Parmenion privacy policy

Last updated: October 21, 2026

The short version

  • Parmenion reads your brand's Meta ad account through Meta's own login, and uses the access you give only to read. It can't change, pause or create ads, and it never posts, replies to, hides or deletes anything as your brand. It asks Meta for one permission, ads_read, which only reads.
  • Your brand's data is used only to run Parmenion for your brand. We don't mix it with other brands' data, sell it, use it for advertising or use it to train AI models. We may fit your brand's own settings and suggestions to your brand's own history, for your brand only.
  • Our code works out the numbers. Where Parmenion uses AI, it labels ads, sorts comments and writes short summaries and answers around those numbers. When Parmenion offers AI on your board, it can also read the files there when you ask it to, and draft items for you to add. Our AI provider doesn't train on your data and, under the settings we use, doesn't keep it after it answers, except where the law requires it or to look into misuse.
  • When Parmenion offers comments, we read the comments on your ads from Meta to find the questions customers ask. We don't keep the names of the people who wrote them.
  • When you disconnect or your plan ends, we delete your ad data within 30 days. You can ask us to delete it sooner at any time.
  • Questions or requests: sam@chibbsmanagement.io.

Who we are

Parmenion is made by CHIBBS MANAGEMENT LLC, an Illinois limited liability company. In this policy, "we" and "us" mean CHIBBS MANAGEMENT LLC. You can reach us at sam@chibbsmanagement.io, or by post at 1640 Marcie Court, Deerfield, Illinois 60015, United States.

Parmenion isn't made, endorsed or sponsored by Meta. We connect to Meta through its official login and API, under Meta's Platform Terms and Developer Policies.

We never ask for, or see, your Facebook or Instagram password. You sign in to Meta in Meta's own window.

We have two roles, depending on the data.

  • Your account: who you are, how you sign in, your billing, the business details you give at sign-up (such as revenue range and new ads a week) and how you use the app. Here we decide how data is used. Privacy laws call this a "controller" or a "business".
  • Your brand's data: the ad data your brand connects from Meta, the comments on your ads, any reviews you upload, your board in the Sandbox and the files you upload to it (when Parmenion offers the board), your goal, winner rules and strategies, and your Parmenion AI history (what your team types to Parmenion AI and what it writes back). Here we act for your brand and on its instructions. Privacy laws call this a "processor" or a "service provider". Your brand is responsible for that data too, and its own privacy policy also applies. Our data processing addendum is part of our terms of service and sets out how we handle your brand's data.
  • Data from Meta: whatever our role under privacy law, CHIBBS MANAGEMENT LLC is the developer Meta holds responsible for the data we get through its platform, and we follow Meta's Platform Terms for it.

Who this policy covers

  • People who visit our website, getparmenion.com.
  • People who use Parmenion, at app.getparmenion.com: brand owners and their team members. This policy calls them "you".
  • People whose data a brand connects to Parmenion. That includes people who comment on a brand's ads, people whose reviews a brand uploads, people who appear in or are heard in a brand's ads, and, when Parmenion offers the board, people named or shown in files a brand uploads. See "People who comment on, review or appear in a brand's ads or files".

Parmenion isn't open yet, and some of the features described here are still being built, such as reading the comments on your ads and your Facebook Pages and Instagram accounts, keeping copies of your ad images and of still frames from your video ads, your brand's board in the Sandbox, uploading files to it, Parmenion AI on the board, the analyzer, the briefer, the ad library and the free trial. This policy says how we handle data in each feature once it's available to you.

This policy covers Parmenion and its website only. It doesn't cover CHIBBS MANAGEMENT LLC's other services, or how Meta handles data on its own platforms. Their own privacy policies apply.

