The short answer: on its own, not fully. But Duolingo is doing more than most people give it credit for.
It builds real vocabulary. Real grammar recognition. And a daily practice habit most learners struggle to maintain any other way. What it has a much harder time building is the one thing most people actually mean by “fluent,” which is the ability to hold a real, unscripted conversation with another person.
That gap is worth understanding properly, because it changes what the app is genuinely good for and where a real person still has to be involved.
Does Duolingo Actually Work?
What the Research Shows
Yes, for what it is designed to do. And there is real evidence behind that, not just marketing.
A 2026 study published in Studies in Second Language Acquisition by researchers at Georgia State University, Université du Québec à Montréal, and Victoria University of Wellington ran the comparison properly. Kim, Payant, Skalicky, and Namkung tracked 183 beginner French learners over 16 weeks across three conditions: classroom only (58 learners), Duolingo only (65), and classroom plus Duolingo (60). They tested proficiency, grammar, vocabulary, pragmatics, and communicative competence at the start and end.
All three groups improved by similar amounts on nearly every measure. The Duolingo-only learners developed communication skills comparable to classroom instruction after one semester.
There was one exception, and it is a telling one. The groups diverged on pragmatics, specifically the tu versus vous distinction, which is the choice between the informal and formal “you” in French. That is not a vocabulary problem or a grammar problem. It is a question of reading a social situation and judging how to address someone in it, which is exactly the kind of judgment that gets built through interaction with people rather than through exercises.
Separately, a study by González-Fernández and de la Viña, listed on Duolingo’s own efficacy page, found that after one semester of English study, Duolingo learners scored higher on proficiency tests than classroom learners. Worth noting that this one appears in Duolingo’s own research collection, unlike the independent university study above.
Where the app consistently performs well:
- Vocabulary building
- Reading recognition
- Grammar pattern recognition
- Building a consistent daily habit
That last one matters more than people assume. Showing up for ten minutes every day, for months, is closer to the actual engine of language acquisition than any single clever method. Duolingo solved the hardest problem in language learning, which is getting someone to come back tomorrow.
What "Fluency" Actually Means
Duolingo’s own team has said plainly that fluency is a vague goal. Language experts tend to avoid the word, because it means something different to everyone using it.
Instead, proficiency gets measured on the CEFR framework, an international standard. This is where a lot of older advice about Duolingo is now out of date. The app used to stop most learners around A2. It has since expanded its major courses, and English, Spanish, and French now carry content all the way to the top of B2.
But course content and conversational ability are two different things, and the table below separates them.
| CEFR level | What it looks like in real life | Does Duolingo teach it? | Can you get there with the app alone? |
|---|---|---|---|
| A1 Beginner | Greetings, basic phrases, ordering food | Yes | Yes |
| A2 Elementary | Simple everyday exchanges on familiar topics | Yes | Yes |
| B1 Intermediate | Handling travel situations, describing experiences | Yes in major courses | Mostly, with extra listening practice |
| B2 Upper intermediate | Spontaneous conversation with native speakers. This is what most people mean by “fluent” | Yes in English, Spanish and French | Not realistically. Reaching B2 in practice takes sustained exposure to native speech and real conversation |
| C1 to C2 Advanced | Academic and professional command | No | No |
So when someone asks whether Duolingo can make them fluent, the honest answer is: the syllabus now goes that far, but the speaking half of it does not come from the app. You can complete B2 content and still freeze in a real conversation, because completing exercises at a level and performing at that level are separate achievements.
Where Duolingo Hits Its Ceiling
The research points to one consistent gap: real, spontaneous conversation. Speaking and live conversational practice stay the weakest part of app-based learning even when reading, writing, and listening scores climb.
There is a simple reason for this.
Holding a real conversation means negotiating meaning, adjusting mid-sentence, and understanding a native speaker who has no intention of slowing down for you. That requires practice with unpredictable, live input. An app can quiz vocabulary and grammar extremely well. It cannot yet replicate a person correcting your pronunciation in the moment, or a conversation that goes somewhere neither of you planned.
