This week's scan of earnings data turned up a familiar but often underweighted signal: consistency. Not a one-off upside surprise, but repeat outperformance across multiple reporting cycles—spanning software, semis, distribution, and industrials—at a time when estimate revisions remain uneven.
Using the FMP Earnings Surprises Bulk API, we screened for companies where execution has persistently cleared the bar set by consensus. The result is a small group of names where expectations have lagged reality for quarters at a time.
In this note, we break down those companies and walk through the exact API-driven process used to surface them—focusing on how to systematically identify earnings beat streaks rather than isolated surprises.
Five Companies With Long Earnings Beat Streaks
Monolithic Power Systems, Inc. (MPWR)
Beat Streak: 26 quarters.
Next quarterly report: May 7 — EPS: $4.89; Revenue: $781M (consensus).
A 26-quarter streak places Monolithic Power Systems in a different category—this is not cyclical variance but a sustained pattern across multiple semiconductor environments. The company operates in analog and power management, areas where design wins tend to translate into long product lifecycles and recurring revenue streams. That stability, combined with exposure to data centers, automotive, and industrial applications, has historically supported predictable execution even as broader semiconductor demand fluctuates.
What stands out is how the company has navigated multiple demand regimes—including inventory corrections and end-market softness—while still exceeding expectations. That suggests internal forecasting is calibrated conservatively relative to actual order visibility, particularly in segments tied to electrification and AI infrastructure. Public filings and recent industry coverage have highlighted continued demand tied to power efficiency in data centers, which aligns with the company's core product positioning.
To evaluate whether the streak reflects structural advantages or timing effects, investors would typically look at gross margin trends and end-market revenue breakdowns within the income statement, alongside backlog indicators where available. Comparing those figures with analyst estimate dispersion can reveal whether consensus is systematically underpricing the durability of demand in power management applications.
Fabrinet (FN)
Beat Streak: 15 quarters.
Next quarterly report: May 4 — EPS: $3.56; Revenue: $1.18B (consensus).
Fabrinet's 15-quarter streak reflects its positioning within the optical and advanced manufacturing supply chain, particularly for communications and data center infrastructure. The company operates as a contract manufacturer for complex optical components, which ties its performance to long-cycle demand trends in networking and cloud infrastructure rather than short-term consumer cycles.
The consistency of beats suggests that demand visibility from key customers—often large networking and telecom equipment providers—has been stronger than what consensus models fully incorporate. Recent industry developments around data center expansion and AI-driven networking upgrades have reinforced the importance of optical interconnects, indirectly supporting Fabrinet's revenue base. These are multi-quarter investment cycles, which can lead to gradual upward revisions rather than abrupt estimate resets.
To assess the signal more precisely, segment-level revenue and customer concentration data are particularly relevant, along with backlog or order flow indicators where disclosed. Pairing that with analyst estimate revisions can show whether consensus is reacting to demand trends with a lag, which is often where repeat earnings beats originate.
Microsoft Corporation (MSFT)
Beat Streak: 14 quarters.
Next quarterly report: April 29 — EPS: $4.04; Revenue: $81.29B (consensus).
A 14-quarter beat streak at Microsoft is less about isolated upside and more about the durability of its forecasting ecosystem. At this scale, consistently exceeding consensus implies not only operational execution but also a pattern where internal visibility—particularly across cloud, enterprise software, and AI-linked demand—remains tighter than what external models capture. The signal here is subtle: expectations are not failing outright, but they are adjusting incrementally behind realized performance.
Recent cycles have reinforced how Azure growth, AI infrastructure spending, and enterprise contract renewals interact to produce earnings outcomes that remain slightly ahead of modeled assumptions. Public disclosures and earnings call commentary have pointed to sustained enterprise demand tied to cloud migration and AI workloads, which tend to be multi-quarter in nature rather than transactional. That dynamic reduces volatility in forward estimates but does not eliminate underestimation, especially when pricing and mix shift toward higher-margin services.