What we collect

What you give us

  • Account details: your name, work email and team.
  • Team invites: the email address of anyone you invite to your team, and any message you add to the invite.
  • Beta invites: while Parmenion is in beta, if we invite you to test it, your email address, with who on our team added it and when. We keep it until you sign in, or until we remove the invite.
  • Brand details: your brand's name, its website (optional) and the country your business is based in. On plans that include it, your brand's look: a logo you upload (PNG or JPG, up to 10 MB) and its colours. We don't read your website for these.
  • Set-up answers, each optional: your monthly revenue range, how many new ads you launch a week, and how you heard about us.
  • What your team tells Parmenion about the brand, when Parmenion offers it: what you sell, the words your brand never uses, and, for a brand on the Parmenion Super Strategy, its onboarding answers (such as brand files, and your winner and goal in your own words). They're your brand's data, kept and deleted with your brand settings.
  • Brand settings: your growth goal, your winner rules, which campaigns test new ads, which sale counts, any labels your team adds to ads, and a profile of your customers if you fill one in. Apart from the labels, they're kept while your brand is on a plan, and deleted as "How long we keep data" says. The labels are kept and deleted with your brand's ad data.
  • Strategies: the strategies your team builds, imports or saves in Parmenion, with their versions. A strategy is a set of rules and settings, with any names and notes your team adds, and holds none of your ad data. Strategies are kept while your brand is on a plan, and deleted as "How long we keep data" says. Our team can see them, as "To run Parmenion for you" says.
  • Tags and your customer map. The tags you put on your ads (such as funnel stage, awareness stage, pain point and angle) and the customer map you fill in (who your customer is, their pains, desires and objections). They're your brand's data. When Parmenion offers the new customer map, it holds your brand's avatars (each a type of customer, written or confirmed by your team; Parmenion AI can draft one from comments on your ads, marked as its own), one shared list of pains, and which avatars each ad speaks to. How aware your customers are is worked out by our code from where your ads and winners sit on the map.
  • Tag rules, your metrics and your 90-day customer value. The rules your team saves to tag ads by their words, the custom metrics it builds (each one's name, what it is, its inputs, filters, formula and any numbers typed into it), the metric cards it picks, and the 90-day customer value it types. They're your brand's data.
  • Your board and the files you upload. When Parmenion offers the board, on plans that include it: the notes, text, docs and flowcharts your team puts on your brand's board in the Sandbox, and the files it uploads there. We accept CSV, PDF, JPG and PNG files, not video or audio. When you upload an image, we remove the location saved in it, such as where a photo was taken. Apart from that, files are kept as uploaded: unlike reviews you upload as reviews, we don't remove names or email addresses from them. Your board and its files are your brand's data, kept and deleted as "How long we keep data" says.
  • Your ad library. When Parmenion offers it: the ads your team uploads to your brand's library (JPG, PNG, WebP, GIF, MP4 and MOV files, or a ZIP of them), with the names and tags your team or the analyzer gives them. We remove the location saved in images and videos, and skip a file you've already uploaded. Your library is your brand's data, kept for your brand only and never shared with another brand.
  • Subscriber counts. When Parmenion offers the recurring-revenue calculator: a CSV of subscriber counts your team uploads. It must hold counts only, never names or email addresses, and we refuse a file that looks like it lists people. We read the counts, drop the file at once and keep the counts as your brand's data.
  • Other people's details in files. Files can hold other people's details, such as your customers' names or email addresses. Upload them only if your brand has the right to, and has told those people what the law requires. Don't upload sensitive details, such as health information, card or other financial account numbers or government ID numbers, or any child's details. Your brand is responsible for the people's details in its files, and we may remove a file that breaks our terms of service.
  • What you ask Parmenion AI: what you type into the Ask box and, when Parmenion offers AI on your board, what you ask it to do there.
  • Angles and concepts. When Parmenion offers the briefer: the angles your team keeps and the ad concepts it writes or asks Parmenion AI to write. They're your brand's data.
  • Messages and feedback: emails you send us, what you tell us on a call you book with us, feedback in the app (such as thumbs up or down) and why you cancel, if you tell us.
  • A conversation you share with us. A Parmenion AI conversation your team shares after a thumbs-down, when Parmenion offers it, with what the AI read for it. It's your brand's data, and we read it only to fix the problem you reported (see "To run Parmenion for you").

What we read from Meta when you connect

In Meta's own window, you choose which ad accounts to share. Meta then gives us access through one permission, ads_read, which only reads. Meta's login also adds two more on its own, email and public_profile. We don't use them, and we don't keep the name or email they return. When Parmenion offers comments, Facebook Pages or Instagram, Meta's window will also ask which Pages and Instagram accounts to share, and we'll ask Meta for the extra permissions those need. We use ads_read to read:

  • your ad account, campaigns, ad sets and ads, their budgets, and their results, such as spend, impressions, clicks, reach, sales and sales value;
  • your ad account's activity history: the budget and status changes made to its campaigns and ad sets, and when. We don't keep who made them;
  • your ads' text, which images and videos they use, and the IDs of the Facebook or Instagram posts they run as;
  • when Parmenion offers them, private copies of your ad images and of still frames from your video ads, which we keep because Meta's links to them expire;
  • when Parmenion offers the analyzer and your team asks it to read a video ad, the video file. We write down the words spoken in it on our own servers (a transcript), and measure how they sound (pace, pauses, loudness, how much of the ad is speech, music under the voice and how much the pitch varies) as numbers. We keep the transcript and the numbers, never the sound or the whole video, which we delete within 1 hour of fetching it;
  • when Parmenion offers them, the Facebook Pages and Instagram accounts you shared: their names and IDs;
  • when Parmenion offers them, comments on your ads on Facebook and Instagram: the text, when it was posted and which ad it's on. See "People who comment on, review or appear in a brand's ads or files" for how we handle commenters;
  • an access token that lets us keep reading, and the ID Meta gives your business.

