Building A Language Learning Platform Is Cool And All, But So What?

Posted on 3/9/2025

5 minute read

Hello Linguaponearners, we meet again! I hope your 2025 language learning journey goals are going smoothly.

First, I want to mention that we are reaching a critical point in Linguapon's beta program. Linguapon is continuously growing through user feedback, and I'm glad to see how far it has come since the beta launch.

I'm taking a brief but somewhat long pause on the series of 'Dissecting Linguapon' blogs. I was next planning a write-up on the Explore module, but I couldn't help but keep asking myself one question: “so what?“

Let me explain. In the very first blog, I listed the challenges that Linguapon hopes to address. These acted as guiding principles or a vague measurement for how I go about adding more features and quality updates to Linguapon. However, each time, I had not made it a habit to look at Linguapon holistically and ask, “so what?“ As in - what is the true value that I'm delivering to a Linguaponearner like yourself?

I found (or rather, stole) an idea from Reddit that really resonated with me and helped me answer the question of “so what?“ It's called SMART.

SMART:

Specific: Is the user clear on what they have to do or achieve?
Measurable: Is progress toward the goal visible?
Actionable: Are the steps to achieve the goal small and actionable enough?
Relevant: Is there a clear purpose or reason for the goal?
Time-bound: Is the time investment needed for each action clear?

I'm going to be cheeky and add an extra one, making it SMARTR.

Raising the bar: How am I providing more value than competitors? (Who is 'SMARTR' about it?)

Now, I'll attempt an exercise of answering these questions with Linguapon. These answers reflect my vision and my actual thoughts at the time of writing.

Specific:

How Linguapon is set up or marketed today is that users take control of their learning in new vocabulary. To ensure the learning isn't entirely random, Linguapon accepts the user's target language, proficiency level, and a topic of interest as parameters. As a result of the Explore flow, the user is guided through a tutorial that populates these parameters and provides suggestions, allowing them to view vocabulary they might or might not already know.

However, the goal or any form of milestone is not immediately clear. There is nothing the user can work toward or aspire to, other than arbitrarily learning vocabulary. I will frame things in a way that gives the user something to constantly aspire toward, while also making the goal flexible enough to allow users to define what they want to achieve. It will be something along the lines of asking the user how many words they want to learn, with a recommended range of words for each proficiency level.

Measurable:

Viewing proficiency levels is a natural target in itself. We understand terms like “beginner” or “intermediate,” but coupling that with the number of learned words as a target makes it more tangible. As humans, we are more receptive to numbers and find progress toward a non-moving target - a simple, understandable number - to be a dopamine hit.

This is a long-winded way of saying: as long as Linguapon remembers how many words the user has learned and projects this alongside the goal, progress becomes measurable.

Actionable:

I talked about Test Me in the previous blog, but since then, things have slightly evolved based on the specifics we're trying to meet. Test Me plays a crucial role in formalizing the actionable step of marking a word as “learned.” If the user answers questions in Test Me correctly, Linguapon evaluates that the user knows that word and remembers this until the user frequently starts getting questions about the word wrong.

Currently, there is a missing link in providing visibility to the user on how many correctly answered questions are required for the system to treat that word as 'learned.' (By default, new words added to the Translation Bank or first encountered in Test Me require one correct answer. Failing to do so sets the user back and would require more correct answers in subsequent attempts.)

Without this information, it's hard to plan accordingly. If I have 3 minutes of spare time and set an actionable goal of learning 3 or 4 words, I wouldn't know whether I've met this actionable step by the end of my 3-minute session.

Relevant:

Progression in a language is arbitrary. We will present the goal in terms of the number of words learned, but is this truly relevant in helping the user's progression in the language they want to learn?

Flashcard apps are popular because it's easy to sift through hundreds of flashcards that the user thinks they know or don't know. Anki, for example, uses a spaced repetition system that prompts the user to recall information and, based on their response, reschedules the card to reappear at a particular frequency. For language learning, words or phrases the user knows - and therefore marks with a particular response - prompt Anki's algorithm to reschedule them for a much later date.

This is useful in triggering active recall, but the problem is that the underlying nature of a flashcard prompts the user to learn by remembering what exactly is written on the back of the card. This means that no matter how often the card reappears and the user eventually memorizes it, it doesn't translate into a deep understanding of the word. A deep understanding would, for example, enable someone to correctly apply the word in any context, spell and identify it correctly in reading and listening comprehension, and pronounce it accurately.

Linguapon differs in that it doesn't give the user free rein to decide whether they 'know' a word or not. Test Me evaluates a user's understanding of a word from multiple angles through a variety of question types. When a user is presented with a question, active recall can be relied upon in some situations, but the beauty of the system is that if the user couldn't recognize the word or character in the example below, they might remember it based on the sound from the audio output. So if two concepts are simultaneously tested for and called upon to answer the question correctly, this would likely effectively reinforce the user's understanding of the character, through association (and validation) with the sound.

If the user has learned 1000 words via Linguapon, the goal for Linguapon is to ensure those 1000 words were truly learned. If not, is Linguapon truly purposeful and relevant for pushing users toward their goals?

Time-bound:

Users like myself tend to suffer from short attention spans. If I knew how much time investment something would take, I would have clear expectations of how much time I need to achieve something - which ties back to the anecdote in the 'Actionable' section. If I knew each question took about 5 to 10 seconds to answer, then wow - I could answer no more than 36 questions in a 3 minute session!

As I previously mentioned, humans love numbers. Knowing how long or how many days it takes to reach 1000 words is motivating. It almost encourages the user to maintain their momentum and continue using the system!

Raising the Bar:

Linguapon is in need of a couple of things that will drive real value into using the service over competitors. One of them is the goal-based architecture that I mentioned earlier.

Another is something that drove me to delay any deep-dive blogs into the other modules. The knowledge system is particularly new to Linguapon, but it is a crucial component in answering the question of “So what?” I would like this to act more as the bloodstream that runs across the entire system. Knowing exactly what words the user knows and doesn't know is the visibility that is missing across components like Explore and Translation Bank.

I will not stop raising the bar nor stop asking myself, “so what?”

Asking “so what?” is important, as this will guide my approach to everything I put into Linguapon going forward. From product design decisions to the language used in marketing, discussions with users or potential users - the true value of something is very difficult to articulate if you don't know the answer to that question. And if you can't articulate the true value, nobody is going to know about it.

Written by Elvin, Linguapon Admin