Advertising measurement
Advertising measurement is optional, and you can use casewise either way. If you are in Australia or New Zealand, it is on unless you turn it off. Anywhere else, or if we can’t tell where you are, it is off unless you allow it. We work out the country from your connection when a page loads, and keep only which of these two rules applies, not the country. The public website is different: wherever you are, we run PostHog there and keep the visit’s campaign and source labels in your browser unless you turn measurement off, and we don’t ask first. If your browser sends a Global Privacy Control signal, we treat it as off. One choice covers advertising measurement and product analytics.
While it is on, we remember the public campaign, ad and creative labels and recognised referring-source categories (such as search, social or email) where available from the visit that brought you here. We do not retain full referring URLs or their paths. We use first-party browser storage shared between gocasewise.com and app.gocasewise.com to keep those identifiers through sign-in, including Google and email codes. We save this source evidence when your account is created, including before email verification, so we can distinguish unverified signups, verified signups and subscription outcomes.
With advertising measurement enabled, a random browser-session identifier connects landing visits and email-code request, failure and completion steps. These step records contain no email address, code or legal content, are not sent to Meta as verified conversions, and are removed after 90 days. We also retain account checkout history, including selected plan, billing interval and checkout handoff, separately from actual payment records. A requested code does not prove email delivery, and an opened checkout does not mean a payment succeeded.
While measurement is on, Meta Pixel on the public website sends public page visits and browser/click identifiers to Meta to measure advertising. On the public website, Meta Pixel runs in Australia and New Zealand unless you have turned measurement off; anywhere else it runs only if you have chosen to allow it, here or in the app. Meta also receives connection information such as your IP address and browser type, and may match visits to its users. Meta may process this information outside Australia. In the app we send manually defined Pixel image events for login/signup and the chat entry, plus verified signup, trial and payment outcomes. These image requests use fixed page addresses without query strings or referrers. We do not load Meta’s JavaScript into the app, use automatic advanced matching, or send form contents. Meta’s own privacy policy governs its processing.
While measurement is on, we also use PostHog to understand how people move from one of our ads to an account and how they use casewise. PostHog runs on the pages of gocasewise.com and on every page of the casewise app, chat included. It records page views, clicks (including clicks that do nothing), form submissions, how quickly the page loads and responds, errors on the page, and a session recording of how the page is used, such as scrolling, clicks and the page layout. Everything you type and all text on the page are hidden in recordings, and boxes where you could write a question are blanked out. Page titles, file names and other details on the page are removed, so PostHog never receives your legal questions, answers, documents or matter details. Links and page addresses keep only the kind of page, ad campaign labels and the plan you chose; ad click identifiers are removed. Each event also carries the public campaign, ad and creative labels and the kind of source (such as search, social or email) of the visit that brought you, and your account keeps those of your first visit. Error messages have email addresses and sign-in codes removed; outside the sign-up, sign-in and plans pages only the kind of error and where it happened are kept. PostHog also receives the steps you take in casewise, such as viewing plans, starting checkout, asking a question (and how many you have asked in that chat), whether an answer arrived, and copying, rating, continuing or downloading an answer or opening one of its sources, with the kind of source (never the words of a question, answer or source). When a trial starts, a payment goes through or fails, or a plan is cancelled, resumed or ends, our server tells PostHog that step with the plan and, when you cancel, the reason you chose from our list (never your note, an amount or card details), only while your choice allows it; we keep a copy of that choice with your account for this. If you answer one of our short questions, such as why you closed the plans without choosing one, PostHog receives your answer and any words you add, including any name or case detail you write there (we take out email addresses and long numbers). PostHog also receives connection information such as your IP address and browser type, and does not store your IP address. A random PostHog identifier is kept in a cookie shared between gocasewise.com and app.gocasewise.com, and expires 90 days after your last visit. While you are signed in, we link this activity to a coded identifier made from your account (never your email address or name) so we can compare visits with signup, trial and payment outcomes; coding it this way does not make it anonymous. PostHog stores this data in the European Union (Frankfurt) and may access it from other countries, including the United States. Recordings are kept for 30 days and other PostHog events for 12 months. Turning measurement off stops recording and removes PostHog’s identifier from this browser. It does not recall data already sent; contact info@gocasewise.com to have it deleted.
Whatever you choose, we count visits that open the app from one of our ads. Vercel receives an ad-arrival event containing the ad’s public identifier and placement. Once the page has loaded, we add one to a daily total for that ad, placement and page, stored by Supabase. These counts contain no cookie, browser, account or click identifier and no IP address, and our reports do not show totals under 5. While measurement is on, Supabase also stores the account attribution record. These providers may process data outside Australia, including in the United States. We do not include your legal questions, documents, answers, client or matter details, unhashed email address or authentication codes in advertising analytics events.
The browser attribution expires after 90 days from the first retained visit. If you turn measurement off, this browser remembers that for about 13 months and your account remembers it until you change it. Outside Australia and New Zealand, an Allow lasts 90 days, and the app then asks again. Account attribution is removed when your account is deleted. While measurement is on, we also send Meta verified signup, trial and payment events (including payment amount/currency), a SHA-256 hash of your email address, Meta browser/click identifiers (_fbp and _fbc) where available, and connection information (IP address and browser type), for matching. These identifiers are linked to your verified account only while measurement is on, expire after at most 90 days from their creation, and are removed from our matching records when you turn measurement off while signed in. A later click is not attached to an earlier conversion. Hashing does not make this information anonymous. Existing earlier consent does not enable these server events. If you created your account before 25 October 2026 and never made a choice, measurement stays off for that account unless you turn it on. Turning measurement off clears this browser’s attribution and stops new collection; when signed in, it also cancels pending server events. Already delivered events cannot be recalled by clearing cookies. It does not change your subscription. Contact info@gocasewise.com to request deletion of previously collected attribution.
For ads that open our login, signup or tagged chat entry directly, we also retain Meta’s ad-click ID (fbclid) while advertising measurement is on. We store it as a first-party _fbc cookie and send it with eligible server events. We read only that click parameter, not other query values, private app pages or your legal content. A first-party _fbp browser identifier is created only while measurement is on. Browser and server conversion events share the same event ID to avoid duplicate counting.
To compare our ad pages, some people who arrive from a Meta ad go straight to our sign-up page instead. A first-party cookie (cw_arm, 30 days) remembers which you got, so you see the same each time; it holds only that, and we count how many people get each version without any identifier. We also try two versions of the top of each ad page. A second cookie (cw_fs, 30 days) remembers which you got, so you see the same each time; it holds only that. If product analytics is on, the version is noted with that page’s events.
You can turn it off or on under Cookie preferences on the public website, Cookies at the foot of the sign-in window, or Cookies and privacy in app Settings. People who turn it off, block storage, change devices or remove tracking parameters may remain unattributed. A tagged visit is not proof that Meta charged for an ad click.