Reading Instagram comments needs a permission Meta calls instagram_manage_comments, which would also let an app hide, reply to or delete them. When Parmenion offers Instagram comments, it will use it only to read them.

When you first connect, we read up to a year of your ad account's history. While the ad account stays connected on a plan, we read new data from Meta about once a day. If we offer you a free audit, we read once, then delete our access.

In this policy, "your ad data" means everything we read from Meta for your brand, plus everything we work out from it, such as winners, labels, summaries, scores, your plan, the analyzer's reads and transcripts, and the dated record we keep of each live suggestion and its inputs so we can check it later.

Reviews you upload

If your team uploads customer reviews, we keep each review's text, rating, date and product. We remove reviewers' names and email addresses when you upload them. A file of reviews your team puts on its board, when Parmenion offers the board, is kept as uploaded, names included (see "What you give us").

What we collect when you use Parmenion

  • Usage: the pages you open, what you click, how far you scroll, which features you use, and your answers to our in-app survey, with your device and browser type, and your country (worked out from your IP address, which isn't kept). See "Cookies and analytics".
  • Error reports and server logs: when something breaks, a report of the error with the page and your browser type, and our servers' logs of requests. We remove tokens, emails and your brand's data first, and keep the logs in our analytics provider for 14 days.
  • Security data: your IP address, browser details, sign-in times and a record of important actions, such as connecting, exporting and deleting.
  • Bot checks: when you sign up, we may run a check that reads signals from your browser to tell people from bots.

Billing

Payments go through Stripe. You pay on Stripe's own page, by card, Apple Pay or Google Pay, so your payment details go straight to Stripe. We never see or store your full card number. From Stripe we get your billing name, email and address, your tax ID if you give one, your plan, your invoices and whether payments went through. Stripe uses your address to work out any sales tax.

  • One free trial per business. To give each business one free trial, we keep your brand's website domain, the domain of the email address you sign up with, and a code Stripe gives us for your card (its fingerprint, never the card number), and match them against earlier trials. We keep them with your billing records, even for a trial that never took money.
  • Payment evidence. For each payment, we keep what the checkout showed you and what you agreed to there, the checkout's IP address and browser, the notices we sent and a summary of how your team used Parmenion in that period, so we can answer a payment dispute.
  • The Parmenion Super Strategy. For this service we invoice your business directly, and you may pay by bank transfer. We keep each invoice and payment's reference with your billing records.
  • Founding access. If you join founding access, Stripe's checkout also asks for your phone number, and may ask for your brand's name and website and your Discord name. Stripe passes them to us with your payment, so we can reach you to book your onboarding call and set up your account. We keep them with your billing records.

On our website

If our website offers free tools, such as calculators, they work out their results in your browser. What you type into them isn't sent to us, and they don't ask for your email. Today our website sets no cookies and runs no analytics or ad tracking (see "Cookies and analytics"). Like any website host, Vercel receives your IP address and browser details when you open a page, to deliver it and keep the site secure.

Our home page has one box where you can join our list. If you give your email address there, your browser sends it straight to Loops, the company that sends our emails, and we use it to email you about Parmenion: its free preview when it opens, launches and offers. After you join, you can also tell us whether you run a brand or work as a strategist, and roughly how much you spend on ads a month. Both answers are optional, and we use them only to choose which of our emails fit you. Every email has a one-click unsubscribe, and we never build these emails from ad data.

