The Earnings Power Framework: Finding 10x Returns in "Expensive" Tech Stocks

Meta at $100 → $330. Remitly at 20% yield. Amazon's 100x return. Here's the pattern they all share.

Hey there,

This week I’m going to get into my favorite framework for valuing and evaluating Tech companies.

Most fast growing tech companies look absurdly expensive until you understand what you’re actually buying.

If you’re like me, you might have spent years passing on Amazon at 100x P/E and Google at 30x because these stocks were never cheap. Meanwhile, those same companies companies multiplied investor wealth at unprecedented rates. The problem wasn’t the companies—it was my valuation framework.

A couple of years ago I discovered Adam Seessel’s Earnings Power Framework from his book “Where the Money Is.” This approach finally clicked allowing me to value tech companies without abandoning value investing principles.

Using this framework, I bought Meta at $100 in December 2022 (currently trading 3x higher) and more recently Remitly at $12 with a 20% implied earnings yield. Today I’m breaking down exactly how this works—the complete methodology you need to calculate true earnings power for any tech company. I’ll go into the following:

• Why GAAP accounting artificially deflates tech earnings by 50-300%

• The 3-step process to calculate true earnings power

• How I valued Meta at $100 when everyone thought it was dead

• Why I target 10%+ earnings yields

• Real examples: Meta, Remitly, and how to apply this today

Let me show you why that “overpriced” tech stock might actually be trading at half its intrinsic value.

---

The Problem: Why Traditional Value Investing Fails for Tech

Traditional accounting principles were designed for the Industrial Age, not the Digital Age.

When Rockefeller built Standard Oil, value came from physical assets—refineries, railroad cars, distribution networks. When Ford built his empire, it required factories, assembly lines, and inventory. Accounting rules evolved to measure this. They worked brilliantly for 100 years.

But then something changed.

Software is eating the world.

~Marc Andreessen

The Accounting Distortion That Hides Billions

Here’s the fundamental problem with GAAP when applied to tech companies:

Physical asset investments get capitalized and amortized. Digital investments get expensed immediately.

Walmart builds a new store:

• Costs $10 million

• GAAP capitalizes this as an asset

• Depreciation expensed over 20-30 years

• Current year impact: ~$400,000

• Reported earnings: Minimally affected

Meta invests in AI infrastructure:

• Spends $10 million on R&D, engineers, data centers

• GAAP expenses this immediately

• Current year impact: $10 million expense

• Reported earnings: Massively depressed

Both companies are investing for future growth. Both are deploying capital to generate future cash flows. But the accounting treatment creates a 25x difference in reported profitability.

This distortion is massive at scale. In 2020, the big five tech companies spent $125 billion on R&D alone. Add sales & marketing for customer acquisition, and you’re looking at $200+ billion in growth investments hitting the income statement as current expenses.

Result: Tech companies appear far less profitable than they actually are.

Amazon returned 2,300x since its IPO despite looking “expensive” every single year. The earnings power was real—just hidden behind Industrial Age accounting.

The Solution: Earnings Power Framework

Adam Seessel’s insight: A tech company’s current income statement is unreliable for measuring its long-term ability to generate cash.

The question to ask instead: What could this business earn if it stopped investing aggressively in growth and maximized current profitability?

Step 1: Project 3 Years Forward (Conservative Growth)

Often Analyst will focus on current or next year’s earnings. This works for mature businesses like Coca-Cola with steady, predictable growth.

Tech companies with low market share in massive markets operate differently. They’re in expansion mode. Every dollar of current profit is reinvested to capture market share and build network effects.

The fix: Project earnings 3 years forward using conservative assumptions.

Why 3 years specifically?

• Not 5 years: Too uncertain, at some point just guessing.

• Not 1 year: By focusing on the short term, you’ll miss the amazing compounding engine that

Meta Example (December 2022)

When I bought Meta at $100 (Link to video at the time), everyone thought it was dead. The stock had cratered from $380 to $100 in one year. Narrative: Zuckerberg’s metaverse spending was lighting money on fire, TikTok was killing them, Apple’s iOS changes destroyed their ad business.

