18 — EPS: $2.88; Revenue: $2.41B (consensus).

Williams-Sonoma's six-quarter beat streak is notable given the broader context of uneven discretionary spending and normalization across home-related categories. Rather than benefiting from a cyclical tailwind, the company has continued to outperform expectations through margin resilience, inventory discipline, and brand-level execution across its portfolio.

The repeatability here suggests that consensus estimates may be anchoring too heavily to sector-wide assumptions instead of company-specific operating behavior. Gross margin stability and free cash flow generation have been recurring differentiators, even as revenue growth moderates. To evaluate whether this pattern holds informational value, investors would be better served examining inventory turns, promotional intensity, and brand-level revenue disclosures alongside standard income statement data. These inputs provide clearer signals than macro consumption narratives alone.

HST Host Hotels & Resorts, Inc.

Beat Streak: 4 quarters.
Next quarterly report: Feb. 4 — EPS: $0.47; Revenue: $1.48B (consensus).

Host Hotels & Resorts' four-quarter beat streak reflects execution within a rate-sensitive, capital-intensive segment where expectations often lag real-time operating data. The signal here is not accelerating growth, but steadier-than-expected performance across occupancy, average daily rates, and cost management in a volatile travel environment.

Because lodging REIT results are heavily influenced by operating leverage, small deviations in revenue per available room (RevPAR) or expense ratios can materially affect earnings outcomes. The persistence of beats suggests consensus may be slow to adjust for property-level performance and regional mix. To contextualize this streak, supplemental operating metrics, property-level disclosures, and balance sheet leverage data are more informative than EPS alone. Tracking how these fundamentals evolve alongside analyst estimate revisions helps clarify whether the pattern is narrowing—or remains analytically relevant.

Interpreting What Repeatable Beats Are Actually Telling Us

Taken together, these five companies illustrate that repeatable earnings beats are rarely about surprise in the traditional sense. Across semiconductors, digital advertising, diagnostics, consumer discretionary, and lodging, the common thread is not sector momentum but forecasting friction. In each case, consensus estimates appear to lag underlying operating realities—sometimes because demand is hard to model, sometimes because cost structures are more flexible than assumed, and sometimes because sector-level narratives overwhelm company-specific data. The signal is not uniform optimism; it is persistence in analytical mismatch.

What separates a meaningful streak from statistical noise is where that gap originates. For NVIDIA and Meta, the disconnect has tended to live inside operating leverage, cost discipline, and secondary revenue streams rather than top-line visibility alone. For Guardant Health, it has shown up in revenue durability and expense cadence despite ongoing losses. Williams-Sonoma and Host Hotels highlight another dimension entirely: consensus anchoring to macro assumptions that fail to capture execution at the brand or property level. In each case, the streak itself becomes evidence that the market's modeling framework—not just its inputs—may be incomplete.

This is where a multi-endpoint analytical view becomes essential. An earnings surprise, by itself, only confirms that expectations were wrong; it does not explain why. Pairing the Earnings Surprises dataset with income statement data helps isolate whether beats are being driven by margin expansion, mix shifts, or revenue timing—but that still leaves room for misinterpretation. Cross-checking repeat beats against operating and free cash flow behavior, as outlined in this discussion of detecting earnings quality erosion through cash flow analysis, adds a critical layer of validation. It helps distinguish durable execution from outcomes shaped primarily by accounting or timing effects.

Layering in analyst estimate revisions and price targets further clarifies how quickly expectations are adapting, while balance sheet data frames whether consistency is being supported by leverage, working capital management, or capital discipline. When those datasets are evaluated together—rather than in isolation—platforms like Financial Modeling Prep allow repeatable beats to function as a diagnostic tool instead of a headline metric.

The broader takeaway is that repeatable beats are less about direction and more about process. They surface areas where consensus models are slow to converge on reality. The analytical edge comes not from observing a streak once, but from tracking whether subsequent data shows that gap narrowing—or continuing to persist—as new information is absorbed.

Building a Repeatability Screen with FMP Data

If the goal is to surface companies that reliably beat expectations—not just those that happen to post an occasional upside—the construction of the screen matters as much as the results it produces. The first principle is to avoid narrowing the universe too early. Starting from a predefined watchlist introduces bias before the data has a chance to speak. A more neutral approach is to begin with the full earnings record, capture every reported outcome, and allow patterns of repetition to emerge on their own. FMP's Earnings Surprises Bulk API is well suited for that role, as it provides a consistent, quarter-by-quarter record of EPS results across a broad set of companies.

As with any automated pull, the first step is simply confirming your API key is active and ready.

