Cloudflare's CEO Just Named the AI Layoff Pattern: Measurers Out, Builders In

TV
Thiago Victorino
6 min read
Cloudflare's CEO Just Named the AI Layoff Pattern: Measurers Out, Builders In

On May 20, 2026, Matthew Prince, CEO of Cloudflare, published an opinion piece in the Wall Street Journal that did something most layoff announcements work hard to avoid. He named the pattern.

In his words (verbatim from the open paragraph): “We haven’t found another example in U.S. business history of a public company growing at more than 30% that laid off more than 20% of its workforce. Yet what we did is likely going to become the norm over the next year.”

Read that twice. Record-setting revenue growth and a one-fifth headcount cut, in the same quarter, in the same company. Prince’s own framing of why he wrote the piece (also verbatim): “This is a story about artificial intelligence, but executives and commentators are misunderstanding how it will disrupt business and who will be affected.”

The subhead the WSJ ran with, also verbatim, is the part every operator needs to read out loud at the next leadership meeting: “The company has less need for middle managers, operations jobs and other measuring positions.”

That last word is the one to underline. Measuring positions.

What Prince Named

The rest of the op-ed sits behind WSJ’s paywall. What we can verify with certainty is what WSJ’s own editorial indexing layer surfaced as the article’s spine keywords: analysts, automation, builders, cut, employ, jobs, layoff, MEASURERS, revenue, sellers. WSJ’s metadata pipeline only emphasizes terms that recur with weight in the body. The capitalized one, the one a publishing system treats as the argumentative backbone, is measurers.

So here is what we can say with high confidence from the verbatim subhead and the keyword spine, and what we are inferring from the rest: Prince’s argument splits the workforce along a single axis. On one side are measurers, people whose primary job is to track, report, summarize, or coordinate numbers and status that AI can now track, report, summarize, and coordinate on its own. On the other side are builders, people whose primary job is to create the things AI cannot yet create unaided: product decisions, customer relationships, code that ships to production with judgment behind it, sales conversations that close.

The cut was not bottom-up. It was middle-out.

A note on what the article also makes clear, per the keyword spine: Cloudflare is hiring at record open-position levels. The layoff is a recomposition, not a downsizing. The seats are not being eliminated, they are being refilled with a different shape of work.

Why This Is the First Named Case

We have written before about the centaur era, where the unit of measurement is the team plus its tools, not the model. We have written about the two percent productivity reality that does not match the productivity narrative. We have written about how marketing and other functions are becoming governance teams.

What we have not had until May 20 is a U.S. public-company CEO putting a name on the cut and the math behind it. Measurers versus builders is Prince’s framing, and it lands because it does what most layoff communications refuse to do: it tells the people who were cut why they were cut, in a category their peers will recognize, and it tells the people who stayed why they stayed.

The companies that try to use this framing without doing the work will get caught. The companies that quietly do the work and never name it will lose the people who could have helped them through it. Prince did both. He named it and he is doing it in public.

The Operating-Model Question Every CEO Now Has To Answer

If you are running a company of any size and the board has not yet asked the version of this question that begins with “why are we not Cloudflare” or its sharper inverse “why are we Cloudflare,” you have maybe one quarter before they do. The question they will actually be asking is this one: which roles in this company are measurers, and which are builders, and what is our plan to recompose, not just to cut.

Three traps to avoid in how you answer.

Trap one: confusing measurers with junior staff. Many of the measurers Prince is describing are senior. They are middle managers whose value used to be coordination, status synthesis, and number-rolling-up. The cut runs through the org chart horizontally, not vertically. Treating it as a junior-cut conversation reads to the room as a hand-wave.

Trap two: confusing builders with engineers. Builders, in Prince’s framing and in the operating reality the keyword spine supports, are not only people who write code. Sellers are explicitly on the builder side of the spine. So are the people who design products, who own customer relationships, who make judgment calls AI cannot make for them. The split is functional, not departmental.

Trap three: assuming the recomposition is a one-time event. If AI capability keeps moving, the line between measurer and builder moves with it. A role that is a builder today, where the human judgment is the load-bearing piece, can become a measurer in 18 months if the surrounding tooling closes the judgment loop. The plan is not a one-time RIF. The plan is a continuous reassessment of what each role’s irreducible human contribution actually is. This is the same discipline we argued for in the output-competence decoupling piece: the role is not the output, the role is the verified judgment behind the output.

What This Means for the Quarter You Are In

Three things to do this week.

First, get the measurers-versus-builders question on the agenda of your next operating review. Do not assign it to HR. Run it with the executive team. The output is a one-page picture of every role above a certain band, sorted into one column or the other, with a short note for each on what the irreducible human contribution is today and what it might be in 12 months.

Second, before you cut anything, hire one builder for every two measurer roles you are uncertain about. This is the Cloudflare move. Prove the recomposition works at small scale before you make it the layoff story. The companies that get this wrong will cut first and discover the builder bench is empty.

Third, write down, in plain language, what your version of measurers and builders looks like inside your business. Do not adopt Prince’s words wholesale. The categories are useful, the specifics are not transferable. A consulting firm’s measurers look different from a SaaS company’s. A regulated business has measurers it cannot legally cut. Your version of the picture is the work.

The piece that makes Prince’s op-ed historically interesting is not the layoff. It is that a public-company CEO put a name on the operating-model change and asked the rest of the market to look at theirs. The boards and the press are about to oblige. Better to have the picture in your hand when they ask.


This analysis synthesizes How I Choose Which Cloudflare Employees to Replace With AI (Matthew Prince in WSJ Opinion, May 2026; paywalled, open paragraphs verbatim and editorial keywords inferred).

Victorino Group helps leadership teams separate the AI work that builds the company from the AI work that just looks busy. Let’s talk.

All articles on The Thinking Wire are written with the assistance of Anthropic's Opus LLM. Each piece goes through multi-agent research to verify facts and surface contradictions, followed by human review and approval before publication. If you find any inaccurate information or wish to contact our editorial team, please reach out at editorial@victorinollc.com . About The Thinking Wire →

If this resonates, let's talk

We help companies implement AI without losing control.

Schedule a Conversation