How we use it

To run Parmenion for you

  • To create and secure your account and team, and sign you in.
  • To connect to Meta and read your ad account.
  • To work out your numbers: winners, testing, scaling, angles, your plan and your suggestions.
  • To keep the files you upload, show them on your board and use them in your strategies and metrics, when Parmenion offers the board.
  • To answer questions in the Ask box and, when Parmenion offers AI on your board, to read your board and its files and draft items for it when you ask.
  • To bill you, and send the emails the service needs: sign-in links, team invites, beta invites, billing emails (including, during a free trial, Stripe's reminder when it starts and ours 24 hours before the first charge), and notices about security incidents, deletion requests and changes to our terms, this policy or our subprocessors, and one email before we close an inactive account.
  • While Parmenion is in beta, when your team invites someone to connect Meta, we're told, so we can give that person the tester access Meta requires until our app is approved.
  • To fix a problem your team reports with Parmenion AI. Anyone on your team can share their own conversation with us after a thumbs-down, when Parmenion offers it, by ticking "Share this conversation, and what Parmenion AI read for it, so we can fix it". Our team reads what was shared only to fix that problem. It can include parts of your board, your files, your ad data, your tags and your customer map. We don't keep it in our tests or use it to train any AI.
  • To help when you ask for support. Apart from strategies (next line), our team looks at your brand's data only: when your team's owner invites one of us to your team for support, as a member of your team, until the owner removes us; to investigate a security problem; to read a conversation your team shares with us after a thumbs-down, when Parmenion offers it (above); when Parmenion offers the board, to check a report that a file on it breaks our terms of service; or, if your brand is on the Parmenion Super Strategy, in a switch-in (below). The invite is the record of a support visit, and we keep a record of every other look.
  • Our team can see every strategy saved in Parmenion, and its versions, whenever needed, to run, fix and support Parmenion, except a Super Strategy's own strategies, which our team sees only in a switch-in. That includes the names, notes and settings your team types into it, but none of your ad data. What a strategy shows when it runs on your brand's own numbers is your brand's data, and our team sees that only as the line above says.
  • The Parmenion Super Strategy. If your brand is on our Super Strategy service, our staff can switch into your brand's account to build and tune your Super Strategy, as your contract says. They can do what a member of your team can, including exporting your brand's data and adding members to your team, and they can save your goal, winner rules and plan setup as an admin would. They can't see billing, connect or disconnect an ad account, delete your data or close your account. Your owner can turn this off in Settings, and sees each switch-in, change and export there. To approve your brand's theme, our staff also see the logo and colours your team uploads.
  • To keep Parmenion secure, stop abuse and meet our legal duties.

To run our business

  • Improving Parmenion. To fix problems and decide what to build, we look at how people use it (see "Cookies and analytics"), your set-up answers, the feedback you give in the app (apart from a conversation you share after a thumbs-down, which we read only to fix the problem you reported), and what you tell us in emails and on calls. We count how often the Ask box is used and whether an answer helped, but we don't read the questions or answers for this, and we don't use your ad data. In our analytics, strategies show only as counts, such as how many a brand has saved and which building blocks they use, and, when Parmenion offers the board, files on your board only as their type and size.
  • How we review Parmenion AI. Where Parmenion uses AI, we count how it's used, for every brand: which features, how often, when they fail, thumbs up or down, and what each use costs us, with the number of tokens. We never read what you ask it or what it writes for this, and we don't use your ad data or your files.
  • Preventing repeat free trials. We match a new trial's website domain, email domain and card code against earlier trials (see "Billing").
  • Where people hear about us. We count the answers to "How did you hear about us?" to see where sign-ups come from. Our team sees only the totals for each answer.
  • Offering larger plans. We use your set-up answers about your revenue range and new ads a week, and how you use Parmenion, to decide who might want to hear about our larger plans. We only tell you about them inside Parmenion. We don't use your ad data, or how you heard about us, for this, and you can ask us to stop at any time.
  • Marketing emails. If you sign up for Parmenion or join our list on our website, we may email you about launches, guides and offers. Every marketing email has a one-click unsubscribe, and unsubscribing doesn't stop the emails the service needs. We never build these emails from your ad data. Where the law requires your consent first, we ask for it.

Your brand's data is used only for your brand

  • We use your ad data, uploaded reviews and the files you upload only to give your brand Parmenion's answers, and only for as long as this policy says.
  • Files you export from Meta yourself, such as an Ads Manager report, and upload to your board when Parmenion offers it, are used only for your brand, kept and deleted with your board, and counted as data from Meta when we report a security incident to Meta.
  • We keep each brand's data separate from every other brand's.
  • We don't combine brands' data, build benchmarks across brands, or train AI models on it, whether ours or anyone else's.
  • Your P1 score, and your brand's own settings and suggestions (such as your win-rate target, or how often each kind of suggestion beat a plain guess on your account), are worked out from your brand's own history, for your brand only, by replaying your past. What we work out for one brand is never used for another, and it's deleted with your brand's data.
  • When your team shares a strategy as a file, it carries the definitions of the custom metrics it uses (each one's name, what it is, its inputs, filters and formula), never a number worked out from your data, and never a file's name, its column names or its rows: a column from one of your files goes as an empty slot that whoever imports it fills with a column of their own.
  • We don't sell it or rent it, and we don't use it for advertising or retargeting, or to build profiles of people.
  • We don't use it to make decisions about people, such as for jobs, credit or housing.
  • For data we get from Meta, we also follow Meta's Platform Terms and Developer Policies. Nothing in this policy overrides them.