To contextualize the streak, the most relevant datasets would be segment-level revenue from the income statement and forward analyst estimate revisions. Tracking how Azure growth rates and operating margins evolve relative to consensus can help determine whether the beat pattern reflects structural underestimation or simply conservative guidance practices. The persistence of small but consistent positive deviations is often where the signal resides.
United Natural Foods, Inc. (UNFI)
Beat Streak: 7 quarters.
Next quarterly report: June 9 — EPS: $0.772; Revenue: $7.79B (consensus).
In contrast to technology and semiconductors, UNFI's seven-quarter streak emerges from a low-margin distribution business where execution is often measured in basis points. Beating expectations consistently in this context points to operational discipline—inventory management, cost control, and logistics efficiency—rather than top-line acceleration alone. The signal here is less about growth and more about margin resilience in a structurally tight industry.
Recent reporting periods have shown how supply chain normalization and cost initiatives can translate into incremental earnings upside, even when revenue growth remains modest. In a sector where consensus estimates tend to cluster tightly due to stable demand profiles, small deviations in operating margin can drive outsized earnings surprises. That dynamic increases the likelihood of repeated beats when internal efficiencies compound over time.
To understand the sustainability of this pattern, the most informative datasets would include operating margin trends and cost of goods sold from the income statement, as well as working capital metrics such as inventory turnover. These indicators help clarify whether the beat streak is driven by temporary cost tailwinds or a more embedded shift in operational performance.
Garrett Motion Inc. (GTX)
Beat Streak: 5 quarters.
Next quarterly report: May 7 — EPS: $0.41; Revenue: $915M (consensus).
Garrett Motion's five-quarter streak stands out given the cyclical and transitional nature of the automotive sector. The company's core business—turbocharging and performance technologies—sits at the intersection of internal combustion engine efficiency and broader electrification trends. Delivering consistent beats in this environment suggests effective cost management and stable demand within its core product lines, even as the industry navigates structural change.
Recent disclosures have pointed to steady aftermarket demand and disciplined capital allocation, which can help offset variability in original equipment manufacturer (OEM) production volumes. In sectors undergoing transition, consensus estimates often reflect broader uncertainty, which can create conditions where companies with stable execution exceed expectations more frequently than anticipated.
To evaluate whether this pattern reflects underlying resilience or timing effects, key datasets would include segment revenue (OEM vs. aftermarket), operating margins, and cash flow metrics from the income statement and cash flow statement. Monitoring these alongside analyst estimate dispersion can help determine whether the beat streak is driven by conservative expectations or genuinely improving operating fundamentals.
What Persistent Outperformance Signals Beneath the Surface
Across these five names, the common thread is not sector or size—it's the persistent gap between execution and expectations. That gap holds even in heavily modeled companies like Microsoft and extends into less-covered segments, suggesting consensus often adjusts incrementally rather than in step with underlying performance.
What differentiates the signal is not the outcome, but the source. In infrastructure-linked names like Monolithic Power Systems and Fabrinet, visibility comes from long-cycle demand and embedded positioning. In contrast, UNFI and Garrett Motion are delivering through cost control and operational precision. The result—repeated EPS beats—looks uniform, but the drivers are not, which is why streaks tend to carry more weight than isolated surprises.
Interpreting that pattern requires stacking datasets rather than relying on a single lens. Earnings surprises establish the signal, but pairing them with margin trends, estimate revisions, and cash flow data helps separate structural execution from short-term variance—an approach that aligns with how multi-endpoint datasets are typically structured within Financial Modeling Prep platform.
Viewed this way, persistent outperformance reflects less about upside events and more about how slowly expectations recalibrate. When that lag repeats across cycles and sectors, it begins to look less like noise and more like a consistent feature worth tracking.
