How to choose a language app when you have a flight booked

Start with the two facts you already have: where you are going, and when you fly. The best language app for travel is the one that works backwards from both, and that test cuts a shortlist quickly.

So the useful answer to this query is not a single app name. It is a shape, and you can hold any candidate up against these five checks:

• Destination and date at the start, not a language in the abstract and an open-ended course that begins with animals and colours.
• Scenarios you pick yourself, so the restaurant, the ticket window and the pharmacy arrive first rather than in unit forty.
• A scheduler that knows your departure date, so nothing important is queued for review after you have already flown.
• Spoken practice from the first session, because the test you are preparing for happens out loud, at a counter, with someone waiting behind you.
• A price shaped like a trip, a one-off payment or a free tier whose limits you understand, rather than a subscription you will have to remember to cancel from an airport lounge.

Those five are the whole of this page. Everything below is how to run each check, in what order, and what the honest trade-offs are when an app passes three of them and fails two.

If there is no date in your diary, a broad course app is a reasonable choice. This guide is for the other situation, where the booking confirmation is already in your inbox and the runway is measured in weeks.

Before the booking, learning a language is open-ended. You can take a year, or three. After the booking you have a fixed number of evenings, most of them already spoken for, and a very specific set of moments you would rather not fumble.

That is a different problem with a different measure of success. You are not preparing to hold an opinion about local politics in a second language. You are preparing to order two of something, ask which platform, apologise properly, say where it hurts, and understand the reply well enough to nod in the right place. Travellers in this position tend to describe the goal the same way: polite, not fluent, and please not that tourist.

The three ways this usually goes

Keep grinding the general course app you already have, and hope the useful material surfaces before you fly. Or scope your study to the trip, accept that it will not make you conversational, and get properly good at a short list of lines. Or wing it with a translation app and a friendly expression.

The second option is the one that matches a date in the diary. The first is where most people end up by accident, three weeks out, because they never checked what the app was planning to teach them next, and there is a page in this section on whether Duolingo is enough to get by on a two-week trip if that is the app sitting on your phone. The third works better than people admit, right up until your phone has no signal in the back of a taxi and the driver is asking you something, which is the subject of the page on what a translator app covers and what you still have to say yourself.

What to stop optimising for

Grammar coverage, vocabulary breadth and daily attendance are all reasonable goals with a year to spend. On a six-week runway they are distractions. Breadth is the enemy of recall under pressure, and attendance is not the same as being able to say the thing. What matters now is a small set of lines you can produce without rehearsing them in your head first, and a plan that puts the most useful ones in your mouth earliest.

The rest of this guide is how to check whether an app does that. If you want the argument against the gamified course apps specifically, the alternatives page in this section makes it properly.

There is no honest universal number, and anyone quoting you one to two decimal places is guessing. What you can do is count moments, which is a much more useful exercise than it sounds.

Take the restaurant. Greeting. A table for two. Do you have a menu in English, and the grace to try without one. What do you recommend. I would like this one. I am allergic to something. Water, still or sparkling. The bill, please. Card or cash. That is nine or ten lines and you have covered most evenings out. Do the same for transport and you get a ticket to somewhere, which platform, does this stop at, what time is the last one, and a way of asking someone to repeat it more slowly. Ten to fifteen lines per situation, each one said out loud until it is automatic, is a realistic evening's worth of work repeated over a few weeks.

On that arithmetic a scenario is a short list first and a longer tail afterwards. A few dozen lines you can produce is a different asset from a few hundred you can recognise, and the second is what long courses quietly give you. So when you are weighing apps, the question is not how large the library is. It is whether the first twenty things the app hands you are the twenty you will personally have to say.