Is Duolingo Effective for Kids?
This is where the answer shifts, and where most reviews stop short.
The gamification that makes Duolingo work for adults works even better on children. Streaks, points, and characters hold attention in a way worksheets never have. For a child building first vocabulary in a new language, that is genuinely valuable.
But three things change with a younger learner:
- The app cannot tell you if your child is actually progressing. It tracks minutes and lessons completed. It does not tell you whether your eight-year-old can answer a simple question out loud without freezing.
- Children need output far earlier than adults do. Adult learners can quietly absorb input for months and convert it later. Children build confidence by being heard and corrected in real time, and confidence is most of the battle at that age.
- School language expectations are specific. If your child is being assessed on reading comprehension in a second language, general app vocabulary will not map neatly onto what the classroom is asking for. That gap is usually where a Reading Tutor working in both languages makes the difference.
None of that makes the app a bad choice for kids. It makes it an incomplete one.
How to Test Where Your Child Actually Is
The streak number tells you your child has been consistent. It does not tell you what they can actually do. These three checks take about five minutes at the kitchen table and will tell you more than any in-app progress screen.
1. Ask an open question they have not rehearsed. Not “how do you say dog,” which tests recall. Ask something real: what did you do at the weekend, what is your favourite food and why. Watch for the pause. A short delay is fine. A long silence, or an answer that arrives in English first and gets translated, means recognition is running well ahead of production.
2. Play audio at normal speed. Find a clip of a native speaker talking at natural pace, not a language-learning recording. A children’s cartoon in the target language works well. Ask your child to tell you roughly what happened. App audio is slowed and enunciated, so this is often where the gap shows up most clearly.
3. Ask them to correct you. Say something slightly wrong on purpose and see if they catch it. Noticing errors requires a different kind of knowledge than passing a multiple-choice exercise, and it is a good sign of real internal grammar.
None of this means the app failed. It is exactly what the research predicts: strong recognition, weak production. The streak was never designed to build the second one.
The encouraging part is that learners in this position usually move quickly once live practice starts, because the vocabulary is already sitting there. It has just never been spoken out loud to another person.
Duolingo vs a Tutor vs Both
| Duolingo alone | Tutor alone | Both together | |
|---|---|---|---|
| Vocabulary building | Strong | Moderate | Strong |
| Grammar recognition | Strong | Strong | Strong |
| Daily consistency | Strong | Depends on homework | Strong |
| Speaking practice | Weak | Strong | Strong |
| Real-time pronunciation correction | Weak | Strong | Strong |
| Understanding fast, natural speech | Weak | Strong | Strong |
| Adapting to how one child learns | None | Strong | Strong |
| Tracking real progress | Minutes only | Skill-based | Skill-based |
| Cost | Free to low | Higher | Higher |
| Realistic ceiling | B2 content, A2 to B1 speaking | B2 and beyond | B2 and beyond, faster |
The pattern is hard to miss. The two tools are strong in almost exactly the places the other is weak.
How to Pair the App With a Tutor
The pairing works best when each side has a defined job instead of both loosely covering everything.
The app owns daily vocabulary and review. Ten to fifteen minutes on the days there is no session.
A bilingual tutor owns everything the app cannot do: conversation, pronunciation, writing, and connecting the language to whatever the child actually cares about.
The loop is what makes it work. A tutor who knows a student is working through the food and travel units can build conversation practice around those exact topics that week. Passive recognition becomes active use inside seven days. Learn it in the app, then use it out loud with a person.
One thing worth avoiding: treating the streak as the goal. A 300-day streak with no speaking practice produces precisely what the research describes. A child’s streak can be thriving while they freeze at a simple question. That is the signal to add live practice, not more app time.
Realistic Expectations by Timeline
Knowing what progress actually looks like keeps anyone from blaming the app, or the tutor, for a timeline that was never realistic.
