They have purposefully avoided large base models and decided to focus on the specific use cases.

The team has plotted all the professions on a graph. A complexity/level of qualification on one axis, available specialist supply on the other.

Typically the more qualified the role, the less supply of it is available.

Hence, they ended up with something like this.

If you look at the bottom right corner - you’ll notice lawyers/judges, doctors (neurosurgenos, cardiologists), investment bankers and psychiatrists.

Operating under the assumption that the highest value upside comes from supplementing those roles with AI - they have purposefully invested into two companies.

First one is Harvey AI, second one is Ambience AI.

When I looked deeper into the case studies behind both - I was fascinated.

On the surface, both are just LLM-wrappers, but at the core are stellar examples of a new product management paradigm.

The one I call “build together with your customer”.

In the past you needed teams to do research, process inputs and produce a prototype. That was the “customer-centricity era”.

Now you can vibe-code tailored bespoke user prototypes in minutes. Adapting the product to the needs of your users as you go. I’d label this a “customer-driven development” era.

So I decided to dig deeper. What does “building with a customer” actually mean in case of Harvey and Ambience? How they acquire customers, how they activate and retain them, what are their pricing and growth strategies.

The full PLG-shebang (with a catch that both companies are sales-led).

Buckle up. It’s going to be fun.

🧬 Article Structure

1. The Legal/Medical Markets. What’s the TAM and the growth projections up to 2030s.

2. Key market players. Looking at Harvey, Ambience - their evaluations, ARR and key inflection points.

3. 🔒 How they grow. A deep-dive into each company growth playbooks. Dissecting all the five layers of the product-led growth (Acquisition, Activation, Retention, Pricing and Referrals/virality).

4. 🔒 Business Model. Sizing up their revenues, costs and estimating profitability.

1️⃣ The Market

How do you even estimate the market that wasn’t there a few years ago?

Well, you go top-down and draw analogies.

Legal Services

For instance, the global legal market today is valued at ~$800b with ~1-1.5m active firms.

Harvey is worth $5b (according to June data), does $75m ARR and services 337 legal firms in US and UK. Share-wise, they are barely scratching the surface. Hence, the multiples on their valuation.

Conservatively, the current day market is estimated as $1.5b and projected to grow to ~$10b with 17-28% CAGR. The downside is legal company conservatism. Many partner firms are aged, operate for dozens (if not hundreds) of years and have deeply-rooted operating principles.

Anyways, even with $10b we’re looking at a meager 1.5% penetration. This is way too conservative for my taste. As a benchmark, it took Amazon 5 years to get to 6-7% of e-commerce penetration in US.

It’s a bullish market with plenty of opportunities.

Healthcare

Healthcare is roughly $9.3 trillion 🤯 global market (number includes both private and public spending). That’s roughly 200k+ hospitals and a numerous amount of various health centers (incl. dental offices) across the world.

As there are dozens of types of doctors, services and jobs - assuming a total penetration is a speculative generalization. But even if we go for a ballpark 1.5%, we’re going to end up with a $140b global market.

Just the AI transcription market alone is assumed to grow to $7b by 2030 with a CAGR of 16-24%.

Ambience AI is currently valued at around $400m (with $35m ARR) and is barely scratching the surface of what’s possible.

2️⃣ Key Players

Before diving into the numbers let me explain for non-legal/non-medical readers what those companies do exactly.

Harvey is a b2b SaaS platform that supplements legal workflows and speeds up legal document drafting, analysis and problem solving. They use an OpenAI and Claude LLMs as well as their proprietary models that were fine-tuned on massive legal datasets for added accuracy.

Ambience is a b2b SaaS platform as well and their core proposition is called Scribe AI. In very simple terms it’s a tool (or a set of tools) that listens to the doctors and their patients, drafts notes and makes medical assessments after each session.

It is deeply integrated into medical databases and fetches user history (using RAG-like tools) to write evaluations. Do not mistake this as pure note taking. Those evaluations are later used for coding and billing customers and insurance companies.

Harvey

Harvey has started at point zero in 2022, grew to $10m ARR in 2023 and they skyrocketed to $50m in 2024.

In 2022 they got their initial $5m funding and started building an OpenAI-based copilot for legal teams. Most of the year was spend drilling into the use cases and prototyping.

A year after, they were able to secure a partnership with a top-tier legal company (e.g. Allen & Overy’s). As their CEO Winston Weinberg states “securing partnerships with key legal firms open doors, because the market is trust-based”.

This has coincided with two Series A and B funding rounds, totaling $103m.