But let’s look at the numbers:

2022 Baseline:

• Revenue: ~$117 billion

• Operating margin: 25% (down from 40% in 2021)

• The stock had lost $650 billion in market value

• Market cap: ~$250 billion

Why the depressed margins?

Two factors:

1. $10 billion annual Reality Labs spending (metaverse)

2. Heavy AI/ML investment to compete with TikTok via Reels

3-Year Forward Projection (2025):

• Revenue: $117B with 5% growth = $135 billion

• I used 5% growth despite historical 20%+ because I wanted to be extremely conservative (and avoid projecting forward the covid bump)

• Reality: Meta’s core business (Facebook, Instagram, WhatsApp) had 3 billion daily users and was still growing outside the U.S.

This projection alone didn’t make Meta cheap. The magic happened in Step 2.

Step 2: Calculate “Steady State” Margins

This is where you reveal the true earnings power hidden by GAAP.

The question: If this company stopped investing in growth and ran for maximum profitability, what margins would it achieve?

Let me show you why this matters with Meta’s situation in December 2022:

Meta’s Reported Situation:

• Operating margin: 25%

• Down from 40% in 2021

• Reality Labs losing $10B annually

• Heavy AI spending to compete with TikTok

But here’s what the bears missed: Meta’s core business (ex-Reality Labs) was still one of the most profitable businesses ever created.

The Steady-State Analysis:

Looking at Meta’s segments:

• Family of Apps (Facebook, Instagram, WhatsApp): Generated all the revenue

• Reality Labs: Pure spending, $10B annual losses

The calculation:

• Take out the $10B Reality Labs drag

• What margins should a mature digital advertising platform achieve?

Historical context:

• Meta’s margins 2018-2020: 35-45%

• Google’s advertising business: 30-35% margins

• These are software businesses with network effects and minimal marginal costs

My conservative estimate:

• Steady-state margin for Meta’s core business: 35%

• This assumes some continued AI spending but not the emergency levels of 2022

2025 Earnings Power Calculation:

• Revenue: $135 billion

• Steady-state margin: 35%

• Adjusted earnings power: $47 billion

Compare this to Meta’s 2022 reality:

• Reported operating income: ~$29 billion (25% margin)

• Adjusted earnings power: $47 billion

• The true earnings power was 60% higher than reported.

Step 3: Apply the Earnings Yield Test

After adjusting for forward earnings and steady-state margins, I target an Adjusted PE < 10.

Adjusted PE = Market Cap/ Adjusted Earnings

Why?

1. Higher adj. PE = Higher margin of Safety

2. The bigger the disconnect the higher the Return when Mr. Market’s “narrative” flips to positive again

Meta’s Earnings Yield - Adj. PE inverted (December 2022):

• Market cap: $250 billion

• 2025 adjusted earnings: $47 billion

• Earnings yield: 18.8%

That’s a 5.3x P/E on adjusted 3-year forward earnings.

For a business with:

• 3 billion daily active users

• Network effects across four platforms

• 98% gross margins

• Proven ability to monetize via advertising

• Still growing user base globally

This wasn’t expensive. This was absurdly cheap.

The market was pricing in permanent margin destruction. Reality: The margin compression was temporary and self-inflicted (Reality Labs spending). Management could flip the switch and return to 35%+ margins any quarter they chose.

One Important Caveat

Just doing a valuation using comparable margins and projecting forward 3 years is not enough. There are 2 other factors that need to work here:

1. The business needs to have a moat- something that allows it to protect it’s margins from erosion over time w/ a long runway for growth:

   1. In the case of META this was the 3B users w/ network effects

   2. As well as Meta’s ability to clone or buy competitors effectively

2. Management needs to be aligned w/ shareholders and rationally allocate capital

   1. Zuckerberg is META’s largest shareholder (2x next largest) and has majority control

   2. He has a history of strong capital allocation (eg. Instagram acquisition)

In this Scenario we needed to be certain that Management would rationalize their spending. Luckily given Zuckerberg’s history of hyper rational behavior, this was easy predict over a 3 year period.