First few months. Recognition grows much faster than speech. Your child will understand far more than they can say. This is normal and temporary.
Within the first year. Short, real exchanges start happening. Simple questions answered without translating in their head first.
Multi-year. Comfortable conversation on familiar topics. This is true regardless of method. The research on high proficiency consistently describes years of sustained input and practice.
Pairing an app with a tutor does not shortcut that timeline so much as it keeps a learner on it. Either one alone tends to stall out.
So Is Duolingo a Waste of Time?
No, and this is worth saying plainly.
The research supports using it as a genuinely useful supplement. It is well suited to building vocabulary, reinforcing grammar, and creating the daily habit that language learning depends on. It is not well suited, on its own, to building the speaking confidence and real-time comprehension that fluency requires.
A long streak is not wasted effort. It is just solving a different problem than the one most learners eventually run into.
Book a free trial session and we will tell you honestly where your child actually is.
References
Kim, Y., Payant, C., Skalicky, S., and Namkung, Y. (2026). Comparing the effectiveness of Duolingo, classroom instruction, and classroom + Duolingo instruction conditions on beginner-level French language development. Studies in Second Language Acquisition. https://doi.org/10.1017/S0272263126101521. González-Fernández, B., and de la Viña, I., via duolingo.com/efficacy/studies. Duolingo course CEFR expansion: blog.duolingo.com/courses-teach-advanced-content/
Frequently Asked Questions
Can Duolingo make you fluent?
Not on its own. Duolingo's major courses now teach content up to B2, the level most people mean by fluent, and research shows it builds vocabulary, grammar, and consistency effectively. But completing B2 exercises is not the same as holding a B2 conversation. Spoken fluency requires live practice with another person.
Does Duolingo actually work?
Yes, for what it is designed to do. Peer-reviewed studies show real gains in vocabulary, reading, and grammar recognition, in some cases comparable to or better than classroom instruction over one semester. Its documented weak point is spontaneous speaking.
Is Duolingo enough to learn a language?
It depends what you mean by learn. For reading, vocabulary, and grammar, the major courses now cover material up to B2 and can carry a learner a long way alone. For speaking and understanding fast natural speech, it is not enough, because those skills only develop through live, unscripted practice.
How long does it take to get fluent with Duolingo?
There is no reliable timeline, because app-only study generally plateaus before fluency rather than reaching it slowly. Learners pairing daily app practice with regular live conversation typically reach comfortable conversation on familiar topics over several years.
Is Duolingo good for kids?
Yes, as one part of a plan. The gamified format holds children's attention well and builds early vocabulary. It does not measure whether a child can speak confidently, and children need spoken output and correction earlier than adult learners do.
What CEFR level does Duolingo get you to?
Duolingo's English, Spanish, and French courses now carry content to the top of B2, having previously stopped around A2. Course coverage and demonstrated ability differ, though. Learners completing that content typically read and understand at a much higher level than they can speak without live conversation practice alongside it.
Is Duolingo better than a tutor?
They do different jobs. Duolingo is stronger at daily vocabulary drilling and habit-building at low cost. A tutor is stronger at conversation, pronunciation correction, and adapting to how a specific learner works. Learners using both tend to progress faster than either alone.
Why can I understand Spanish but not speak it?
This is the most common outcome of app-only study, and it is expected. Recognition and production are different skills. Apps train recognition heavily and production lightly. Speaking improves through live, unscripted practice where you have to produce language under time pressure.
Does the Duolingo streak actually mean anything?
It measures consistency, which genuinely matters in language learning. It does not measure ability. A long streak with no speaking practice reliably produces strong recognition and weak conversation, so the streak is best treated as a habit tracker rather than a progress score.
Should my child use Duolingo and a tutor at the same time?
That combination works well when each has a clear job. The app handles daily vocabulary review between sessions. The tutor handles speaking, correction, and tying the language to schoolwork or family use. A tutor who knows which units your child is on can build sessions around them.
