Wave AI: Data-Driven ICP for $7M ARR Note-Taking Mobile App

Wave AI, a $7M ARR mobile app by Josh Mohrer, didn't start with a meticulously crafted Ideal Customer Profile (ICP).

It started with a problem: his 70-year old dad, a retired doctor couldn't remember what his own physicians told him.

Patient Zero: An Accidental Discovery at a Doctor's Office unlocked Wave AI's Value

Josh Mohrer's father is a physician—a smart, articulate, medically trained professional. But when his own health issues emerged, he struggled to relay the details of his complex appointments.

"Despite being a retired doctor, he would come back from these very important meetings with his doctors and be sort of not really able to answer our questions in the specificity that we would bring," Josh recalls. "My parents are not losing it; they are smart people... they just had trouble doing that".

Josh handed his dad the prototype Wave app with a simple instruction: record the next appointment and send him the summary.

The result was transformative. Suddenly, complex medication names were captured perfectly. Treatment plans were organized and clear. The information that would have been forgotten or misremembered was now preserved.

That moment crystallized Wave's true value, rooted in Josh's core belief: "life's most important things happen away from your desk". The app wasn't just for office meetings; it was for high-stakes conversations where being fully present matters more than taking notes.

The GPT System That Mapped Wave AI’s ICPs

Instead of guessing who his ideal customers were, Josh let the data reveal them. He developed a clever system to understand how people were actually using Wave.

The Meta-Analysis Strategy:

Josh runs every recording summary through a second AI filter to categorize the content. He explains, "I actually run every summary through a meta filter... it asks the question, 'From these lists of things, what is this?' and [the AI] responds... then I save that into a database".

This simple analysis revealed 3 unexpectedly distinct user segments, each with its own clear patterns.

ICP #1: Business Professionals recording Meetings and In-Person Sales Calls

• Usage Volume: Roughly 50% of all recordings.

• Job to be Done: Capture in-person meetings, sales calls, and client discussions where typing on a laptop feels rude or distracting.

• Defining Characteristic: The app's usage is a mirror of the traditional work week. "We're obviously dead on Saturday, Sunday—like literally -90% of recording volume," Josh notes.

• Psychographic Profile: "I need to be fully present in conversations to build rapport, but I can't afford to miss critical details."

ICP #2: The Church Community recording Sermons

• Job to be Done: Record and archive sermons, Bible studies, and spiritual messages for later reflection.

• A Surprising Discovery: This segment was an unexpected revelation. "The biggest surprise has been that on Sundays, Church dominates Wave usage," Josh says. "I don't know if it's pastors or congregants... but there's major, major Church usage".

• Psychographic Profile: "These words are meaningful, and I want a perfect record to study and remember them exactly as they were delivered."

ICP #3: Medical Patients using it for Doctor Visits

• Usage Volume: 5% of recordings, but the source of the most passionate feedback.

• Job to be Done: Navigate complex medical information during appointments and easily share accurate updates with family members.

• The Origin Story: This was the original inspiration for Wave and what Josh calls a "real breaking point" in realizing the app's potential.

• Psychographic Profile: "I'm overwhelmed with information and stress. I need a tool to ensure I understand my treatment and can explain it clearly to my loved ones."

Feeding Meta Ads the Right Signals: Clean Data, Low CAC

These distinct ICPs are the fuel for Meta's powerful ad algorithm. When Josh sends a purchase event to Meta, he's providing a rich signal. The algorithm doesn't just see "a sale"; it sees a pattern.

Josh doesn't need complex ad funnels. He provides clean conversion data from distinct user groups, and the algorithm optimizes delivery to each segment independently. He calls Meta's ad engine "the best version of that that humanity has ever known" because it is "able to find my guy and my gal" with stunning efficiency.

This is why his $30 Customer Acquisition Cost (CAC) is profitable. Meta isn't targeting blindly; it's delivering pre-qualified users who already have the exact problem Wave solves.

The lesson from Wave's success is clear: Stop guessing who your customers are. Build something genuinely useful, listen to the data, and let your users tell you exactly who they are and what they need.

Top Tweets of the day

1/

I don't think the 2-3 year timeline is accurate but both genders have mostly the same needs. And it makes sense how something useful for one gender can be useful for another one.

Skincare, for example, were only female-focused until everyone saw Korean males. And plastic surgery was also mostly done by females unless you were a celebrity. Now even males do it. Again, Korea is #1 at plastic surgery.

This phenomenon has a useful parallel to ages too. Earlier, only younger people were focused on mental exercises (puzzles for brain building) and physical exercises (gym for muscle building) but they are even more useful for older people. For example, doing puzzles reduce chances of Alzheimer's or some disease that makes you forget stuff. I read some book that Japenese men do puzzles and that's why they have less ratio of Alzheimer's in older people. Early research also shows it to be true for muscle building. Older people benefit more from muscle building but it's still early days so there isn't more noise on socials.

2/

TurboNote's TikTok strategy is epic.

3/

Sora 2 effect in action.

1 day to add $4K MRR & 1M+ views by copying competitor's playbook. The irony is that the competitor hired real actors, so it probably cost exponentially more.

Rabbit Holes

1. How to validate your startup idea (10k MRR) by Founder's Cafe

2. How to have great taste by Max Stoiber

3. Forbes Marketplace: The Parasite SEO Company Trying to Devour Its Host by Lars Lofgren

What’d ya think of today’s newsletter? Hit ‘reply’ and let me know.

Do me a favor and share it in your company's Slack #marketing channel.

First time? Subscribe.

Follow me on X.

More Startup Spells 🪄

1. Dara Denney's YouTube Comment Section Hack (LINK)

2. TikTok ‘Axes of Virality’ Framework Behind 950M Faceless Views (LINK)

3. Hampton's AI Report (LINK)

4. Scarcity Marketing Tactics: How Fake Urgency Boosts Conversions (LINK)

5. SetApp Dynamic Pricing: How Browser Fingerprinting Determines Your Subscription Cost (LINK)