How Meta Played Out

December 2022 purchase price: $100 Current price (November 2024): ~$330 Return: 230% in under 2 years

What changed? The market realized exactly what the Earnings Power Framework showed:

1. Reality Labs spending, while continuing, wasn’t permanent margin destruction

2. Reels successfully competed with TikTok (140 billion daily plays)

3. AI improvements restored ad effectiveness post-iOS changes

4. Operating margins recovered to 38% by 2024

The earnings power was always there. It was just hidden behind aggressive growth spending that GAAP treated as current expenses rather than investments.

The Broader Pattern: It Repeats Everywhere

This framework applies across the digital economy. Any company that’s:

• Asset-light (software-based)

• Investing heavily in growth

• Operating in large, growing markets

• Building durable competitive advantages

Historical examples:

• Amazon (2012-2018): Appeared expensive at 100x+ P/E while building AWS and logistics. True earnings power hidden by aggressive reinvestment. Stock returned 10x.

• Netflix (2015-2018): Operating near breakeven due to content spending. Market feared cash burn. Steady-state margins for streaming: 25%+. Stock went $100 to $400.

• Shopify (2017-2020): Running at losses due to R&D. Platform margins at scale: 25-35%. Stock went $100 to $1,500.

The pattern: Growth spending depresses current earnings → Market treats as “unprofitable” → Eventually realizes it’s actually highly profitable capital allocation → Stock reprices violently upward.

Critical Guardrails: When NOT to Use This

This isn’t a license to justify any valuation. You need discipline.

Only apply to companies with:

1. Clear competitive moats:

• Network effects (Meta, LinkedIn, Airbnb)

• Low-cost advantages (Amazon)

• Brand power (Apple)

• Switching costs (Adobe)

No moat = no earnings power. Growth without competitive advantage attracts competition that destroys margins.

2. Verify spending is working:

• Customer acquisition costs trending down

• Lifetime value trending up

• Unit economics improving with scale

• Market share gains sustained over years

If spending billions on growth but not gaining share, that’s poor capital allocation, not hidden earnings power.

3. Large addressable markets:

• Total market needs to be multiples above current revenue

• Market must be growing, not shrinking

• Realistic path to capturing significant share

4. Management capital discipline:

• Clear articulation of return hurdles

• Spending becoming more efficient over time

• Ability to articulate path to profitability

Bezos always articulated clear ROI frameworks. If management just spends with no discipline, earnings power is questionable.

The edge isn’t complex. It’s using accounting that reflects how digital businesses actually create value. Most investors compare tech companies to the S&P 500 using reported earnings. They’ll keep passing on “expensive” stocks that 10x.

You now have the framework to see what they’re missing.

---

A Stock Trading at a 20% Earnings Yield

Right now, I’m sitting on a position that mirrors the Meta setup from 2022—but earlier in the cycle.

The market sees: “No earnings, declining from highs, too risky.”

The Earnings Power Framework reveals: A 5x P/E on 3-year forward earnings power.

This is the exact same pattern that delivered 230% returns with Meta in under 2 years.

What I’m seeing:

• $2.2B market cap, down 60% from highs

• Growing 30%+ annually in a $2 trillion market

• Currently breakeven—every dollar reinvested in high value growth

• Clear path to 20%+ steady-state margins

• $440M earnings power by 2027 = 20% earnings yield today

For context: I bought Meta at a 5.3x adjusted P/E. It returned 230% in under 2 years.

The full analysis below includes:

• 3-year projections with conservative assumptions

• Detailed margin analysis using comparable models

• Why the competitive moat protects these economics

• Exact entry price, position sizing

This is the research paid subscribers get 4-8 times per month, plus full portfolio transparency and real-time updates.