Meta cut 21,000 employees in 2022–2023. Then it posted $40.6B in net income in 2024 — the most in its history.
That is not a coincidence. It is the business case for mass layoffs during a period of peak revenue. The tech layoffs graph looks paradoxical only if you assume headcount and profit move together. They don't — and for big tech, they now move in opposite directions.
The Tech Layoffs Graph: 2022–2025 by the Numbers
The cumulative data tracked across Layoffs.fyi and public SEC filings tells a consistent story: over 700,000 tech jobs cut in four years, with the peak coming in 2023.
| Year | Jobs Cut | Primary Drivers |
|---|---|---|
| 2022 | ~164,000 | Rate shock, valuation reset, hiring overcorrection |
| 2023 | ~260,000 | Efficiency mandates, post-pandemic correction, activist pressure |
| 2024 | ~150,000+ | AI displacement begins, Intel restructuring, Cisco cuts |
| 2025 | ~90,000+ | AI functional replacement, continued role elimination |
Track real-time layoff data on the Value Add VC Layoffs Dashboard.
The Overhiring Hangover Was the Real Story
Between 2020 and 2021, big tech added headcount at a pace that had no precedent. Zero interest rates made capital free. Remote work removed geographic constraints. COVID accelerated digital adoption. Companies hired for a world that didn't last.
Meta
Peak: ~87,300 (Sep 2022)
Current: ~70,000 (2024)
−20% headcount
Amazon
Peak: ~1.6M (early 2022)
Current: ~1.5M (2024)
−6% headcount
Alphabet
Peak: ~190,000 (2022)
Current: ~180,000 (2024)
−6% headcount
The "Year of Efficiency" — Zuckerberg's framing in 2023 — was not spin. Meta's operating margin went from 20% in 2022 to 41% in 2024. The math was straightforward: fewer employees, same revenue, drastically higher margins.
Why AI Is Accelerating the Tech Layoffs Graph
The 2022–2023 wave was about correcting past mistakes. The 2024–2026 wave is structurally different: AI is now replacing entire function categories, not just trimming fat.
Klarna
AI doing the work of 700 customer service agents
Headcount dropped from ~5,000 to ~3,800 while revenue held
Duolingo
Reduced contractor headcount citing AI productivity gains
Shifted budget from human content creation to model fine-tuning
Shopify
AI-first hiring mandate — teams must prove AI can't do a job before hiring
Permanent structural cap on headcount growth regardless of revenue
Intel
15,500 jobs cut in August 2024 — largest single-announcement layoff of the cycle
Restructuring toward AI chip architecture, eliminating legacy roles
Revenue Per Employee: The Metric Driving Every Decision
Activist investors, public market analysts, and internal finance teams now track revenue per employee as a primary efficiency metric. When it compresses, pressure builds. When it expands, it validates the cuts.
| Company | Revenue/Employee (2022) | Revenue/Employee (2024) | Change |
|---|---|---|---|
| Meta | ~$1.1M | ~$2.1M | +91% |
| ~$1.4M | ~$1.7M | +21% | |
| Microsoft | ~$900K | ~$1.1M | +22% |
This is the real benchmark driving tech CEO decisions. Revenue per employee is now a proxy for AI integration success — and every earnings call gets asked about it. See full earnings data on the Big Tech Earnings Dashboard.
What This Means for Founders and Operators
Opportunities Created
- ✓ Senior tech talent available at pre-2021 comp expectations
- ✓ Ex-FAANG founders with deep domain expertise entering the market
- ✓ AI-native startups can hire for outcomes, not headcount
- ✓ Lean team benchmarks are now table stakes — not a differentiator
What Founders Should Watch
- ⚠ Enterprise hiring freezes compress SaaS sales cycles
- ⚠ HR tech and workforce tools face category headwinds
- ⚠ Consulting and services business models face AI displacement
- ⚠ Revenue per employee is now the benchmark LPs and VCs apply to you too
The tech layoffs graph isn't going flat.
AI is making the case for permanent headcount compression — and every CFO in Silicon Valley has seen the Meta numbers.
Track live tech layoff data on the Value Add VC Layoffs Dashboard. Originally published in the Trace Cohen newsletter.