This week's earnings scan surfaced a small but telling cluster of names that continue to separate themselves from the broader tape. While market attention has rotated toward macro resilience and selective operating discipline, a handful of companies are quietly extending clean earnings-beat streaks—quarter after quarter—without dramatic multiple expansion or headline risk.

The signal comes directly from the FMP's Earnings Surprises Bulk API, which allows a top-down sweep of quarterly EPS outcomes before any ticker-level bias enters the process. In this article, we break down how that dataset highlights five companies sustaining measurable execution momentum, and why repeatability—not one-off surprises—is increasingly where conviction is forming.

Five Companies With Long Earnings Beat Streaks

UPWK Upwork Inc.

Beat Streak: 15 quarters.
Next quarterly report: Feb. 11 — EPS: $0.23; Revenue: $197.25M (consensus).

A 15-quarter earnings-beat streak places Upwork in rare territory, particularly for a platform business exposed to cyclical hiring and discretionary enterprise spend. The durability of the streak suggests less about short-term demand spikes and more about cost discipline, pricing mechanics, and gradual improvements in take-rate efficiency. In a market that has periodically questioned the sustainability of freelance marketplaces, the data shows a company that has repeatedly cleared a low but persistent bar for execution.

What makes the signal notable is its consistency through multiple macro backdrops—post-pandemic normalization, tech layoffs, and uneven SMB demand. Rather than dramatic upside surprises, Upwork's pattern reflects incremental beats that compound over time. This is often where operating leverage quietly shows up first. Income statement detail—particularly contribution margin and sales-and-marketing efficiency—helps contextualize whether the streak is being driven by structural improvements or temporary expense restraint.

Going forward, the earnings cadence itself is the primary tell. Monitoring quarterly revenue per client, active client counts, and margin progression alongside the earnings record can help determine whether the streak continues to reflect underlying platform health or simply disciplined expectation-setting.

ADI Analog Devices, Inc.

Beat Streak: 8 consecutive quarters.
Next quarterly report: Feb. 18 — EPS: $2.25; Revenue: $3.10B (consensus).

Analog Devices' eight-quarter beat streak stands out given the well-documented cyclicality in semiconductors. While parts of the chip sector have moved through inventory corrections and demand digestion, ADI's results suggest a steadier end-market mix and a balance sheet capable of absorbing volatility without operational whiplash. The streak aligns with the company's exposure to industrial, automotive, and communications infrastructure—segments where demand tends to move slower, but with more visibility.

Here, the signal is less about growth acceleration and more about earnings quality. ADI's ability to exceed expectations across a downcycle implies resilience in gross margin management and disciplined capital allocation. Analysts tracking the earnings record often pair it with cash flow metrics, backlog commentary, and segment-level revenue disclosure to assess whether beats are being supported by backlog conversion or pricing stability.

For readers watching this name, the relevant datasets extend beyond quarterly EPS. Inventory levels, order trends, and regional revenue breakdowns from the income statement and balance sheet offer useful context for interpreting whether the streak reflects normalized conditions—or continued insulation from broader semiconductor swings.

FIVE Five Below, Inc.

Beat Streak: 5 quarters.
Next quarterly report: March 18 — EPS: $3.44; Revenue: $1.60B (consensus).

Five Below's five-quarter streak comes against a retail backdrop marked by uneven discretionary spending and shifting value perceptions. The company's ability to repeatedly exceed expectations suggests that its price-point positioning continues to resonate, even as consumer behavior remains selective. Unlike retailers dependent on promotional intensity, Five Below's earnings pattern points to merchandising discipline and traffic resilience rather than margin giveaways.

The streak matters because it coincides with an environment where consensus estimates have been cautious across specialty retail. Clearing those bars repeatedly indicates tighter execution at the store level—inventory turns, shrink management, and category mix all play a role. Same-store sales and gross margin data are particularly useful in understanding whether the earnings beats are volume-driven or cost-driven.

As the company heads into its next report, the earnings record is best read alongside store expansion metrics and operating margin trends. Those datasets help clarify whether consistency is coming from scale efficiencies or from maintaining relevance in a competitive value segment.

AMBP Ardagh Metal Packaging S.A.

Beat Streak: 4 quarters.
Next quarterly report: Feb. 26 — EPS: $0.02; Revenue: $1.28B (consensus).

Ardagh Metal Packaging's four-quarter streak is notable given its capital-intensive profile and sensitivity to input costs. Packaging companies often struggle to deliver consistent earnings surprises due to contract timing, volume variability, and commodity exposure. AMBP's recent record suggests tighter control over those variables, particularly in an environment where beverage demand has been stable but not accelerating.

The signal here is subtle but important: modest beats in a low-margin business often point to incremental efficiency gains rather than revenue inflection. Analysts tend to look at EBITDA margins, contract renewals, and free cash flow conversion to determine whether earnings consistency reflects sustainable operating improvements or favorable timing effects.

For AMBP, balance sheet data and cash flow statements add essential context. Tracking leverage ratios and capex intensity alongside the earnings streak helps frame whether operational discipline is translating into financial flexibility, not just quarterly EPS outcomes.

GH Guardant Health, Inc.

Beat Streak: 3 quarters.
Next quarterly report: Feb. 19 — EPS: -$0.43; Revenue: $267.59M (consensus).

