top of page
Focused Group Discussion.png

Insights & Resources

Articles, insights, and strategies to help founders grow smarter.

Your AI startup hit 2M ARR. So why is your next funding round stalling?

  • Writer: Claudia Crangasu
    Claudia Crangasu
  • Jul 11
  • 2 min read

Updated: Sep 26

ree

You've got early traction. You've proven the tech works. But whether you're preparing for Series A or trying to scale post-Series A, investors keep asking the same question: "Who exactly is your customer?"



The reality? Your ICP evolved faster than your positioning strategy. Following up on my recent article about AI positioning strategy, where I discussed how "AI products don't fail because of poor tech, they fail because users don't understand why they matter".



I just worked with a $2.5M ARR AI company preparing for Series A. They were burning $200K/month with a 6-month sales cycle and inconsistent deal sizes. 

The problem wasn't their product. Their original ICP had evolved, but they were still positioning for tech personas, instead of the decision makers who were actually buying their products.



The pattern I see in AI companies that successfully scale from $1M to $10M ARR:


  • They redefine their ICP every 3-6 months as product capabilities expand

  • They position from customer budget backward, not tech features forward

  • They know exactly who has purchasing authority (it changes as you move upmarket)


Andreessen Horowitz on ICP Misalignment: “A poorly defined ICP is the hidden cause of everything from high CAC and low conversions to bloated product roadmaps.” – Joe Morrissey , a16z Partner .

Sequoia Capital Pat Grady on AI GTM Strategy: “Start with vertical- or function-specific applications. Work from the customer back, not the technology out.” (Grady led Sequoia’s $75M investment in OpenEvidence — now a $1B company.)

The Positioning-ICP connection I've learned: the key isn't replacing humans, it's enabling them to do their best work. But here's the catch: you can't enable someone if you don't know exactly who they are. 47% of failed startups never tracked marketing ROI, they never knew if their ICP assumptions were right*



The Framework That's Working for Seed-Series A Companies

Whether you're raising Series A or scaling post-Series A, here's my 5-step ICP evolution process:


  1. Map your expanded capabilities to new buyer segments: Your AI improved, so did your addressable market size

  2. Interview your highest-value customers : What made them buy? Who else influences their decisions?

  3. Test messaging across different company sizes: Your startup positioning won't work for 500-person companies

  4. Track metrics by customer segment: Where's your highest LTV and fastest sales cycles?

  5. Align your entire GTM on one primary ICP : Chasing everyone means converting no one


In this Wild Wild West of AI, your ICP isn’t just marketing, it’s your survival strategy. The startups hitting unicorn status? They’re absolutely essential to someone specific.

If you're an AI company ($1M-$5M ARR) struggling with:


  • Inconsistent deal sizes and unpredictable revenue

  • "Promising" investor meetings that don't convert to term sheets

  • Marketing that worked 6 months ago now falling flat

  • Difficulty explaining who your "real" customer is


Let's talk. I work with 2 scaling AI companies at a time as their Fractional CMO, specifically helping them nail the positioning that unlocks their next growth phase.



*Sources: 



bottom of page