Signals Desk Weekly | Multi-Year CAGR Strength Taking Shape Across Five Names (Sept 7-11)

Most of the widest revenue-to-EBITDA spreads in a five-year screen are not operating leverage at all. They are companies whose base year happened to be terrible. Run the calculation across the large-cap universe and the top of the list fills with pandemic troughs, one-time charges and near-zero starting margins, all of which produce spectacular growth rates and tell you nothing. Strip those out and what remains is a much shorter list: EMCOR Group, NVIDIA, Amazon, Williams Companies and Cboe Global Markets.

The analysis uses FMP's Income Statement API to compare five-year revenue and EBITDA CAGRs across the group, all measured over the same FY2020 to FY2025 window and all starting from a base year with a defensible operating margin. This article works through the five names, the mechanism behind each spread, and how to build the same screen with the quality gate already attached.

Key Takeaways

• The base year does most of the work in a five-year CAGR. Names with starting EBITDA margins near zero produced spreads above 100 percentage points in this screen and were removed, because a company going from a 2% margin to a 10% margin generates arithmetic, not evidence.
• EMCOR's spread comes from a genuine margin transformation: EBITDA margin moved from roughly 4% to nearly 12% while revenue nearly doubled, driven by the electrical work behind data-centre construction.
• Amazon shows why EBITDA is not the end of the analysis. Operating earnings compounded at twice the revenue rate, but trailing free cash flow has turned negative on AI infrastructure spending.
• Two of the five require a definitional caveat before their margin percentages mean anything, because Williams and Cboe both report revenue lines that include large pass-through components.

Five Names Where Earnings Compounded Faster Than Sales

EMCOR Group, Inc. (EME)

5-Year Revenue CAGR: 14.07%
5-Year EBITDA CAGR: 40.14%

EMCOR carries the widest clean spread in the screen at roughly 26 percentage points, and the mechanism is the simplest to verify. Revenue roughly doubled over the five years while EBITDA rose more than fivefold, which means EBITDA margin moved from about 4% to nearly 12%. For a specialty construction contractor, a margin in the 4% range is normal rather than depressed, so the starting point here is not flattering the arithmetic the way a trough year would.

What changed is the type of work. The most recent quarter showed revenue up nearly 20% to $5.15 billion with operating margin above 10%, and remaining performance obligations reaching a record $17.1 billion, up 44% year over year. Electrical construction grew 24% with segment margin expanding 210 basis points, and the network and communications category that houses data-centre work rose 45% in electrical and more than doubled in mechanical. Management has described AI facilities as carrying substantially more revenue content per megawatt than conventional builds because of cooling complexity.

The offsetting detail matters. Mechanical construction margin contracted 110 basis points on a shift toward larger cost-plus contracts, and the company has spent roughly $750 million acquiring five electrical contractors whose combined margins will take time to normalize through purchase accounting. That makes the composition of future margin the thing to watch rather than the headline. FMP's Financial Statement Growth API is the practical way to track whether the revenue and earnings growth rates stay separated or begin converging as acquired revenue dilutes the mix.

NVIDIA Corporation (NVDA)

5-Year Revenue CAGR: 66.90%
5-Year EBITDA CAGR: 90.97%

NVIDIA's spread is about 24 percentage points, but the striking part is where it starts. The base fiscal year already carried a 34% EBITDA margin, which is a strong result for a semiconductor company in any year. Expanding from there to roughly 67% while revenue compounded at 67% annually is a different phenomenon from a recovery story, and it is the only case in this screen where an extremely high growth rate sits on an equally high starting margin.

The most recent quarter extends the pattern rather than moderating it. Revenue of $96.2 billion more than doubled year over year, data centre accounted for $89.0 billion of that, and gross margin held at 75.0%, up 2.6 points from a year earlier. Operating expenses grew around 10% sequentially against 18% sequential revenue growth, which is the arithmetic that keeps the spread open. Guidance points to roughly $108 billion for the following quarter, explicitly excluding China data centre compute.

The analytical question is no longer whether leverage exists but where it terminates. Gross margin guidance stepping down to 74.0% is a small signal that pricing and mix have limits, and a company at this margin structure has far more room to compress than to expand. FMP's Financial Ratios API puts gross, operating and EBITDA margin on one quarterly series, which is where any flattening would appear before it reaches the growth rates.

Amazon.com, Inc. (AMZN)

5-Year Revenue CAGR: 13.18%
5-Year EBITDA CAGR: 26.52%

Amazon's earnings compounded at roughly twice the pace of its revenue, lifting EBITDA margin from about 13% to 23% on a revenue base that nearly doubled to over $700 billion. At this scale, ten points of margin is an enormous absolute number, and the source is well understood: the mix has shifted toward cloud and advertising, both of which carry far higher incremental margins than retail.