1. Pull Bulk Earnings Surprises

Begin by hitting the Earnings Surprises Bulk API, which aggregates every quarterly EPS surprise — positive or negative — for the year you specify:

https://financialmodelingprep.com/stable/earnings-surprises-bulk?year=2025&apikey=YOUR_API_KEY

Sample Response:

[

{

"symbol": "AMKYF",

"date": "2025-07-09",

"epsActual": 0.3631,

"epsEstimated": 0.3615,

"lastUpdated": "2025-07-09"

}

]

From here, the first cut is mechanical: isolate the entries where epsActual > epsEstimated. That gives you the universe of names that beat expectations at least once during the period — essentially a raw pool before you evaluate whether any of them can deliver that result consistently.

2. Retrieve Company-Level Details

With that universe in hand, the analysis moves from identifying events to evaluating consistency. For each ticker that cleared the first filter, pull its complete quarterly earnings history using the Earnings Report API:

https://financialmodelingprep.com/stable/earnings?symbol=AAPL&apikey=YOUR_API_KEY

Looking at the complete sequence of reported quarters makes it possible to evaluate frequency and clustering. This is where judgment enters the workflow. Some analysts require three or more consecutive beats to qualify as a streak; others impose minimum surprise thresholds or remove near-zero deviations. The parameters can be adjusted, but the intent stays the same: separate sustained execution from statistical noise.

By this point, the screen has moved beyond identifying isolated surprises. What began as a broad event scan turns into a structured assessment of earnings reliability, highlighting companies where internal forecasting and operational control have proven more consistent than the market's expectations over time.

Broadening the Universe as Coverage Scales

A repeatability screen works best when its scope is expanded deliberately rather than all at once. The natural starting point is a segment of the market where estimate behavior is familiar and well constrained. Under the Free plan, coverage is centered on widely followed companies such as AAPL, GOOGL, and JPM, where analyst models tend to cluster closely and revisions are more incremental. That environment makes it easier to pressure-test streak definitions and confirm that the screening logic behaves as intended before introducing additional complexity.

Once the framework proves stable there, moving into the Starter plan changes the character of the data. Broader U.S. equity coverage brings in smaller-cap names and niche industries, where analyst coverage thins out and estimate ranges widen. That dispersion is not a drawback—it becomes a useful stress test, helping distinguish earnings consistency driven by execution from consistency that merely reflects loose or uneven forecasting.

Expanding further through the Premium plan extends the same approach to international markets, including U.K. and Canadian listings. The underlying methodology remains intact, but the operating context shifts. Variations in reporting standards, business mix, and margin structures introduce new interpretive considerations without requiring the screen itself to be rebuilt.

At every stage, the discipline is the same: scale only after the prior layer holds. By sequencing expansion this way, the signal remains interpretable as coverage grows, and earnings repeatability continues to function as a practical analytical input rather than being diluted by scale alone.

From Individual Workflow to Firmwide Analytical Standard

When a screening process demonstrates that it holds up under repeated use, its value inside a firm naturally shifts. What starts as a desk-level solution becomes a candidate for institutional adoption—less about individual efficiency and more about establishing a shared analytical reference point. At that stage, the priority is no longer how quickly one analyst can run the screen, but whether the same earnings patterns are being identified, filtered, and interpreted consistently across coverage teams.

That evolution is typically driven by analysts closest to the work. As a workflow becomes embedded in active coverage, those users are the ones who codify definitions, resolve gray areas, and make assumptions explicit. In practice, that effort replaces fragmented spreadsheets and slightly different logic across sectors with a common framework—one that can be reviewed, challenged, and improved without being rebuilt from scratch each time.

The benefits of standardization surface quickly. Shared dashboards supplant one-off models, methodological changes become visible and auditable, and governance improves because inputs and rules are clearly defined. Just as importantly, conversations across teams move away from reconciling numbers and toward interpreting what the data implies. The result is less duplication, fewer inconsistencies, and more time spent on analysis rather than maintenance.

Once a workflow reaches that point, scaling it through a platform-level setup—such as FMP's Enterprise plan—becomes a matter of durability rather than expansion. It allows a proven desk-level process to support firm-wide usage, auditability, and continuity without compromising the underlying methodology, effectively turning an individual solution into shared research infrastructure.

Keeping Earnings Consistency in Motion

Earnings consistency only matters if it's re-tested as expectations evolve. Used properly, tools like the FMP Earnings Surprises Bulk API keep that evaluation active—allowing repeatable beats to be monitored as estimates reset and operating conditions change. The real work isn't identifying a streak once, but staying attentive to whether it continues to reflect something the market hasn't fully priced in yet.

Want more? Explore our earlier article: Weekly Signals Desk | Concentrated Analyst Revisions via the FMP API (Week of Jan 19-23)

Disclosure: Signals Desk content is provided for informational and analytical purposes only and does not constitute investment advice or trade recommendations. The analysis reflects interpretation of market data and publicly disclosed or third-party information, including data accessed via Financial Modeling Prep APIs, at the time of publication. Signals discussed are probabilistic, can be wrong, and may change as market conditions and consensus data evolve. This content should be considered alongside broader research, individual objectives, and risk assessment.