How we use AI

  • Our code works out every number in Parmenion. Where Parmenion uses AI, it adds the words around those numbers.
  • What AI may do:
    • label each ad, for example whether it's a new idea or an edit of an old winner, and which customer question it answers. When it isn't sure, it says "Not sure" rather than guess, and Parmenion shows how often its labels agree with your team's;
    • when Parmenion offers the analyzer and someone on your team asks, read each ad's picture, on-screen text, words and, for a video, its transcript, to say what it's made of, such as its hook, angle, the pain it speaks to, awareness stage, format and style. It says "Not sure" where it's unsure, marks style labels as "inferred", and never labels anyone's emotions. The analyzer's diagnosis of an ad against your own recent numbers is plain counting in our code, with no AI;
    • when Parmenion offers the briefer and someone on your team asks, find angles and write ad concepts from your brand's own data and what your team types (such as what you sell);
    • when Parmenion offers it, draft avatars on your customer map from the comments on your ads, for your team to confirm;
    • sort comments and reviews by the questions and worries customers raise;
    • write the short summaries on your pages;
    • answer questions you ask in the Ask box;
    • turn a request you type into a draft strategy, and explain a strategy's results in plain words;
    • when Parmenion offers Analyze, explain a chart in plain words when you click it;
    • when Parmenion offers AI on your board, and you ask it to there or in a strategy, read what the task needs from your board and its files, such as a CSV's columns, a PDF's text or what an image shows, and draft items for you to add or discard, such as notes, docs, flowcharts and strategies.
  • What we send: only what the task needs. That can be the numbers our code worked out, your ads' text, images and still frames from video ads where a task needs them, and, when Parmenion offers the analyzer, their on-screen text and transcripts, comment and review text, your question, what your team types for the briefer, and, when Parmenion offers AI on your board, the parts of your board and its files that the task needs. Files go the same way as everything else, under the same agreement and settings, with any names in them. We never send an AI provider your access tokens, your billing data, or commenters' names or IDs from Meta.
  • What AI can't do: it has no direct access to our database. It can only call a few read-only functions our code gives it, and they return only your brand's data. It can't see other brands' data or change your ads (nothing in Parmenion can), and it can't send messages. When Parmenion offers AI on your board, it only drafts there: what it makes lands as a draft, marked as the AI's, and nothing is added until someone on your team adds it.
  • Our AI provider is Anthropic (Claude). We send requests to Anthropic directly, and only under a written agreement that bars it from training on your data. Under the settings we use, it doesn't keep your data after it answers, except where the law requires it or to look into misuse of its service.
  • We don't make images. We don't use your ads, images or videos, or the images and files you upload, to make new ones.
  • What Parmenion AI writes is marked. Everything Parmenion AI makes that your team sees carries a "Parmenion AI" mark, where Parmenion offers each feature: Ask answers, items and drafts on your board, Analyze's text, the labels and tags it sets ("Not sure" included), the names it gives groups of customer questions, avatars and lines on your customer map, transcripts, and the angles and concepts it writes. A line keeps its mark until someone on your team edits it (it then shows "Edited") or removes the mark, one at a time, a selection or all at once, and we record who removed which mark and when. If your brand is on the Parmenion Super Strategy, our staff can remove a mark in a switch-in too, and the record names them. AI can get things wrong, so check anything important before you act on it.
  • No automated decisions about people. Nothing Parmenion produces decides anything about a person with legal or similarly serious effects.

People who comment on, review or appear in a brand's ads or files

This part is for people who don't use Parmenion, but whose comments, reviews, images or other details a brand shares with us.

