Who we are
Duor Digital OÜ is an independent product studio registered in Estonia with a distributed team. We design and build mobile and web products end to end, and we also ship products of our own, so we know what it means to carry something past launch rather than hand it over and leave. The team stays small on purpose: the people you speak with are the people doing the work.
How we would approach the problem
Your survey findings ring true to us, and we would treat the three problems as one design problem rather than three features.
The structured path comes first. A learner should never have to decide what to study next; the app should always know. We would build the 60-lesson course with your curriculum partner as a single visible path, so progress feels like moving along a road rather than ticking through a library.
Speaking in the first week is the objective we would design around most carefully. Beginners avoid speaking because it feels exposed, so we would make the first spoken exercises very short, private, and forgiving, with pronunciation feedback that points at one thing to fix rather than a score. Getting someone to say a sentence out loud on day two matters more than any other metric in the first year.
For the community feature, we would start with light, moderated pairing: short, structured exchanges between learners at similar levels, with clear prompts, so no one is dropped into an open chat with a stranger. We have built pairing and cohort features before and learned that structure, not freedom, is what keeps people coming back.
Two practical points. First, the content system: your team must be able to change a lesson on a Tuesday afternoon without asking us, so we would build and test that with your editors early, not at the end. Second, the hosting budget. Staying under €20,000 a year for 25,000 active learners is realistic if speech processing and AI features are designed with cost in mind from the start, and we would treat that limit as a design constraint, not an afterthought.
We would begin with the user research: at least 40 learners across the launch languages, including people whose first language is one of the others, since a French speaker learning Italian and an English speaker learning Spanish do not need the same first week.
Similar work
Our own product, Qiao, is a mobile-first app for international students moving to China. It ships in English, Russian and Chinese, is built around a single guided path with stages, and pairs newcomers with people who have been through the same thing, so the questions of multilingual onboarding, structured progress and moderated matching are ones we have already worked through in practice.
We also built the AI layer of Healy, a consumer health app whose whole premise is daily motivation: personalised plans, well-timed nudges and progress tracking. Keeping people engaged after the first enthusiastic week is a problem we have spent real time on, and 30-day retention above 35% is the kind of target we are used to being measured against.
Details of both are on our site.
Availability
We can commit a dedicated team for a 6–9-month build starting in February 2027, and we would welcome the chance to submit a full proposal in January.
Kind regards,
Duor Digital OÜ
[email protected] · duor.digital