Funding has allowed them to hire top legal talent (from firms like White & Case, Latham & Watkins, Skadden e.t.c.). Some sources claim that, Harvey has employed up to ~80 ex-legal advisors, effectively embedding them both into their product (aka “design partners”) and GTM processes. Smart.

All of those factor combined with a funding momentum - allowed them to scale from 37 to 240 customers in a year and $50m ARR (almost x5 over the previous year).

Ambience

Ambience AI significant inflection point kicked in between 2024-2025.

There were two key factors. First one, they were able to secure an exclusive five-year deal contract with Cleveland Clinic, covering 300 different medical specialists.

This happened after a rigorous AB-testing trial, where their Scribe AI tool has shown 95%+ accuracy is patient assessment and x2-3 ROI on the time saved by each clinician.

Contract attracted attention and funding, Kleiner Perkins (mentioned above) alongside the OpenAI fund have co-invested a total of $70m into Ambience.

Key take-aways

There are two key take-aways here:

• both companies spend at least a year fine-tuning their products for very specific legal and medical use cases;

• both has secured partnerships with high-profile trusted clients, attracting other customers and funding.

Although it’s not disclosed, but I would not be surprised if Ambience went the same path as Harvey and hired medical workers as a part of their GTM strategy.

3️⃣ How They Grow

Read more

The AI hype-train has ignited the no-code/low-code platform market, sending valuations sky-high.

Cursor is reportedly worth $9.9B, Windsurf nearly sold to OpenAI for $3B before talks collapsed, and Lovable’s latest $150M raise values it at $1.8B valuation.

This is truly unprecedented.

Six decades ago, reaching a $1 billion valuation typically took around 15 years - and only after an IPO.

By the 2020s the average tech start-up needs about six years to reach that bar, while AI-driven companies can do it in just three - all through private funding rather than public markets.

Yes, AI is obviously an accelerator. But as with every new, shiny trend, there’s blind optimism in extreme returns (read as — burning cash).

In this article, I’ll dissect both the market and the core players: how they grow (the full PLG breakdown), what’s their business model and how sustainable it is in the long run.

This is a deeply researched post (40+ sources) and your definitive source on the market and the players.

🧬 Article Structure

1. The No-Code / Low-Code Market. What’s the TAM and the growth projections up to 2030s.

2. Key market players. Looking at Cursor, Windsurf and Lovable - evaluation, ARRs, key inflection points.

3. 🔒 How they grow. A deep-dive into each company growth playbooks. Dissecting all the five layers of the product-led growth (Acquisition, Activation, Retention, Pricing and Referrals).

4. 🔒 Business Model. Sizing up their revenues, costs and estimating profitability.

Total words: 3070, reading time: ~25 minutes

1️⃣ No-Code / Low-Code Market

When you think about the NC/LC, it’s useful to draw analogies.

IDE Market

What was the IDE market like in the pre-Cursor era?

Back in the 2000s, it was a niche market valued at < $1 B. Most coding platforms were bundled into the operating systems (Microsoft Visual Studio with Windows, or Xcode with macOS).

With the emergence of mobile and web apps and open-source IDEs, demand spiked, driving the market up to $3.7–5.4 B (~17–20 % CAGR between 2010–2020). Tools like Eclipse and VS Code started becoming more popular.

Now, with AI as an accelerating factor, platforms like Cursor and Windsurf have broken out with a combined valuation of $12.9 B, essentially doubling the market.

This was made possible because of two key factors:

• Novel recurrent monetization of software engineers (who are eager to pay the cost for the productivity gains);

• Breaking into fringe user segments (vibe coders or non-engineers tinkering around with coding).

Web/App Development

This market isn’t entirely new. Freelance or studio web development was a thing dating to the late 1990s. Back in the day, a single bespoke portfolio web project could easily cost you $5–50 k. Hence, the entire market was lucratively estimated at $10–15 B.

This all changed when companies like Wix (founded in 2006), WordPress (2003), and Weebly (2006) started eating into it. Now you can develop a similar portfolio web project for a fraction of the price and time—about $20—and do so in a few days.

Despite their growth, the whole website-builders market is sized at $2.5–2.8 B, which is at best a 5 % share of the larger pie for app and web development.

Fast-forward to 2025: players like Lovable, Base44 (purchased by Wix for $80 M), and Orchid (recent Y Combinator graduate) come onto the stage. Lovable is already estimated at ~65 % of the entire website-building market.

What all of these platforms are doing is stretching the boundaries of what was once possible and breaking into novel user segments.