  • What we read. When Parmenion offers comments, we read the comments people post on a connected brand's ads on Facebook and Instagram, through the brand's own access. We do this to count the questions and worries customers raise, such as "Isn't it too heavy?", so the brand can make ads that answer them.
  • What we keep. The comment's text, when it was posted and which ad it's on.
    • We don't keep commenters' names, usernames or profile pictures.
    • We don't store commenters' Meta IDs. We store a code made from each ID with a secret key, a different key for each brand. Without our key, no one can link the code to a person, and codes can't be matched across brands. With it, we can find your comments when you ask us to delete them, and count how many different people asked something. We treat these codes as personal data.
    • We remove @-mentions of other people from the text.
  • What we never do. We never send commenters' names or IDs from Meta to AI, or show them in Parmenion. We don't contact commenters, reply to or hide comments, build profiles of people, or use a comment for any brand other than the one whose ad it's on.
  • Reviews a brand uploads as reviews are handled the same way. We remove reviewers' names and email addresses when they're uploaded. A file of reviews a brand puts on its board, when Parmenion offers the board, is kept with the names in it, like any file (below).
  • If you appear in a brand's ad, we read the ad for that brand only, and, when Parmenion offers them, keep copies of its images and of still frames from its video to analyse it. We don't try to identify anyone in it. When Parmenion offers the analyzer, it may write down what you say in a video ad and measure how it sounds (such as pace and loudness), as numbers, for that brand only. We don't match faces or voices, make face or voice templates, label who is speaking, or label anyone's emotions or traits, and we keep no audio.
  • If you're in a file a brand uploads, such as a spreadsheet, a PDF or a photo (brands can upload files when Parmenion offers the board), we keep it for that brand only, with any names in it, on its instructions, and delete it with the brand's data. The brand's team can see it on its board and, when Parmenion offers AI on the board, Parmenion AI may read it when someone on the team asks. The brand is responsible for having the right to share your details with us, and for telling you what the law requires. We don't try to identify anyone in it. Requests about it go to the brand: contact the brand, or email us with the brand's name, and we'll pass your request on. We don't delete a file, or a review in one, at your request ourselves, unless it breaks our terms, such as a file with a child's details (see "Children").
  • If you delete your comment on Facebook or Instagram, our copy is deleted with the rest of the brand's data. To have it deleted sooner, email us.
  • To ask us to delete your comment or review, email sam@chibbsmanagement.io with the brand's name and a link to your comment, or its text. We'll delete it from our systems within 30 days and tell the brand. You can also ask the brand directly. A review in a file on a brand's board is handled like any file (above).

Who we share data with

We share data only in the ways listed here.

  • Service providers (subprocessors). These are companies that help us run Parmenion, under written contracts that let them use the data only to provide their service to us:
    • Vercel, for hosting, background jobs, file storage and the bot check;
    • Neon, for our database;
    • Anthropic, for AI;
    • PostHog, for product analytics, error reports, server logs, feature rollouts, our in-app survey, heatmaps, and Parmenion AI's counts and cost (never its text);
    • Loops, for sign-in, account and marketing emails, for the list you can join on our website, and for our one billing email: the reminder 24 hours before a free trial's first charge, which carries your brand's name, its plan and price, and when the trial ends (no card digits, nothing about ads);
    • Stripe, for payments and billing emails, such as receipts, failed-payment notices, renewal reminders and its reminder when a free trial starts;
    • Google Workspace, for our email inbox and internal records.

    Our subprocessors page says what each one gets, where it works from, and whether it handles customer data yet. We add a new one to that page, and tell team owners by email, at least 30 days before it starts handling your data, or before an existing one gets a new kind of data. In an emergency, such as replacing one to keep Parmenion running or secure, we'll tell you as soon as we can, and you can still object.

  • Stripe's own uses. Stripe also uses some payment data for its own purposes, such as preventing fraud and meeting its legal duties, under its own privacy policy. It never gets your ad data.
  • Your team. People on your team can see your brand's data in Parmenion.
  • Our staff and strategies. Our staff can see the strategies saved in Parmenion as they're saved: their rules, settings, names, typed numbers and each strategy's own notes, the definitions of the custom metrics a saved version uses (each one's name, what it is, its inputs, filters, formula and typed numbers, as typed), and their versions. They never see what a strategy shows when it runs on your numbers, except as a member of your team your owner invited for support, or to investigate a security problem. They never see your tags, tag rules, customer map, custom metrics or 90-day customer value, except a custom metric's definition inside a saved strategy version. They see the notes and docs on your board, and the files on it, only as a member of your team your owner invited for support, to investigate a security problem, or to check a report that a file breaks our terms. The exceptions to these three limits are a Parmenion AI conversation your team shares with us after a thumbs-down, which can include any of them (see "To run Parmenion for you"), a member of our staff your owner invited to your team, and a Super Strategy switch-in, where staff see what a member of your team sees. A Super Strategy's own strategies are seen only in a switch-in.
  • What we count. We count how many strategies are saved, imported, shared, applied and deleted, which blocks strategies use (from the strategies staff can already see), and how drafts turn out. We also count when a strategy is removed from your plan or refused (and why), when ads are tagged or untagged (the kind of tag only), and when your customer map is saved. We also count when a tag rule is saved, switched off or on, or deleted, how many ads a rule tags as new ads arrive, when a custom metric is made, saved or deleted, or refused (and why), and when your 90-day customer value is saved. When Parmenion offers the board, we count when a file is added to or removed from it (its type and size only). Where Parmenion uses AI, we count how it's used: which features, how often, when they fail, thumbs up or down, and what each use costs us. Counts only: never a strategy's recipe, name, note, a tag's value, your customer map's words, a tag rule's words, a metric's name, formula, inputs or typed numbers, your 90-day customer value, a file's name or contents, or anything typed to or written by Parmenion AI.
  • Your own tools, when you tell us to. If you export your data, we send your brand's data where you point it. That tool's terms and privacy policy then apply. Our terms of service require you to keep using data from Meta within Meta's rules.
  • Meta. When we read your data, Meta sees our requests. Meta's terms also let it check how we handle data from Meta. If Meta asks, we give it our list of customers (each business's name and contact details), the companies that handle data from Meta for us, and answers about how we use that data. We report to Meta any security incident involving data from Meta.
  • Our professional advisers, such as lawyers and accountants, when they need it to advise us. They must keep it confidential, and they get ad data only if a legal claim or legal duty needs it.
  • When the law requires it, for example a valid court order. We check that each request is lawful and push back on any that isn't, or that asks for too much. We give only what's required, keep a record, and tell the customer affected unless the law forbids us.
  • To protect people. We may share account details when we believe in good faith it's needed to stop fraud, abuse or harm to someone. We don't share your ad data for this unless the law requires it.
  • If our company changes hands. If CHIBBS MANAGEMENT LLC merges or is sold, data may move to the new owner, who must keep this policy's promises. Data from Meta moves only if Meta's terms allow it. We'll tell you before your data moves to the new owner, so you can delete it first.

