Speekify – language learning online service

Speekify is an online service for people who want to speak a foreign language fluently and confidently.
The MVP was made by me and two friends of mine, Roman Smirnov and Denis Lukyanov, over three months in the evenings and on weekends. We had a plan to launch it in such a short time and we managed it with a delay of just two weeks. We have every reason to be proud :) So let's talk about the project and my part in it.

My role involved designing the web application and iOS app – everything from UX to visual design, including illustrations, and customer research as well.

Our users are mostly well-educated women accustomed to working with online services. They are familiar with Facebook and Twitter and have used language-learning services like Duolingo or similar.

They don't want to be bored with grammar lessons, have little free time to learn, and need a clear solution that delivers value, feels modern, fun, and joyful.

That's why the firm's style is colorful, clean, and joyful. The app interface contains lots of illustrations, big buttons, and short text captions.

This is my first project fully made with a mobile-first approach, and I really love it. The thing is, when you design mobile-first, you leave only the important stuff on the screen. You don't make sidebars, columns, two-level menus, etc. You just show the user what they need right now and nothing else.

Then, when you migrate the design to desktop, it looks perfect – nothing extra, a lot of space, one CTA button, clear and simple.

After a user signs up, they get a series of lessons. Each lesson contains an original audio track divided into phrases, and the user should repeat it fragment by fragment.

Then they move to the next step, where they make a recording and Speekify automatically reviews it, indicating whether it is similar to the original audio or not.

I divided the user's workflow into 4 steps: listen and repeat; record; send for review; listen side-by-side with the original track. On each step, the user has only one CTA button and can perform only one action – listen, record, or send. I've also made big buttons to make it easy for users to understand what to do next.

A user can return to the previous step to train more or to re-record. But this action is secondary, so the "Back" buttons are less visible than CTA buttons.

The project seems tiny, but my Sketch file contains 56 screens already. And it's only the MVP so far.

When a user sends his recording for review, Speekify estimates it in percentages – how similar the pronunciation is to the original audio track. If the percentage is more than 70%, Speekify gives the user a reward – a medal, cup, or diamond. This feature is very engaging. Users try to improve their recordings to get a diamond.

Estimation icons are a starting point for gamification. I'm planning to add achievements, labels, or other game mechanics to involve users more deeply in the learning process.

From the beginning, I wanted to make Speekify friendly and tell our users that we are close to them and open for communication. So, I used informal texts, emojis, and made quite a lot of illustrations for the application and landing page: intro screens, congrats pages, "enough for today" screens, error 404 pages, etc.

A lot of work, actually :)

Users show a lot of interest, but we have some issues and a huge backlog as well. We need to improve user engagement and retention, set up an onboarding process, solve technical issues, explain training method features, make an iOS app, and so on.

Anyway, I'm extremely happy and excited about the project. It solves real user problems, helps people become more confident and fluent. The app is clean and friendly, and users understand quite well how to use it. We receive vivid feedback and have a lot of fun in the process.