Anthropic’s ARR grew from $9 billion in late 2025 to $30 billion as of April 2026. Claude Code reportedly represents $2.5 billion of that, pretty wild for a product released only a year ago. While initially built for developers, GTM operators without formal coding skills are now among the most enthusiastic Claude Code users out there.

To understand what’s going on I asked readers: How are you using Claude Code (and Cowork) for GTM?

Maja Voje from GTM Strategist and I surveyed 200 GTM operators who’ve used Claude Code or its sister product Cowork. The survey was conducted in March-April 2026. This report wouldn’t be possible without the help of readers — THANK YOU for contributing!

Today I’ll share the six most interesting takeaways that stood out to me. (For premium subscribers, keep reading to get the full 37-page PDF report.)

Claude Code is among the fastest-growing tech products in history. Cowork, which was released in January, is very quickly catching up.

The GTM operators we surveyed (admittedly an early adopter crowd) said they’re split between whether their primary Claude product is Chat (30%), Code (31%), or Cowork (32%). The remainder primarily use another AI tool.

Maja’s shorthand: think in Chat, build in Code, and operate in Cowork. A cheat sheet for when to use each product:

Claude products pull from the same underlying setup and context engineering. This means Cowork is an easier starting point because it lives on your desktop rather than requiring you to use a terminal. Things like tool connections (via MCP), projects, skills, and file management can get built in Cowork, then inherited by Code for those who need extra power.

Claude Code and Cowork users rave about their experience. Literally. 92% say these products saved them time (that’s expected). Fully 67% said the products enabled something previously impossible. Meanwhile, 55% said they’ve already replaced a tool or vendor because of it – normally ChatGPT, an agency, or a contractor.

What’s so exciting about Claude Code and Cowork is that they create a system for AI-native work where it finally feels like investing in AI will create compounding advantages. It also feels like AI works with you rather than replacing you – each Claude setup benefits from your specific knowledge, preferences, and style (although that begs the question about who ‘owns’ this if and when you change jobs.)

With the right setup, Claude knows your business intimately, (almost) everything can live in one place, AI is trained on your exact work preferences and specifications, and it increasingly has access to the 3rd party systems it needs. Intelligence compounds as Claude keeps learning.

The minimum viable setup includes detailed company context (via a CLAUDE.md file and files that explain your ICP, personas, positioning, competitors, etc.), custom skill files, and tool connections.

More sophisticated users also adopt:

Setup pro-tips from serious Claude users (Maja has a great step-by-step post, too):

Code and Cowork empower the GTM generalist. The average GTM operator adopts 3.5 different use cases including productivity (81%), content creation (69%), product marketing (64%), growth marketing (56%), and GTM engines and prospecting (54%). The most untapped use cases are in lifecycle marketing (32% adoption) and selling (23% adoption).

We then asked which use case has had the biggest GTM impact.

From there we drilled down into specific GTM workflows, asking about more than 40 tangible ways of adopting Code and Cowork. The most common were creating social media posts (#1), building websites, landing pages or microsites (#2), building tools for SDRs (#3), adapting messaging by segment (#4), and building structured content pipelines (#5).

The most universal Claude frustration is around credits and costs, which was mentioned by roughly half of respondents in an open-ended field. This includes running out of credits, getting throttled, hitting usage limits (even on the Max plan), and getting blocked on a project until credits reset. In the words of one GTM operator, “I'm on a Max plan and I still run into model usage restrictions 2-3x per day.”

Here are five ways to save credits and reduce costs according to serious Code and Cowork users:

Beyond credits, users get frustrated by memory and context limits, integration gaps, inconsistent results, and difficulties around managing workflows.

There are challenges with sharing context across a team or organization as well. “It’s not easy to share across teams,” a GTM operator told us. “If a coworker comes up with a skill that's useful, we'll have to build the skills separately rather than sharing.” Another said, “If you update the skill, you have to pray your colleagues are actually going and updating it, too.”

One user’s complaint isn’t easily solvable, however: “The human necessities for sleep, meals and toilet breaks.” 😂

So far 27% of Code and Cowork users say it has already replaced an existing GTM tool. Another 30% said while it hasn’t replaced a tool yet, it will soon.

But there wasn’t much consensus around which tools might get replaced aside from perhaps simple automation or enrichment tools, ChatGPT, spreadsheets, analytics tools (ex: Tableau). Contractor and agency spend seems to be on the chopping block before tools themselves.

Meanwhile, for many users the power of Claude is that it connects to their existing tech stack – making some GTM tools even more useful. The GTM tools that Code and Cowork users love the most are:

Premium subscribers get special access to the full 37-page PDF report. This includes more survey data and insights on Claude Code (and Cowork) adoption, how GTM teams use Claude, limitations to be aware of, and advice to get the most out of Claude.