Guardant Health's three-quarter beat streak carries a different signal profile than the others on this list. With expected losses still embedded in consensus estimates, the streak reflects narrowing losses or revenue outperformance rather than outright profitability. In diagnostics and life sciences, that distinction matters: consistency often shows up first in expense control and test-volume scaling before margins turn positive.

What stands out is the repeatability. One earnings beat in this space can be attributed to timing or trial-related noise; three in a row begins to suggest improving visibility. Revenue growth rates, test adoption trends, and operating expense trajectories from the income statement help anchor whether the streak reflects structural progress or temporary cost alignment.

For GH, the most informative datasets extend beyond headline EPS. Monitoring quarterly revenue composition, R&D spend, and cash runway provides a clearer picture of how earnings performance fits into the company's longer operational arc, without relying on speculative inflection narratives.

Interpreting What Repeatable Beats Are Actually Telling Us

The real analytical lift comes when earnings data is not treated as a self-contained signal. Earnings beats gain context when they are cross-checked against underlying fundamentals and broader conditions—an approach explored in FMP's breakdown of how earnings surprises align (or diverge) from macroeconomic signals. In practice, pairing earnings streaks with income statement and cash flow data helps distinguish between beats driven by structural efficiency and those benefiting from transient conditions.

That layered approach reflects how repeatability screens are typically used on institutional desks. Earnings outcomes sourced from FMP become more informative when evaluated alongside margin trends, working-capital dynamics, estimate revisions, and even insider activity. When those inputs move in parallel, consistency begins to look intentional rather than incidental.

Viewed this way, beat streaks function less as signals to pursue and more as filters for focus. They narrow a broad universe to companies where the reporting cadence itself carries informational weight—supporting a more grounded read on operational momentum without leaning on directional forecasts or narrative extrapolation.

Building a Repeatability Screen with FMP Data

If the objective is to find companies that clear expectations repeatedly—not sporadically—the workflow needs to start at the dataset level, not the ticker level. The most efficient way to do that is to scan the full universe of quarterly EPS outcomes first, then narrow the focus only after the data begins to separate itself. FMP's Earnings Surprises Bulk API is designed for this kind of wide-angle pass, allowing you to work from coverage breadth to signal depth rather than the other way around. Before running the process, confirm that your API key is active.

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

Once that pool is established, the analysis shifts from occurrence to persistence. For each ticker that passed the first screen, pull its full quarterly earnings history using the Earnings Report API:

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

Reviewing the full earnings sequence allows you to count consecutive beats and distinguish one-off surprises from sustained execution. How strict the criteria should be depends on the use case. Some teams flag three consecutive beats as meaningful; others require a minimum percentage surprise to reduce statistical noise. Regardless of where the threshold is set, the purpose is the same: to translate raw earnings outcomes into a repeatability profile that captures both frequency and clarity of the signal.

Broadening the Universe as Coverage Scales

A practical way to validate a repeatability screen is to start with limited coverage and expand only once the logic proves sound. The Free plan offers a sufficient testing ground, covering widely followed names like AAPL, GOOGL, and JPM. That scope is large enough to confirm that the streak logic behaves as expected—without introducing unnecessary noise from thinner or less consistent reporters.

Once the mechanics are stable, the Starter plan becomes the natural next step. It opens access to the full U.S. equity universe, allowing the screen to operate across a broader mix of sectors and market capitalizations. That expansion matters because repeatability signals tend to look different once smaller caps and less-covered industries enter the sample.

For teams looking to extend the same framework internationally, the Premium plan adds U.K. and Canadian listings, enabling the screen to run across multiple geographies without structural gaps. At that point, the methodology remains unchanged—the only variable is coverage depth—making it easier to compare consistency patterns across regions using the same analytical lens.

From Desk-Level Insight to Firmwide Standard

A repeatability screen proves its real usefulness once it moves beyond an individual analyst's toolkit and becomes part of the firm's shared analytical fabric. At that stage, the value is no longer speed or convenience—it's alignment. When multiple desks reference the same earnings-streak logic, the signal turns into a common point of orientation for portfolio managers, strategists, and risk teams, reducing the drift that comes from parallel models and one-off interpretations.

Analysts play a central role in that transition. By formalizing streak definitions, surprise thresholds, and supporting metrics around a consistent dataset, they effectively champion a standard others can trust. Shared inputs mean conclusions are easier to compare, assumptions are transparent, and downstream use—whether in sizing frameworks, factor attribution, or risk reviews—becomes easier to trace and validate. The result is less fragmentation and fewer debates about whose numbers are “right.”

Sustaining this at scale requires infrastructure that supports governance as much as analysis: shared dashboards instead of personal spreadsheets, version control instead of silent revisions, and clear data lineage instead of manual handoffs. This is where an environment like the Enterprise plan fits naturally into the workflow, providing the scaffolding for teams to build once and reuse consistently. At that point, the earnings-streak framework stops behaving like a model owned by one desk and starts functioning as an institutional signal—portable, auditable, and understood across the organization.

Turning Past Beats Into Forward Indicators

Once earnings streaks are established, the focus naturally shifts from cataloging history to observing whether discipline holds as new data prints. Continuously refreshing the framework through the FMP's Earnings Surprises Bulk API turns prior consistency into a live reference point—highlighting which companies continue to reinforce their execution profile and which begin to diverge. At that point, the streak itself becomes less about the past and more about how the signal evolves quarter by quarter.

Want more? Explore our earlier article: Five Stocks Standing Out: Target-Gap Signals Pulled from the FMP API (Dec 1-5)