If we stretch it to the extreme, any doctor or lawyer could potentially become a software engineer. Instead of hiring a company, they can now build their own appointment-booking systems that are tailored to their needs and have great UX—all without any coding knowledge or even understanding of the underlying concepts.

The barriers to entry for people without expertise have never been so low in history, and this is just a starting point.

The low-code/no-code market is already estimated at $45 B, and some optimistic analysts claim it’s going to grow to $197 B. My personal view is that those numbers are not over-exaggerated.

2️⃣ Key Players

For the purposes of the current analysis, I’ve taken the three core players that have publicly available information on their growth numbers and speculative data on their costs—Cursor, Windsurf, and Lovable.

I’m confident that the number of visible players will increase x10 within the upcoming year, but more on that later.

Crazy ARRs

Cursor hit $500M ARR in 36 months after launch, Windsurf $100M in 18m, Lovable $75M in 24m.

The inflection point for Cursor was at 24 months. They have moved into a larger enterprise segment and doubled their sales headcount (getting from ~8.3m/month to 25m/month revenue). By month 36 the canonical hockey stick curve has kicked in.

For Windsurf the inflection point was at month 12. It was driven by their enterprise sales strategy as well, but with a different twist. They have increased their distribution network via the channel partner program (e.g. AWS Parter Network, Value Added Resellers, Managed Service Providers e.t.c.).

Lovable has invested zero into sales (more on that later) and grew organically. The inflection point for them came around ~12 months after launch and was a combined effect of LLMs getting better at coding (Release of a novel Claude Opus) and their no-cost social media efforts.

I’d speculate that for both Lovable and Windsurf the hockey stick curve is yet to come and we’re to see at least a x10 uptick in growth.

3️⃣ How they grow

Read more

It’s a result of a set of well-planned and executed efforts.

This set of efforts goes under the universally accepted codename “growth playbook”.

The most successful companies I’ve worked at were very intentional about those playbooks. They thought them through and continuously perfected them with ongoing experimentation. Here’s the ROI, here’s the payback. This lever we work with, this one we discard… rinse and repeat.

I’ve seen how ecomm and mobility marketplaces grow and their “secret sauce recipe” behind it.

However, I wanted to take a broader look at how different industries approach growth.

I’ve looked into 30+ growth playbooks of companies ranging from SaaS (Canva) to Platform (Nvidia), to uncover common patterns.

These materials were scattered across various articles and podcasts, so compiling them required some exploratory work.

What surprised me most was that while playbooks differ by industry, the fundamental principles within each market are largely the same.

Understanding these principles and applying them to your emerging business can dramatically increase your chances of success.

Let’s dive in.

Today’s article

1. Growth Playbooks. Tested growth tactics divided by industry (SaaS, Marketplaces, Platforms, FinTech, Utilities)

2. Core Principles. Core growth principles applicable to any company at any stage.

3. TL;DR. Key take-aways from the article for people on the go (2 minute read)

+ A bonus → An AI podcast of the article for those who’d prefer listening instead of reading.

📝 Word count: 3563 words
⏱️ Reading time: ~27 minutes

🌱 Growth playbooks

A few key concepts before we get started.

Inflection points

Inflection points are actions that solve the critical bottleneck needed for company growth at a specific point in time.

For instance, if a startup is heavily acquiring users, but leaking them at a mid or bottom of the funnel, these are critical bottlenecks. Similarly, resolving a supply constraint during the early stages of marketplace growth can be a crucial bottleneck.

Successfully overcoming these challenges unlocks an upward growth trajectory - this is what defines an inflection point.

Growth playbooks should be revisited and updated with each new inflection point.

PLG/SLG

In essence product-led growth means that the value of a product attracts user acquisition at close to zero cost.

Sales-led growth requires proactive sales and account management to acquire users.

There’s a widely circulating graphic of PLG iceberg with 8 layers. But in essence it means that your product should be so simple and straightforward that it sells itself. The problem, the pricing and the learning are so effortless that an arbitrary user can figure it out and start getting value out of it.

Chosing PLG over SLG or a combination of thereof is definitely a part of the company’s growth playbook.

Beachhead segment

In simple terms, always start with a specific niche for your target users.

The more specific the better. It’s labelled “beachhead” as a wartime analogy from US troop invasion into Normandy.

1️⃣ SaaS (Utilities)

⚡ Inflection points

Companies like Canva, Dovetail, Miro, Notion and others all have similarities.

Read more

When you objectively look at their product and the value proposition it’s literally paper thin (pun intended). It’s an outlandishly expensive tablet (full package with folio goes for €677) that performs only a single function - taking notes. Everything else is either missing (backlight), clunky (reading e-books) or comes under a paid subscription (basic sync with your laptop).