We don't sell personal information we collect through Parmenion or its website. We don't share it for cross-context behavioral advertising, which means ads based on what you do on other sites and apps. We don't share it with other companies for their own marketing.

Cookies and analytics

Our public website, getparmenion.com, sets no cookies and runs no analytics today. The app uses a few cookies and similar tools (such as local storage in your browser):

WhatWho sets itWhyCan you turn it off?
Sign-inParmenionKeeps you signed in and protects your sessionNo. Parmenion can't work without it
Bot checkVercelTells people from bots when you sign upNo. It's needed for security
Product analytics in the appPostHogShows us which pages and features get used, for up to 1 yearYes. See below
Your analytics choiceParmenionRemembers your yes or no on this browserYou can clear it in your browser
Trial lengthParmenionWhen you open one of our offer links, remembers only how long your free trial is (3 or 7 days), for 30 daysYou can clear it in your browser
PaymentStripe, on its own checkout pageTakes payment and checks for fraudNo, if you want to pay. See Stripe's privacy policy
PreferencesParmenionRemembers small choices, such as a tab you pickedYou can clear them in your browser

Analytics (PostHog). Where we run product analytics, we use PostHog, set up this way:

  • PostHog records the pages you open, what you click, how far you scroll, errors, survey answers and your device and browser type.
  • We use PostHog's US Cloud. PostHog uses your IP address to work out your country, then discards it.
  • Inside the app, analytics is tied to an account ID, not your name or email.
  • We don't record your screen or sessions.
  • PostHog never gets your ad data. We don't let it record the text on your screen in the app, and the page addresses and titles it records use our own IDs, never Meta IDs, ad names, numbers or comment text. For Parmenion AI, it gets the feature, model, token counts and cost, never what was asked or written. For our payments, it gets the plan, the amount and our ID for your brand, never names or card details.
  • We turn off PostHog's AI features, and the setting that lets PostHog train its own models on anonymised data.
  • Our public pages run no analytics today. If we add it there, it will run without cookies. Inside the app, we ask first if you're in the EU, the UK or Switzerland.
  • You can ask us to turn analytics off for you at any time. Where Settings offers an analytics switch, you can also use that.

No ad tracking. We don't use advertising pixels, cookies or server-side ad tracking, such as the Meta Pixel or Meta's Conversions API, on our site or in the app. We don't upload our users' emails or phone numbers to ad platforms to target or measure our own ads. If this ever changes for our public website, we'll update this policy first.

Browser signals.

  • Do Not Track: there's no common standard for these signals, so we don't respond to them. We don't track you across other companies' sites anyway.
  • Global Privacy Control: we treat it as a request to opt out of any sale or sharing of your data. We don't sell or share it.