Phrazl is built to that arithmetic rather than to a syllabus. You pick a destination, a departure date and up to three scenarios from restaurant, transport, hotel, emergency and shopping, and each scenario's phrase set is pre-curated and ordered by trip-relevance rank rather than by grammatical difficulty. A paid trip pack holds 50 phrases per scenario, which is 150 for three, and the top of that ranking is where your first week goes. The free preview gives you the first 20 phrases of a trip, which is enough to judge the ordering before paying for anything, and the free preview page walks through what those twenty look like. If they do not look like your holiday, that is useful information and it cost you nothing. The phrasebook page in this section explains how the ranking is built.

Spaced repetition is the right engine for a fixed list of phrases. It shows you a line just as you are about to forget it, then stretches the gap each time you get it right, so your minutes go to the phrases that are slipping rather than the ones already solid.

It has one blind spot, and on a booked trip it is the whole problem. Classic spaced repetition assumes an open horizon. Get a card right a few times and the algorithm happily schedules it for three weeks out, or six, because it is optimising for remembering the phrase in a year. If your flight is in eighteen days, a phrase scheduled for day thirty-one has effectively been dropped from your trip, and nothing in the interface will tell you that.

What to look for instead

Ask whether the app takes a departure date at all, and if it does, whether it uses that date as a deadline or as decoration. A deadline-aware scheduler pulls intervals inwards as the date approaches, so the last review of every phrase lands before you fly rather than after.

Phrazl uses an SM-2 spaced repetition scheduler with interval compression calibrated to your trip's departure date as a hard deadline, which is why the date is the second thing it asks for after the destination. The free preview includes the departure countdown and the review schedule, so you can see what the plan looks like before you buy anything.

And what progress should mean

One more thing to check while you are in there: what the app counts as progress. Phrazl shows a single trip readiness bar labelled as curriculum coverage completed, with no streaks and no XP. That is a measure of how much of your trip's phrase set you have worked through, not a prediction of how any conversation will go. Nobody can promise you the second thing. The argument for one bar instead of a streak is laid out in full on the trip readiness page.

Reading a phrase and saying it are different skills, and only one of them is being tested when you reach the front of the queue. Any app that lets you feel finished without ever opening your mouth is training you for the wrong exam.

So the check is simple. Is there audio on every phrase, not just on a sample. Can you replay it as many times as you need without leaving the card. Can you slow it down when the run of consonants defeats you. And is there at least one mode where you have to produce the line rather than pick it from a list.

Phrazl has three learning modes: flashcard, listen and repeat with audio playback, and fill in the blank. Audio sits on every phrase in the free preview, so the out-loud part starts in the first session rather than after a paywall, and in a trip pack every phrase plays at speed or slowed down to copy, with a written pronunciation guide alongside it. Those screens now live in a phone app shell, a bottom tab bar with a More sheet over a single column, so the thing reads and installs like a phone app rather than a website squeezed onto a handset, which matters when you are practising standing up on a train.

One honest note, because travellers ask about this before they ask about anything else: the audio is a synthesised voice, not a native speaker, and the preview and plan copy say so before you pay rather than after. It is useful for rhythm, stress and where the emphasis falls. It is not a person from the place you are going. Judge it in the free preview with your own ears rather than taking a marketing sentence for it, which is good advice for every app in this category.

A second honest note, by destination. Fill in the blank hides a word inside the phrase, which works where the writing separates words. For Japanese, Chinese and Thai it falls back to hiding the whole phrase instead, so on those trips the out-loud work really rests on flashcards and listen and repeat. Worth knowing before you choose a mode to live in.

Whichever app you choose, the habit matters more than the voice. Say the line out loud, in the room, at the volume you would use in a shop. The first time you mispronounce something in public, you want it to be the second or third attempt, not the first.

Most language apps are priced as habits, because a monthly fee suits a product that wants you back every day for two years. A trip is not a habit. It is an event with a start date, an end date and a moment shortly afterwards when you no longer need the app at all.

That mismatch is why price shape deserves a decision of its own, taken before the feature list gets a vote. Three shapes are on the table, and each has a real trade-off.

