What we know about you, and what we don’t.

Part one — this website

This website serves public pages, a beta-list form, and a private Twin Training page for internal testers. The interactive demo uses temporary state in the current browser tab. Reloading the page discards that demo state.

There are no cookies. There is no analytics package, no tag manager, no advertising pixel, no heat-mapping, no session recording, and no embedded social media widgets. Page resources, including images, styles and scripts, are served from this site. Its browser security policy restricts resource loading and script-initiated connections; ordinary page navigation and email links remain available. You can inspect the requests by opening your browser’s network inspector and reloading the page.

The interactive demo does not leave your browser. Ratings you give, dishes you save and plans you build in the demo are held in your tab’s memory. They are not written to storage, not sent anywhere, and not associated with you. Reloading the page discards all of it. The demo does not create an account, send a message or invitation, or reserve a table — the people and venues in it are fictional.

Hosting request information

The site is hosted by Vercel Inc., which keeps ordinary web server logs the way every host does: the requesting IP address, the time, the page requested, and the browser’s user-agent string. These exist so the site can be delivered and so abuse can be blocked. We do not use them to build a profile of you and do not combine them with anything else. Vercel’s handling of that data is governed by its own privacy policy.

If you join the beta list

If you join the beta list, Love Salone receives your email address, phone type (iPhone, Android or other), general region, how you found us, your selected interest, confirmation that you are 18 or older, and permission to email you about early access. Vercel processes the request for us and stores it in private storage. We use it only to invite people to the beta in waves; it does not create a marketing subscription or guarantee an invitation. Entries are kept for 12 months and then deleted by a daily retention run. You can ask us to delete your entry sooner at hello@tastetwins.app. The form does not ask for your name, street address, ZIP code, health information or free-text notes.

If you use Twin Training

Twin Training is limited to signed-in internal testers. Supabase stores your account ID, dish answers, selected reasons, and an optional short “Something else…” note you write about what you like or dislike in a dish. That text is private to your account and kept with your answer. We may use AI to read notes like this and improve the choices. Notes are not used directly in matching or shared with other testers. Authorized operators review any suggested new choices before they are added. Please do not enter medical or contact details. You can read your own training answers and undo the latest one through the page; authorized Love Salone operators can review them to improve the reason choices. Training answers remain until you undo them, delete your account, or ask us to remove them at hello@tastetwins.app. Signing in keeps a session in this browser tab.

If you confirm a food rule, we save the excluded food and any exceptions to your account so Twin Training can skip those dishes. You can pause, edit or delete these rules. Rules do not store a medical reason or verify allergens.

Part two — the TasteTwins mobile app

What it collects

• Your reactions to dishes. Ratings you give on the app’s seven-point scale, from strongly disliked to strongly liked. This is the core of the product; without it there is nothing to learn from.
• Sensory ratings. Numeric intensity values across the app’s taste dimensions — how much crunch, salt, heat, acidity and so on a dish had for you — and the categorical descriptors attached to them.
• Taste Drivers. Optional selections, made after a meal, from a fixed vocabulary describing what actually made the experience good or bad.
• Dietary exclusions you declare. Allergies and lifestyle restrictions you enter yourself, used to filter out dishes you should not be shown.
• Coarse location, only while you are searching. City or neighborhood level, processed at the time of a venue search so results are near you. There is no continuous or background location tracking.
• Account basics. The email address or sign-in identifier you register with, and the profile details you choose to enter.

What it does not collect

• Government identification numbers of any kind.
• Genetic, biological or biometric data. “Taste DNA” is a name for your learned food preferences. It is not a genetic test, it involves no saliva sample, no DNA sequencing and no genetic data of any kind. We chose the name because it is memorable, and we would rather say plainly what it is not.
• Protected health information or medical records. TasteTwins is not a medical device and does not diagnose, treat or give clinical advice.
• Background GPS logs or travel history.
• Card numbers or financial account details. If payments are ever introduced, they will run through a PCI-compliant processor and the card details will not reach us.
• Your contacts, photo library, microphone or calendar, except for a specific photo you deliberately choose to attach.

How it is protected

• Row-level security is enabled on every application table, so a query can only ever return rows the requesting account is entitled to.
• The access policies do not permit anonymous row operations. An unauthenticated request cannot read or write application data.
• Your taste vectors are stored separately from your profile content, and an authenticated account can only insert and manage its own.
• Data is encrypted at rest with AES-256 and in transit with TLS 1.3.
• Access follows role-based controls.

Your control

For help exporting or deleting account data, email hello@tastetwins.app. You can delete your account in the app from Profile → Settings → Delete account. We keep a minimal hashed deletion record so the same request is not processed twice.

Research, and why it is a separate decision

Aggregate patterns in food preference are genuinely useful to public-health and academic researchers, and part of why this company exists is to make that data available to them. That is a completely separate decision from using the app, and it works like this:

• It is opt-in and specific. Nothing about your data goes to a research partner unless you have separately and explicitly agreed to it. Consent to use the app is not consent to be in a dataset.
• Declining costs you nothing. Every feature works identically whether you opt in or not. You will not be nagged, degraded, or shown a worse product for saying no.
• You can withdraw. Turning it off stops your data being included in anything released afterwards.
• Safeguards are being built, and are not finished. Statistical disclosure controls — thresholds ensuring no released record can be traced to a small enough group to identify someone — are a design requirement that must be implemented, tested and documented before any dataset is shared with anyone. They are not in place today. We are saying so rather than describing a planned control as an existing guarantee.
• No brokers, ever. We do not sell, rent or trade individual profiles to data brokers, ad networks or marketers. This is not a business model we intend to leave room for.

Children

TasteTwins is not directed at children under 13 and we do not knowingly collect their information. If you believe a child has given us data, write to hello@tastetwins.app and we will delete it.

Your rights

Depending on where you live, you may have the right to see what we hold about you, correct it, delete it, obtain a portable copy, or object to particular processing. We extend all of these to everyone who asks, regardless of jurisdiction, because maintaining two standards is more trouble than simply having one good one. Email hello@tastetwins.app and we will respond within 30 days.

We do not sell personal information, and we do not share it for cross-context behavioral advertising.

Changes

If this policy changes materially, the date at the top changes and app testers are notified in the app. Superseded versions are kept, so a change can always be pointed to rather than quietly made.

---

Questions, or something here that does not match what you have observed? Tell us — we would rather fix a document than defend one.