Does Duolingo actually work for a trip? What two months buys you
Two months of daily Duolingo buys you real things. You will recognise written words, you will catch numbers and place names, and you will be able to open an exchange. What it does not reliably buy you is the second sentence, which is where most actual interactions live. On a seven-day trip that gap is the whole question, because a trip does not test your foundation, it tests roughly forty specific sentences you will say out loud, several of them twice a day.
So does Duolingo work for travel? It works at what it is built and measured for. Duolingo's own 2024 research report concluded that learners of English, Spanish and French with minimal prior proficiency can start a conversation after four to six weeks of app use, at the A1 level of the CEFR scale, based on 140 participants who averaged 90% accuracy across the study's conversation tasks. A1 is the level where you can begin an exchange. It is not the level where you can steer one. The published first-person experiments are useful precisely because they show where the steering fails.
Two months of German, seven days in Germany
The cleanest test is Morgane Croissant's account for Matador Network: fifteen minutes a day, a 70-day streak, then a seven-day trip to the island of Rügen. She came back able to handle bakeries, trains and hotels at the level of a simple request. She also came back with a very precise description of the ceiling. She "could order salad but I could not ask for a specific dressing". She "could ask for water or the bill at a restaurant but I did not know how to say I didn't want the sparkling kind". When she set off, she had just started learning to count to ten.
The detail that should stop you is which phrase earned its keep. The one she used most was "Ich spreche kein Deutsch, sprechen sie Englisch?" Two months of daily practice, and the most valuable sentence in the set was the one that opts out of the language. Her own estimate afterwards was that five to six months would have been closer to the mark.
Ninety days is better, and the seam is in the same place
A second Matador experiment ran French for a 90-day streak and then went to Brussels, and the result was genuinely better. The app's "getting around" material covered hailing a taxi and the prepositions needed to direct someone to a hotel, and four days of trying to stay in French mostly held. The failures were structural rather than lexical: "if I had the vocabulary to express a complex idea, I often lacked the grammatical structure". A museum audio guide landed at about a third comprehensible. Rapid native speech about a local public holiday did not land at all.
Duolingo is not hiding this. Cindy Blanco, a senior learning specialist there, framed the aim as giving you "a really solid foundation that you can take with you anywhere". That is an honest description of a course path. It is also a description of something other than a departure date.
A course and a trip want the phrases in different orders
A course path is sequenced by what teaches well. Simple present before past, regular verbs before irregular, numbers wherever they fit the unit. That is why someone can reach day 70 and still be learning to count to ten. Nothing is broken there. The sequence is just answering a different question than yours.
A trip is sequenced by what you spend first and most. On a week in Portugal you will say some version of "a table for two", "the bill, please", "does this have nuts in it", "which platform", "what time is the last train" and "is there a pharmacy near here" before you say anything in the past tense. Those sentences do not form a grammatical progression. They form an itinerary.
Three specific gaps follow from that:
- Ordering. Trip-critical phrases are scattered through a course rather than front-loaded, so the thing you need on day one may sit behind twenty units.
- Deadline. A streak is designed to run forever. Your flight is on the eleventh. Nothing in the course compresses to meet it, so the review schedule and the calendar never negotiate.
- Out loud. Tapping a tile in the right order is recognition. Saying the allergy line to a waiter who is already walking away is production, and the second one is what the trip grades.
What we built instead
Phrazl starts where the trip starts. Onboarding is three steps: the destination, the departure date, and up to three scenarios from restaurant, transport, hotel, emergency and shopping. From that you get a pre-curated phrase set for that language and scenario combination, ordered by trip-relevance rank rather than by grammatical difficulty, across the first ten languages at launch.
The departure date is not decoration. The scheduler is SM-2 spaced repetition with the intervals compressed against your departure date as a hard deadline, so the review pace tightens as the trip approaches instead of drifting past it. Practice runs in three modes: flashcard, listen-and-repeat with audio playback, and fill-in-the-blank. Progress is a single bar labelled as curriculum coverage completed. No streaks, no XP, and no claim that a percentage predicts how a conversation will go, because coverage is what the bar actually measures.
Two things worth saying plainly before launch. The phrase audio in the first release is a synthesised voice rather than a native-speaker recording, and the app says so on the screens where you hear it. And the emergency scenario carries a disclaimer on the selection screen and in the terms: those phrases are for general communication practice, not a substitute for contacting local emergency services or a qualified interpreter. The first twenty phrases of a trip are free to try, and when payments open a trip pack is a one-time purchase at $9.99 for a single destination with up to three scenarios, or $19.99 for the full bundle, with a returning-traveller discount on a second destination.
When Duolingo is still the right call
If you want the language rather than the trip, keep the course. Six months of it will take you somewhere a phrase set never will, and the published accounts above are evidence for that, not against it. The mismatch only appears when the calendar is short and fixed. Two months of a course path leaves you fluent in the opening line. Two months aimed at your actual itinerary should leave you fluent in the forty sentences the week will ask for, which is a smaller ambition and a much better week.
Is two months of Duolingo enough for a one-week trip?
It is enough to open an exchange and handle a simple request at a bakery, a station or a hotel desk. The published two-month German experiment found the ceiling arrived at the second sentence, with the writer able to order a salad but not ask about the dressing. For a fixed one-week trip, a phrase set aimed at your actual scenarios covers more of what you will say.
Does Duolingo teach you to speak out loud?
It includes speaking exercises, and its own research reports A1 conversation-starting skills after four to six weeks of use. The gap travellers describe is production under pressure: having the words for a complex idea but not the structure to assemble them while someone waits. Saying phrases aloud against a recording, repeatedly, is what closes that.
How many phrases do I actually need for a week abroad?
Fewer than people expect, but they have to be the right ones. A week of restaurants, transport and a hotel runs on a few dozen sentences, most of which you repeat daily. Phrazl builds its sets around the destination and the scenarios you pick, ordered by trip relevance rather than grammatical difficulty.
Can I use both a course app and a trip phrase set?
Yes, and they do different jobs. Keep the course if you want the language long term, since several months of it goes somewhere a phrase set will not. Use a departure-calibrated phrase set for the trip that is already booked.