Product Strategy Proposal: How We Modeled Anthropic's Enterprise Playbook 18 Months in Advance

If you want to build real product sense, you have to escape tutorial hell.

After 14 years of building products and recently transitioning to full-time solopreneurship, my philosophy has always centered on Applied Practice. You don't learn strategy by reading another framework; you learn it by placing a bet, documenting it, and letting the unforgiving market grade your paper.

In November 2024, nearly two years ago, I put this into practice during the Product Sense & Strategy course by PM Curve. The capstone was a grueling two-phase simulation. Phase 1 was a solo elimination round where we had to pitch a strategic product initiative. Those shortlisted moved to Phase 2, joining a team to build a comprehensive Go-To-Market and product strategy presentation.

We had a few products to pick from for our capstone—Spotify, Anthropic, and a few others. I picked Anthropic.

At the time, the entire landscape of major LLM players was obsessively focused on B2C chat interfaces. But I hypothesized a different path for Anthropic based on their unique safety positioning, the broader market sentiment around AI, and the glaring gap between early consumer adoption and absolute enterprise hesitation.

My Phase 1 pitch got me shortlisted, and I teamed up with a brilliant group of PMs—Pushkar, Ankit, Krishna, and Praharshitha—for Phase 2. We ended up winning the capstone. (You can see the timestamped receipt of our win in my original LinkedIn post here).

Now that it’s July 2026, I wanted to pull these artifacts back out and grade our scorecard against what Anthropic actually shipped.

Phase 1: The Hypothesis (AI Safety & Compliance)

My solo pitch focused on an "AI Safety Toolkit." The premise was simple: enterprises will not adopt LLMs at scale if they fear data leakage, compliance breaches, or autonomous hallucinations. I proposed Prompt Shields (injection detection), Adversarial Testing, and a rigorous Compliance Framework.

📄 Read the Phase 1 Proposal (Opens in new tab)

Phase 2: The Execution & GTM (Claude Studio)

For the final group presentation, we translated that safety moat into a tangible product ecosystem: Claude Studio. We pitched an enterprise hub focused on solving sub-problems impacting productivity, like integration and workflow disruption. Our core bets included:

• Automated Workflows via Computer Use: Delegating repeated work to computer agents.
• Role Groups: Granular access controls mapping to company hierarchies.
• Knowledge Integration: Connecting Claude directly to Confluence, Notion, and Slack securely.

📄 Read the Phase 2 Presentation (Opens in new tab)

The 18-Month Scorecard: Hits and Misses

Looking at Anthropic today, treating this capstone as a live strategic simulator paid off. Here is how our bets panned out:

✅ HIT: Enterprise Productivity & Workflow Adoption
We called out that complex enterprise systems require seamless AI integration, and built our pitch around unlocking productivity. Today, those exact tenets of enterprise workflow adoption are the foundation of the recently released Claude Co-work ecosystem.

✅ HIT: Computer Use & Autonomous Agents
In late 2024, Anthropic had just launched their rudimentary computer use capabilities. We rightly doubled down on it, pitching "Computer use for Automated workflows" to delegate repetitive tasks. Today, that bet on autonomous activity is exactly what Claude Code has become—a massive engine operating independently in the developer and browser environments.

✅ HIT: Professional Services
On our business model slide, we explicitly mapped out "Professional services" as a revenue stream to support Claude Enterprise solutions. Fast forward to today, and the "Forward Deployed Engineer" (FDE) is one of the hottest, most critical roles in the AI space, proving that enterprise AI requires hands-on integration services.

✅ HIT: The Compliance Moat
We pitched a standalone AI Safety Toolkit. Anthropic took it a step further and embedded those exact concepts natively. With their compliance API integrations and governed data partnerships, they brought the models directly to sensitive enterprise data without moving it.

❌ THE MISS: Our Financial Projections
Slide 2 of our deck projected an optimistic 2026 valuation for Anthropic of $40 Billion with a revenue run rate of $1 Billion.
The reality? As of May 2026, Anthropic's revenue hit an astonishing $47 Billion. I am going to go ahead and blame my conservative Indian upbringing for that hilariously safe financial estimate. We nailed the product roadmap, but the sheer velocity of the AI market broke every standard SaaS financial model.

The Takeaway for Product Managers

If you want to hone your strategic skills, stop just reading theory. Pick a product. Write a strategy. Publish it. Then, come back a year or two later and see how the market humbled or validated you.

Want more product teardowns and applied practice frameworks?
Drop your email below, and I’ll send my latest essays and simulator updates straight to your inbox.

Work with me:

I am currently taking on select consulting projects for founders and offering 1:1 mentoring for aspiring/mid-level PMs looking to bring rigorous, applied strategy to their careers. Click here to book a chat.