Privacy


Grammarism Logo

This page describes exactly what Grammarism records, why, and what it does not do. It is written to be accurate rather than comforting, so that you can decide for yourself.

Two third parties. All code and files are self-hosted, and there are no fonts, images or social buttons loaded from anyone else. The two exceptions are Google Analytics and Google AdSense. Google Analytics loads on every page and records that the page was opened; Google is sent the address of the page, the page that referred you and your IP address, and handles them on its own terms. It is not given the answers you type, your scores or which worksheets you download.

Ads, and the pages they run on. Google AdSense places the advertising, and its script is loaded from googlesyndication.com. It is loaded on the exercise pages and on no others: this page, the home page, the worksheet and index pages and the exercises embedded in partner sites carry no advertising script at all, so nothing of Google's advertising is contacted from them.

Where they are on an exercise page. Eight places, of which a wide screen uses seven. Three sit in the sidebar, one above the “Also Completed After” card and two under “Related Pages”; one is a grid of ad links under the closing row of buttons, below the last question; and one is a bar fixed to the bottom of the screen, which stays there for the whole visit and has no close button. The rest stand between the questions: on a phone, after the sixth, the eleventh and the sixteenth; on a wider screen, after the eighth and the fifteenth, because the sidebar's two have moved up beside the questions there instead of falling below them.

Which of those you get is decided in your own browser, from the width of the window, before any ad is asked for — and the ones your screen does not use are removed from the page rather than hidden, so no ad is requested, loaded or counted for a slot you were never going to see.

What the ads involve. On those pages Google and the advertising companies it sells through set cookies and similar identifiers of their own in your browser, and can read what any script on a page can read: your IP address, your browser and device, the address of the page and the page that referred you. They use it to decide which ad to show, to avoid showing you the same one over and over, and to count views and clicks; some of them also use it for interest-based advertising built from your browsing on other sites. That is held by them, under their terms rather than ours — Google's notice is at policies.google.com/technologies/ads. We hand them nothing of ours: not your identifier, not your answers, not your scores, and not which worksheets you download.

Your choices about ads. Our own notice at the bottom of the page is an acknowledgement, not a consent choice — it is about the cookie described below and asks nothing about advertising. Where European or UK law requires a consent choice for personalised advertising, it is Google's to ask, through the consent message that loads with its ad script; where no such law applies, no dialog appears. Either way you can refuse the cookies yourself: block third-party cookies in your browser settings, clear your browsing data for this site, or use an ad blocker — none of that stops the exercises, the marking or the downloads working. Interest-based advertising from Google can be switched off at myadcenter.google.com, and the industry opt-out pages are youronlinechoices.eu in Europe and optout.aboutads.info in the United States.

Our cookie. Grammarism itself stores a single cookie, named user_id. It holds a random-looking identifier and nothing else, it is never shared, and it expires after one year. Its purpose is to recognise a returning visit. Google Analytics sets two more of its own, both named beginning _ga, which it uses to tell repeat visits apart. On the exercise pages, which are the only ones that load the advertising script, Google and its advertising partners set further cookies of their own; those are not ours, and we neither choose nor control them.

Stored in your browser, not sent to us. A few things are kept in your browser's own storage and never leave your device: whether you chose the light or the dark theme, your progress through an exercise set, and the fact that you dismissed the cookie notice at the bottom of the page. Clearing your browsing data removes all of them, along with the cookie above.

Grammarism exercises on someone else's site. Our exercises can be embedded on partner websites. When one is, the answers you type into it and how far through you got are kept in your browser under that site's own storage rather than ours — there is no other way to stop your work being thrown away the moment you close the page, because a browser gives an embedded box no lasting storage of its own. It is still only your answers and your progress: no identifier, no cookie, and nothing that leaves your browser. Other scripts running on that site can read it, in the same way they can read anything else that site stores. It is removed by the “Clear Previous Answers” button under the exercise, by clearing your browsing data for that site, and automatically ninety days after you last answered.

How that identifier is produced, stated plainly. If you arrive without the cookie, the identifier is computed by hashing a combination of: the network block your IP address belongs to (not the full address — the last part is discarded), your browser's name, the language list your browser sends, and whether you are on a mobile device and which operating system family it reports. The hash cannot be reversed back into those values, and your full IP address is never stored. But we want to be straightforward with you: this is derived from your network and your device, so under data-protection law it counts as personal data, and we treat it that way rather than claiming otherwise.

What is recorded against that identifier. Each page you open (the address, and the page that referred you); the answers you type into exercises and whether each was correct; when you ask to be shown an answer; your score when you finish a quiz; which worksheets you download, so the daily limit can be enforced; and when a browsing session starts and ends. This is used to find questions that are unclear or wrongly marked, to see which material is useful, and to enforce the download allowance.

What is never collected. Your name, email address, date of birth, gender, precise location, payment details, or any account credentials — there are no user accounts on this site at all. Nothing you record here is sold, and nothing beyond what is described above is shared with anyone. None of what we record is given to the advertising companies, and none of it is used to choose the ads you see.

If you have a question about any of this, or want records associated with your identifier removed, please get in touch.

Be safe!

And do not open all the boxes...