The most recent quarter shows the mechanism operating at full strength. Net sales rose 20% to $200.6 billion while operating income grew 43%, with AWS operating income up 63% and AWS revenue growth accelerating to 37%, the fastest in eighteen quarters. Advertising advanced 26%. Two items need separating from that picture: a large non-operating gain from an investment holding inflated net income far beyond operating income, and it does not touch EBITDA in any meaningful operating sense.

The more consequential caveat is capital intensity. Property and equipment purchases rose by tens of billions year over year for AI infrastructure, and trailing twelve-month free cash flow has swung from a substantial inflow to an outflow. Operating leverage that does not reach free cash flow is a different proposition from operating leverage that does, and FMP's Cash Flow Statement API is where the gap between the two series becomes visible. Whether the spend eventually converts into the same margin structure is the open question underneath an otherwise clean five-year record.

The Williams Companies, Inc. (WMB)

5-Year Revenue CAGR: 9.13%
5-Year EBITDA CAGR: 18.52%

Williams shows about 9 percentage points of spread, with EBITDA margin rising from roughly 41% to 62%. Before reading anything into those levels, the definitional point has to be made: a midstream operator's reported revenue includes commodity sales that pass through at little or no margin, so the ratio moves with gas prices as much as with efficiency. The growth rates remain comparable because the same definition applies at both ends of the window, but the margin percentages are not comparable to an industrial company's.

With that caveat in place, the underlying trend is real. Adjusted EBITDA of $1.92 billion in the most recent quarter rose 6%, available funds from operations rose 10%, and full-year adjusted EBITDA guidance was raised by $200 million. The growth is contracted rather than commodity-driven: transmission projects entering service, gathering volume increases, and storage revenue. Dividend coverage near 2.3 times funds from operations indicates the earnings are converting.

The complication is what comes next. Williams agreed to acquire Momentum Midstream for up to $5.5 billion, adding a large Haynesville gathering platform, alongside a joint venture bringing $5.34 billion of third-party capital into its power projects and growth capital expenditure guided at $7.3 to $7.9 billion. Pro-forma leverage moves to roughly 3.75 times. A company financing that much growth changes its own denominator, and FMP's Enterprise Values API is the right reference for watching whether EBITDA growth is outpacing the net debt being added to produce it.

Cboe Global Markets, Inc. (CBOE)

5-Year Revenue CAGR: 6.59%
5-Year EBITDA CAGR: 15.22%

Cboe has the narrowest spread at roughly 9 percentage points and the slowest revenue growth, which makes it the clearest example of leverage rather than expansion. The same caveat as Williams applies in a different form: exchange operators report gross revenue that includes liquidity payments and routing costs, so net revenue is materially lower than the headline figure and margin ratios calculated on gross revenue understate the actual economics.

The operating picture is unambiguous. Record net revenue of $731.6 million grew 25% year over year, adjusted earnings per share rose 45%, and adjusted operating margin expanded from 63.7% to 70.4%. Options drove it, with revenue up 30% on average daily volume up 26% and revenue per contract up 6%. Index options volume rose 32%. The company raised its organic revenue growth target for the year to the mid-to-high teens.

The structural point is that an exchange's cost base is close to fixed, so incremental volume converts to earnings at very high rates, and this works identically in reverse. Two signals sit against the quarter: options market share slipped slightly and equities share fell more than a point, meaning the volume boom is doing the work rather than competitive position. Recurring data revenue is the offsetting stabilizer, and its growth guidance was also raised. FMP's Key Metrics TTM API normalizes the returns and cash-generation measures that separate a volume cycle from a durable margin reset.

Where the Extra Margin Actually Comes From

The useful output of this screen is not the ranking. It is the fact that four fifths of the candidates with the largest spreads had to be thrown out, and the reason each one failed says something about how the metric behaves. Companies with base-year EBITDA margins near zero generated spreads above 100 percentage points purely because dividing by a small number produces a large answer. Pandemic-affected businesses produced 30 to 40 point spreads that measure a return to normal. Companies carrying a large impairment or litigation charge in the base year produced the same effect through the accounting rather than the economy. Separations and large acquisitions produced spreads by restating what the company is. None of these is operating leverage, and all of them will sit at the top of an unfiltered list.

What survives that filter divides into two mechanisms. EMCOR, NVIDIA and Cboe are cases of incremental margin: the additional unit of revenue costs materially less to produce than the average unit, whether through segment mix in construction, an operating expense base growing at half the rate of revenue, or a fixed cost structure absorbing higher volume. Amazon and Williams are cases of composition: the business is selling a different mix than it was five years ago, weighted toward cloud and advertising in one and toward contracted infrastructure in the other. The first mechanism is more fragile because it reverses on the way down. The second is more durable but slower and usually financed.