• A free tier. Costs nothing, and the only question is where the wall sits. Check whether the free content reaches your situations or stops just before the useful part.
• A subscription. Sensible if you genuinely intend to keep learning after the trip. If you take one, set the cancellation reminder on the day you subscribe, not on the day you land.
• A one-time purchase. Matches the event, and the trade-off is that you pay up front on less evidence, which is why a free preview before it matters so much.

Phrazl is priced as the third shape: a free 20-phrase preview, then a one-time $9.99 trip pack, or $19.99 for the full bundle if you already have more than one flight booked, with a returning traveller discount on a second destination. The pricing page carries each plan's limits in full, there is a separate page on how the one-time checkout works, and another on what a second destination costs, so this page will not restate them. One limit worth knowing while you are deciding: any single trip takes up to three of the five scenarios, so covering all five at one destination means setting up a second trip for it, bundle or not.

Payments run through Stripe. Worth being straight about the stage: Phrazl is not open for purchase yet, so today there is a waitlist rather than a checkout, and no live payment has been taken.

Once you have chosen, the plan is the easy part, and it is the same plan whatever app you picked. Count backwards from the date instead of forwards from tonight.

Weeks one and two: one situation only

Start with the situation you will meet first, which for most trips is either the taxi rank or the restaurant. Take ten lines, not fifty. Say each one out loud, standing up, at the volume you would use in a room with cutlery noise. You are not trying to remember it, you are trying to make your mouth stop hesitating in the middle. Ten lines, fifteen minutes an evening, is enough to make the first evening abroad feel rehearsed.

Weeks three and four: add the second and third

Transport next, because the cost of getting it wrong is a missed train rather than a slightly odd dinner. Then whichever of the hotel, the pharmacy or the market applies to you. By now the first set should be coming back to you unprompted, which is what the review queue is for. If you are still having to look at the restaurant lines, do not add the third situation yet. Three half-learned sets are worse than two you can produce.

The last week: no new material

The week before you fly is for the lines you half know, the numbers, and the two sentences nobody practises and everybody needs: could you say that again more slowly, and sorry, my pronunciation is terrible. Resist the urge to add vocabulary at the airport. It will not stick and it will crowd out what has.

The only part of this worth handing to software is the timing, because deciding what to revise tonight is both tedious and easy to get wrong. That is what Phrazl's scheduler is for: SM-2 intervals compressed against your departure date as a hard deadline, so the last sight of each phrase lands before you fly rather than during someone else's holiday. The single readiness bar tells you how much of the set you have worked through. What it cannot tell you is how the conversation will go, and no app can.

Emergency is one of the five scenarios you can select, because pharmacy, police station and the sentence about an allergy are exactly the moments travellers are most afraid of, and pretending otherwise would be silly.

It also needs a plain limit, stated on the scenario selection screen and mirrored in the terms of service: these phrases are for general communication practice only, and they are not a substitute for contacting local emergency services or a qualified interpreter. Practise the line for it hurts here, and still call the number. Save the local emergency number in your phone before you fly, note whether your travel insurance has an assistance line, and treat the phrases as something that buys you thirty seconds of understanding, not as medical or legal help.

That is a fair test for any app in this category. If a product invites you into the emergency use case without saying where its usefulness ends, it is being careless with you.

This is the last check and the quickest one. You are handing over an email address, a travel date and, if you buy, a payment record, so it is reasonable to ask where those go before you sign up.

Phrazl signs you in with email and password, and stores a short list: your email, name and hashed password, your trips, your review history per phrase, purchase records and any support request you send from the account screen. Hosting is Amazon Web Services. Deletion is self-serve, and the delete account button clears the account, trips, review history, purchase records and support requests at once. Two things that are not in place yet, since you would rather hear them here than find out later: there is no export of your review history, and no two-factor authentication. The data page in this section sets out the full list, the retention, the three companies involved and the one caveat on deleting a single trip you have bought a pack for.

Whatever you choose, run this check on it. An app that cannot tell you what it keeps is telling you something anyway.