VC & InvestingMarch 11, 2026ยท9 min read

The Era of Valuation Compression

How 71 public tech companies went from peak ZIRP multiples to the efficiency era โ€” and which sectors are winning the recovery.

TC
Trace Cohen
3x founder, 65+ investments, building Value Add VC

Quick Answer

Tech valuations compressed a median of 52% from 2021 ZIRP peaks โ€” EV/Revenue multiples dropped from 17.4x to 8.1x across 71 public companies. Semiconductors bucked the trend, tripling on AI infrastructure demand, while traditional SaaS fell ~60%. Markets now price on efficiency and AI proximity, not growth alone.

For more than a decade, SaaS had one of the simplest narratives in venture capital. If a company was growing revenue quickly and building predictable recurring revenue, investors assigned very high multiples. Growth was the primary signal. Then everything changed.

I built a dashboard tracking valuation multiples across 71 public tech companies since 2021. The data tells a clear story โ€” and the sectors don't move together.

The Headline Numbers

8.1x

Median EV/Revenue today

Down from 17.4x at peak

52%

Median multiple compression

Since 2021

23.3%

Median revenue growth

Still healthy

41.7

Median Rule of 40 score

Above the threshold

Many public software companies are still growing at healthy rates and meeting Rule of 40 standards. What changed is not growth itself. What changed is how investors value that growth.

The Sector Rotation

The bigger story is not just compression. It is where capital moved:

Semiconductors
5.8x โ†’ 19.3x+233%
Cloud Infrastructure
14.4x โ†’ 8.0x-44%
SaaS (traditional)
~15x+ โ†’ ~4โ€“8x~-60%
Fintech
depressed โ†’ quietly re-ratedโ†‘

Semiconductors are the clearest example โ€” multiples tripled almost entirely due to the AI infrastructure cycle. Meanwhile, traditional SaaS saw their valuation frameworks reset.

The Spread Between Winners and Losers

Best stock since 2020

+1,108%

Worst stock since 2020

-89%

Total spread: 1,197 percentage points. Public tech is no longer moving as a single category. Only four companies maintained EV/Revenue multiples above 15x every year:

Snowflake
CrowdStrike
Nvidia
ARM

All four sit directly in the AI infrastructure ecosystem.

Notable Individual Company Moves

Spotify1.4x โ†’ 7.0x EV/Revenue (+401%)
Meta2.2x โ†’ 7.2x after the 'year of efficiency' reset
NvidiaPeaked at 54x โ€” highest multiple among large public tech
Bill16.8x โ†’ 3.7x (-78%)

The SaaS model itself did not break.

But the valuation environment has become much more selective. Markets are no longer pricing software companies as a single category โ€” they're separating companies based on efficiency, profitability, and how directly they participate in the AI economy.

Explore the full interactive dashboard with 71 companies, 8 sectors, and 590 data points on the SaaS Valuations Dashboard at Value Add VC. Originally published in the Trace Cohen newsletter.

Frequently Asked Questions

How much have tech valuations compressed since 2021?

Median EV/Revenue multiples for public tech companies fell 52% from their 2021 ZIRP peaks โ€” from 17.4x down to 8.1x. Compression wasn't uniform: semiconductors tripled while traditional SaaS companies saw declines of 60% or more.

Which tech sectors are outperforming in valuation multiples?

Semiconductors dramatically outperformed, with median multiples rising 233% from 5.8x to 19.3x driven by AI infrastructure demand. Only four companies โ€” Snowflake, CrowdStrike, Nvidia, and ARM โ€” maintained EV/Revenue multiples above 15x consistently across the period.

What metrics do investors use to value tech companies post-ZIRP?

Post-ZIRP investors weight efficiency and profitability alongside growth. The Rule of 40 (revenue growth rate plus profit margin) has become a key benchmark, with scores above 40 commanding premium multiples. Proximity to the AI economy now functions as a separate valuation premium.

Why did Nvidia's valuation multiple peak so high?

Nvidia peaked at 54x EV/Revenue โ€” the highest multiple among large public tech companies in the dataset โ€” because it sits at the center of the AI infrastructure build-out. As the dominant GPU supplier for AI training and inference, investors priced in years of exceptional demand growth.

Explore 41+ free VC tools, dashboards, and recommended startup software.