Testing which one is present requires leaving the income statement, and the endpoint coverage on the FMP platform is what makes that a query rather than a project. The screen itself scales through the Income Statement Bulk API, which applies one ruleset across the universe rather than reconciling per-company pulls. The first follow-up is always cash: comparing the EBITDA series against the Cash Flow Statement API establishes whether operating earnings are reaching free cash flow or being absorbed by working capital and capital expenditure, which is precisely the distinction that separates Amazon's record from EMCOR's in this group.

Balance sheet capacity is the second test, because leverage financed with debt is not the same signal as leverage generated internally, and the Financial Scores API gives a fast read on that before a deeper look. Expectations are the third. The Price Target Consensus API shows whether published targets already reflect the margin structure, and the Financial Estimates API shows whether forward revenue and earnings forecasts are extrapolating the spread or assuming it narrows. A spread the market has fully absorbed is a description of the past. A spread that forward estimates assume will close, in a business where the mechanism looks structural, is the more interesting configuration, and it is the one this screen is built to surface.

Building a Consistent CAGR Screening Framework

Building a useful CAGR screen is less about the formula itself and more about maintaining discipline in the underlying dataset. The calculation is straightforward; what determines whether the result is meaningful is data consistency. Every company needs to be evaluated using the same reporting periods, identical financial line items, and the same time horizon. Once those inputs are standardized, growth rates become comparable across industries, capital structures, and business models. The workflow below shows how to structure that process using FMP's Income Statement data — starting with a single company and then scaling the exact same logic across a broader universe.

Step 1: Pull Income Statement Data

Begin with a single symbol to establish the baseline. Query the standard Income Statement API to retrieve the full set of historical reporting periods needed for the calculation.

As long as your API key is active, one request gives you the raw time series you'll be working with. For example:

Endpoint:

https://financialmodelingprep.com/stable/income-statement?symbol=AAPL&apikey=YOUR_API_KEY

Step 2: Gather Historical Figures

From the JSON output, select the specific metric you want to analyze — revenue, EBITDA, EPS, or another line item. Arrange the values in proper chronological order before doing any math. This step is easy to overlook, but it's critical: CAGR only makes sense when the starting and ending points are clearly defined and consistently ordered.

Step 3: Calculate CAGR

Once the first and last data points are set, calculate CAGR using the standard formula:

CAGR = (Ending Value / Beginning Value)^(1 / Years) - 1

This reduces several years of performance into a single annualized figure, making it easier to compare growth profiles across companies without getting lost in interim volatility.

Step 4: Scale Screening with Bulk API

After validating the method on one symbol, broaden the workflow using the Income Statement Bulk API:

https://financialmodelingprep.com/stable/income-statement-bulk?year=2025&period=FY&apikey=YOUR_API_KEY

Running the same calculation at scale lets you build filters — for instance, highlighting companies that clear a five-year revenue CAGR threshold — while ensuring every ticker is processed under the same ruleset. Once the bulk pull is in place, updating or rerunning the screen is effectively a single action.

Scaling the Framework Without Changing the Methodology

The strength of this type of screen comes from consistency, not complexity. Once the formula, reporting periods, and financial line items are defined, the objective is to keep the methodology fixed while gradually widening the universe being tested. Expanding coverage should not require changing the framework itself — only the number of companies moving through it.

That's why the workflow is easiest to validate in a smaller environment first. Within the Basic plan, the Income Statement endpoints provide enough historical coverage to align reporting periods properly, normalize the selected metrics, and verify that the CAGR calculations are producing comparable outputs across companies. At this stage, the emphasis is less about scale and more about making sure inconsistencies in filings or missing data are not distorting the screen.

From there, expanding into the Starter plan simply broadens the sample size. The screening logic remains identical, but the larger universe makes sector-level comparisons more useful. Patterns that initially appear company-specific can then be evaluated against peers, industries, or market-cap cohorts to determine whether the operating leverage signal is isolated or part of a broader trend developing within a segment of the market.

The Premium plan extends that same structure further by increasing historical depth and geographic coverage. The underlying process still does not change. What changes is the scope of observation — allowing the same framework to be applied across wider datasets without introducing new assumptions or altering the screening criteria midstream. That continuity is what makes the process repeatable over time rather than dependent on one-off observations or isolated market conditions.

Watching the Spread Rather Than the Rank

A five-year spread is a description of a window that is about to move: next year the base rolls forward, and several of these gaps will look different for no operating reason at all. Running the same screen on the same ruleset each quarter through the Income Statement API and Income Statement Bulk API is what turns that into a series worth reading rather than a snapshot worth ranking.

If you found this useful, you might also like: Weekly Signals Desk | Concentrated Analyst Revisions via the FMP API (Aug 31-Sept 4)

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.