Building a Repeatability Screen with FMP Data
Screening for consistent earnings outperformance works best when you start from the entire dataset rather than a curated list of names. Pre-selecting companies introduces bias upfront—especially when the objective is to uncover patterns that may not be obvious within familiar coverage. A cleaner approach is to begin with the full distribution of earnings outcomes and let the filtering process surface repeat performers on its own.
The FMP Earnings Surprises Bulk API is the practical entry point for that process. It standardizes quarterly EPS results against consensus estimates across a broad equity universe, which makes it possible to scan for surprise patterns without needing to stitch together multiple sources. Before running any calls, the only requirement is a valid API key.
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 doesn't prove its value by covering more names—it proves it by holding up as conditions become less controlled. The cleanest way to test that is to expand the dataset in steps, starting where information is most complete and gradually moving into areas where estimates are less stable and coverage is thinner. The objective is straightforward: confirm that the signal persists as the environment becomes more variable.
The starting point is the Free plan, where the universe is concentrated in heavily followed large-cap companies such as AAPL, GOOGL, and JPM. This is the most modeled segment of the market—analyst coverage is deep, consensus estimates are tightly grouped, and earnings expectations tend to be well anchored. If a streak-based framework produces consistent results here, it suggests the methodology is functioning as intended rather than reacting to inconsistencies in the data.
Expanding into the Starter plan introduces a broader set of U.S. companies, including smaller-cap and more specialized names. At this level, estimate dispersion naturally widens as analyst coverage drops off. That shift acts as a built-in stress test. When the same screening criteria continue to identify repeat performers under these conditions, the signal begins to look less like a byproduct of tightly modeled forecasts and more like a reflection of underlying execution.
The next step—through the Premium plan—extends the analysis across additional geographies, including markets like the U.K. and Canada. The framework itself doesn't change, but the inputs do. Differences in reporting standards, sector composition, and estimate reliability introduce another layer of complexity. Applying the same logic across these regions ensures that what's being measured—earnings consistency—remains comparable despite those variations.
From Desk-Level Model to Institutional Framework
When a screening process continues to produce stable signals across multiple earnings cycles, it stops being a personal tool and starts to look like institutional infrastructure. What begins as an analyst's method for tracking repeatable earnings beats becomes a candidate for how the broader research platform defines and measures consistency. At that point, the shift is less about expanding coverage and more about aligning methodology—establishing a shared way to interpret operational reliability across sectors.
In practice, that transition is usually driven from within the analyst ranks rather than mandated from above. The people closest to the data—those refining filters, resolving inconsistencies, and testing assumptions quarter after quarter—are the ones who surface what actually holds up. As those workflows mature, they naturally expose the inefficiencies of fragmented approaches: siloed spreadsheets, slightly different definitions across teams, and duplicated effort in reconciling results. A standardized framework replaces that with a common reference point, allowing different coverage groups to work from the same underlying logic.
The operational impact is immediate. Shared dashboards take the place of isolated models, making outputs visible across teams rather than confined to individual desks. Changes to thresholds or screening criteria become transparent and reviewable, instead of being embedded in private files. That visibility improves auditability—inputs, calculations, and assumptions are clearly defined—and creates a foundation for governance. Time spent reconciling discrepancies across teams is reduced, shifting focus toward interpreting what the data is actually indicating.
At that stage, scaling the workflow becomes less about efficiency and more about preserving consistency as adoption widens. Centralized infrastructure, such as the Enterprise plan, allows a process that has already been validated at the desk level to operate across the firm with unified data access, version control, and shared visibility. The goal is not to alter the analysis, but to ensure that as more teams rely on it, the methodology remains intact and comparable across the entire research organization.
Keeping Earnings Consistency in Motion
Earnings consistency is not a static signal—it evolves as expectations reset and new data comes in. Re-running the screen through the FMP Earnings Surprises Bulk API keeps the focus on names where execution continues to outpace consensus, rather than those where the signal has already been absorbed.
Want more? Explore our earlier article: Weekly Signals Desk | Five Notable Valuation Disconnects from the FMP API (March 30 - April 3)
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.
