I am not long or short Duolingo.

TLDR: The strength of Duolingo lies in the whole product offering that creates a sticky platform and happy users, this is not easily replicated by companies like Google simply because they'd need to invest billions to compete. Replicating a function like translation, is not the same as replicating a system which would be a bundle of features, not just translation.

So let's complete this case study between Duolingo and Google with focus from the product offering of both companies as it relates to online education.

One of the apps that Google recently launched is called "Little Language Lessons" and this is basically what the bears are talking about when they say Duolingo doesn't have a moat: any company can integrate LLM to build a new Duolingo, from scratch.

What bear thesis is true?

Google AI, Gemini, can replicate the same function as Duolingo. This is done through Google's Little Language Lesson's ("LLL") product launch.

It's an experimental set of small AI tools for situational language practice.

What does that even mean?

If you want to learn Spanish, type in "ordering coffee in a cafe" as a prompt and Gemini proceeds to provide you with vocabulary, tips and phrases to learn as if you're in a cafe. You log in, you try to remember the sentence construction and words, and you're done.

Where the bear case is weak

"Google can just build Duolingo from scratch", or even worse, "any company can build a Duolingo"

But I don't believe that to be the case from a business perspective.

• Duolingo's Structured Progression vs. Google's Ad-Hoc Queries:

Duolingo is a system, with a whole product offering that all work together seamlessly. It provides a full curriculum with leveled skills, adaptive paths, and long-term tracking (daily goals, unit completion, and multi-year streaks), this enables it to be a habit-forming app, key for product stickiness.

In contrast, LLL relies on adhoc prompts for tools like Tiny Lesson (situation-specific phrases)

It does not provide a cohesive and systemic progression

• Gamification and Retention Mechanisms vs. Minimal Engagement Hooks:

Duolingo's ecosystem builds STICKINESS through gamified elements like leagues, badges, friend challenges, motivational reminders, and interactive storylines which encourages daily return and social competition.

Little Language Lessons offers no such ongoing incentives, and functions more as quick-reference app without streaks or community-driven motivation

• Ecosystem Integrations and Multi-Subject Expansion vs. Isolated Tools:

Duolingo delivers a whole product with integrations like LinkedIn score exports, school programs, English proficiency testing, and extensions into music/math/chess, creating a versatile platform for real-world application and credentialing.

Little Language Lessons remains limited to Gemini's core multimodal capabilities, without broader ecosystem ties or non-language subjects, making it a supplementary tool rather than a standalone solution.

In other words, it's not something you'd get attached to, because Google hasn't even developed it for stickiness.

Also consider that even without LLM's, any company could have launched a meaningful competitor against Duolingo in previous years. But they'd need both the engineers, product leaders, scientists and funding to compete. So the issue isn't exactly about the "LLM replacing Duolingo as a product", but the issue is more about can Duolingo's whole product offering, be compelling enough for users that they don't just use it for adhoc learning, but rather continous usage.

An easy counter to my above point is that Google can expand its product offering to create a systematic ecosystem like Duolingo.

Sure, but consider the CapEx, investment and how non-traditional this will be for Google.

It already has multiple failed experiments ranging from Google Glass(remember?) to Social media so competing from this angle risks losing the core competency race for Google. They don't need to do this, because the ROI for a multi trillion company is low

Add to all of this, Duolingo sits on $1B in net-cash after accounting for debt, 40% net-income margin, PE of 11, and all the while being down 70% in past one year.

I write all of this because Duolingo to me seemed ridiculous at >$150 share price, but now, attractive.