Nihongo Tree

Privacy

This page covers the Nihongo Tree website (nihongotree.com) and the Nihongo Tree app for iPhone and Android. Short version: no accounts, no ads on the site. Your study progress stays on your own device. The website counts visits with Google Analytics, which uses cookies, and with Vercel Web Analytics, which doesn't, and measures our ads on Reddit with Reddit's pixel, which uses a cookie. The app counts how it is used with Google Analytics, without cookies or an advertising ID. Effective 2026-11-19.

What is stored on your device

Nihongo Tree saves your progress and settings on your device only: which words you marked known or turned off,flash-card options, and display choices such as furigana, JLPT filters and playback speed. None of it is sent to us. On the website it also remembers your answer to the analytics question below.

On the website this lives in your browser's local storage. Clearing your browser's site data for nihongotree.com removes it, and another browser or device starts fresh.

In the app it lives in the app's own storage, with a second copy in the app's preferences so that your phone clearing the first copy to free space doesn't lose your progress. Deleting the app removes both. If your phone backs up its apps (iCloud on iPhone, Google backup on Android), that backup can include this data, as it would for any app; that is set in your phone's settings, and we never see it.

What we measure

We use Google Analytics, loaded through Google Tag Manager, to see how people use the site: which pages they read, how they found it, roughly where they are (country and city, worked out from the IP address), their device and browser, and how long they stay. Google Analytics sets cookies (named _ga and_ga_…) to tell a returning visitor from a new one; by default they last up to two years. Google processes this data under its own privacy policy.

We advertise the podcast on Reddit. To see whether those ads lead to visits and new listeners, Tag Manager also loads Reddit's advertising pixel. It tells Reddit which pages you open here and when you follow a link to Apple Podcasts, Spotify or the App Store, and sets a cookie (named _rdt_uuid) so Reddit can match a visit to an ad you saw there. Reddit processes this data under its own privacy policy. Tag Manager loads nothing else, and Google's advertising features are switched off.

In the European Economic Area, the UK and Switzerland, neither Google Analytics nor the Reddit pixel sets cookies unless you say yes. If your browser is set to a European time zone, the site asks once; the answer is kept in your browser. Elsewhere the cookies are on by default, and you can turn them off below.

We also count page views with Vercel Web Analytics, which sets no cookies and shows us, in aggregate, page paths, referring sites, country and device type. Neither analytics tool is used to show you ads.

The app uses Google Analytics for Firebase, the same Google Analytics account as the website, to count how it is used: which pages you open, and when you play a word or a Daily conversation, open flash cards, mark a word known, turn the reminder on or off, or pick a search result (not what you searched for). With each event it sends your device type, app version, rough location (worked out from the IP address) and a random ID that Firebase creates for this installation, so a returning user can be told from a new one. It doesn't use your phone's advertising ID, doesn't ask to track you across other apps, and sends no name, account or study progress. The same rule as the website applies: in a European time zone the app asks once and sends nothing until you say yes; elsewhere it is on, and you can turn it off below. Deleting the app removes its ID.

The app has no Vercel Web Analytics, no crash reporting and no advertising tools.

Your analytics choice

Google Analytics and Reddit pixel cookies on this browser: …

Turning them off also deletes the Google Analytics and Reddit pixel cookies already on this browser.

Servers that see your requests

The website, the word audio and the Daily conversations' audio are served from nihongotree.com, hosted on Vercel. The Daily conversations, their transcripts and the podcast feed come from our own server at api.nihongotree.com, partly by way of nihongotree.com, which passes those requests on to it. The app downloads word audio and the Daily conversations from these same two addresses; apart from Google Analytics while it is on, it talks to nothing else.

Like any web server, these receive your IP address and browser or device details with each request in order to send the page or file back. Our server at api.nihongotree.com keeps a standard access log of those requests (the IP address, what was asked for, whether it worked and the browser or app name). The log is overwritten as it fills up, which at today's traffic means each entry lasts about ten days. Vercel keeps its own short-lived request logs for nihongotree.com. With no accounts, nothing in these logs is linked to a person.

The podcast

The Daily conversations are also a podcast. Podcast apps such as Apple Podcasts and Spotify fetch the feed from nihongotree.com and the episodes through OP3 (op3.dev), an open podcast statistics service that counts downloads for us and then sends the app on to our server. OP3 sees each download (such as the IP address and the podcast app's name) under its own privacy policy. Listening on the website or in the app doesn't go through OP3. What a podcast app records about your listening is up to that app and its own privacy policy.

Notifications in the app

The app can remind you once a day when the new Daily conversation is out: around the time you usually open the app, and around 6:30 am Mountain Time until it knows that. To time it, the app remembers on your device only, for the last two weeks, the time of day you first open it each day. It is off until you turn it on from the Daily page. The reminders are scheduled on your phone by its own notification system: there is no push service, and no device token or other identifier is sent to us. Turn them off on the Daily page or in your phone's settings.

Speech

If a recording can't play, for example when you're offline, Nihongo Tree can read the word aloud with a Japanese voice on your device instead. In the app on Android that is the phone's own text-to-speech engine; on iPhone and in a browser it is the built-in speech voice. The words are never sent to us. Some browsers offer voices that the browser maker provides online, under its own policy.

What we don't do

  • No accounts or sign-in, so we hold no names, emails or passwords.
  • No cookies in the app, and none on the website apart from Google Analytics and the Reddit pixel.
  • No ads on the site or in the app. The only advertising tool is the Reddit pixel above, which measures our own ads on Reddit. If that changes, this page will say so first.
  • No selling of data, and no sharing beyond the services named on this page.

Questions and changes

Questions about privacy: write to support@nihongotree.com. If this page changes, the date at the top changes with it.

Nihongo Tree is made by Anthem Technology.