How on earth do those guys sell 500k of those devices a year and keep on growing? What’s the secret sauce there? Let’s figure it out.

📊 Market

Depending on the lens reMarkable falls both in the E-Ink tablet (€2.6b in 2024) and E-reader ones (€7.0b in 2024). Since e-ink tablet market is pretty noisy and includes the hardware produced for other purposes (e.g. bus stop signs), let’s stick with the e-reader ones.

Market is forecasted to grow to €10b, but still it’s moderately sized and the growth rates aren’t extraordinary (6.3% CAGR).

Additionally, the market is pretty saturated. You have large players like Amazon Kindle (selling up to €400m of Kindles a year) and a long-tail of other smaller businesses like Kobo, Boox, Pocketbook (the long list of obscure names goes on).

The full prices vary from ~€370 to €50 depending on the screen size and functionality.

From this perspective, reMarkable takes roughly ~4% of the market share, has the worst price tag and subpar functionality. Crappy place to be in, but those folks still grow. How?

🎯 Positioning

Ok, e-reader market is noisy and overloaded with complex functionality. How do you disrupt it?

Folks at reMarkable identified a very specific niche. We won’t be just another dull e-book reader, we’ll become a note-taking device. While other companies offered swiss-army knife products, those guys created differentiation. Additionally, there’s focus. We’ll be extremely good at note-taking at the cost of everything else (fancy functions, book reading, even customer support e.t.c.).

So you get a novel niche (a micro blue ocean) with a focus on excellence in this niche.

💎 Value prop

💰 Price

You might say that €677 for a note-taking device is outlandish. It’s true, that’s a huge premium for such a marginal utility. You can make notes on paper at a fraction of the price (even the fanciest Moleskine is going to cost /20 less than the price).

But there are two factors that work in reMarkable’s favor. First, since it’s a micro blue ocean of a market, there’s literally no price anchor (nothing to compare this device with). Second, it’s an object of desire (more on that later).

🛍️ Convenience

If you browse over the Reddit threads, people are claiming that it’s much more convenient to carry a single reMarkable instead of countless notepads.

But is that really so or is there a bias?

The average Moleskine weighs ~180 grams while the weight of reMarkable 2 tablet is ~403 grams (almost half a kilo). With the Folio, the weight jumps up to 878 grams, which is getting closer to a laptop and 22% higher than an iPad (685 grams).

Another argument would be that you can store endless notepads in your memory and open them anytime you like. My counter-argument here would be that people on average open ~5 of their last written notes and hopelessly forget about the ones made long ago. It’s a nice theoretical argument, which doesn’t really stand ground.

Thickness is where it beats the analog-world alternatives - with 4.7mm versus ~10+ mm of Moleskine.

All things considered, it’s not dramatically more convenient to carry and use reMarkable instead of an analog notepad. So I’m biasing that there’s a bias towards emotions.

❤️ Emotions

reMarkable is definitely an object of desire.

There are two major factors contributing to this. First, it delivers an experience with its sleek design and writing capabilities. Secondly, the company employs trendsetters in select local groups (e.g. writers, tech influencers e.t.c.) to cultivate that social proof and exploit our motor neurons.

Experience.

reMarkable stands out among a vast array of unimpressive e-book readers. Design is minimal and sleek. Even the cases have a premium vibe. There’s really nothing to compare this device to in terms of aesthetics.

Writing and sketching experience is where the team has invested a lot of focus. The response rates are the fastest and even the sound the e-marker makes resembles that of an analog pencil.

The team has over-invested in experience and even employs two Design VPs - one responsible for Visuals (Didrik Rasmussen) and the other for UX Applications (Brynjar Barkarson).

Social proofing.

The company invests to build social proof through targeted communities. With novelists (Margaret Atwood), personal growth trainers (Guy Kawasaki), techies (Tim Kendall), and even Olympic medalists on board (Brandi Chastain), the team purposefully aims to influence specific communities.

Those influencers create a perception of higher status, efficiency, focus, and achievement. Everything you can be if only you have purchased a reMarkable.

💰 Financials

Are those guys profitable? Sustainable in terms of business?

📈 Funding and EV

Up to date those guys have secured 24m+ in disclosed funding and recently took on a €43m debt with 11.7% interest (in Oct 2023) (assuming a 15-year timeframe, it’s €6m principal+interest payouts annually).

It’s a smart move as VC money is expensive (considering their current valuation of €1b with revenue of €244m in 2022). As the CFO statement goes this money will be channeled into US and UK expansion.

💵 Revenue and profits

Read more