How long we keep data

DataHow long we keep it
Your ad data, uploaded reviews and the tags on your ads. Also, when Parmenion offers them, what's worked out from an ad account's data: the analyzer's reads, transcripts and sound numbers, its checks and accuracy history, the avatars linked to an ad, concepts Parmenion AI writes from that account's data, and the dated record of each live suggestion and its inputsWhile the ad account is connected on a plan. Deleted within 30 days after that ad account is disconnected, our access to it stops working or your plan ends, or within 7 days after you ask us to delete it. If you connect that ad account again or restart your plan within the 30 days, nothing is deleted.
A video file the analyzer fetches, when Parmenion offers itDeleted within 1 hour of fetching it. We never keep the sound or the whole video.
Your brand settings (such as your goal and winner rules), strategies, tag rules, your customer map (its avatars and pain list), your custom metrics and metric cards, and your 90-day customer value. Also, when Parmenion offers them, your ad library, the angles your team keeps, concepts your team writes, your subscriber counts, your brand's logo and colours, what your team tells Parmenion about the brand, and a Super Strategy onboarding's answersWhile your brand is on a plan. Deleted within 30 days after your plan ends, or within 7 days after you ask us to delete your brand's data. If you restart your plan within the 30 days, nothing is deleted. They belong to your brand, not one ad account: disconnecting an ad account, or our access to one stopping, doesn't delete them, but facts in them drawn from that account's data go with it.
Your board in the Sandbox and the files you upload to it, when Parmenion offers the boardWhile your brand is on a plan. Deleted within 30 days after your plan ends, or within 7 days after you ask us to delete your brand's data. If you restart your plan within the 30 days, nothing is deleted. Your board belongs to your brand, not one ad account: disconnecting an ad account, or our access to one stopping, doesn't delete it, but what we worked out from that account's data leaves your board with that data, such as its numbers and anything Parmenion AI made from it.
A file you delete from your board (with a CSV's rows), when Parmenion offers the boardDeleted within 7 days. Until then, undo can bring it back. It leaves our backups 7 days after that.
A board you delete, when Parmenion offers the boardIts notes, docs, versions and files are deleted within 7 days, and undo can't bring them back. They leave our backups 7 days after that. Its strategies stay.
A strategy you delete, when Parmenion offers the boardKept for 24 hours, with its versions, so undo can bring it back. Then it's deleted, and it leaves our backups 7 days after that.
A file you delete from your ad library, when Parmenion offers itDeleted within 7 days, and leaves our backups 7 days after that. The names and tags the analyzer gave it go with it.
A subscriber-count file you upload, when Parmenion offers itThe file is dropped as soon as its counts are read. If you remove the counts, they're kept 24 hours so undo can bring them back, then deleted.
Meta access tokensDeleted within 1 hour after the connection ends, a free audit is built, or we act on your request to delete your data.
A free audit, if we offer you one, and the ad data behind it30 days after the audit is ready, unless you start a plan. Our access to your Meta account is deleted within 1 hour of the audit being built, so we don't keep reading your ads.
What you type to Parmenion AI and what it writes back, in the Ask box and, when Parmenion offers AI on your board, on the board, and a conversation your team shares after a thumbs-down, when Parmenion offers it, with what the AI read for it90 days, and always deleted with your brand's ad data. What you add to your board stays with your board (strategies keep their own clock), and what a shared conversation holds from a file goes when the file does. Our logs and usage records note that a question was asked, never what it said.
Account detailsWhile your account is open. Deleted within 30 days after you close it. If you have no plan and haven't signed in for 12 months, we may close your account, after emailing you once, 14 days before.
Brand details and set-up answersKept with your brand, and deleted with your brand's data.
Beta invitesUntil the person signs in, or we remove the invite.
Marketing email contacts (your email address, name, plan and sign-up date, and whether you unsubscribed)While your account is open, then deleted within 30 days after it closes. If you unsubscribed, we keep only your email address, on a do-not-email list, so we never email you again.
People who join our list on our website (your email address, the date you joined, and your two optional answers, if you gave them)Until you unsubscribe or ask us to delete it. If you haven't opened a Parmenion account within 24 months of joining, we delete it then. If you open one, the row above applies instead. If you unsubscribed, we keep only your email address, on a do-not-email list.
Records that you accepted our terms, that your team agreed to what we read before connecting Meta, and your analytics consent choice (EU, UK and Switzerland). For a brand on the Parmenion Super Strategy, also the list of switch-ins and each change to the owner's switchWhile your account is open, then 12 months.
Our staff's working notes on a Parmenion Super Strategy serviceOur own records, not your brand's data, so they aren't in your export. Deleted 12 months after the offer ends.
Usage analyticsIn PostHog, events for up to 7 years, which PostHog doesn't let us shorten; server logs, 14 days. No text from Parmenion AI. In our own usage records, 12 months. None of it holds your ad data. If you close your account, we delete your usage records, and delete you, with your events, in PostHog.
Security logs12 months. IP addresses go after 90 days.
Support messages24 months after the last message.
Billing records, including what we check to give each business one free trial (website and email domains, and Stripe's code for the card)As long as tax law requires, 7 years after your brand's newest billing record, even for a trial that never took money.
Payment evidence: the checkout's IP address and browserWhile your plan runs, then 12 months after your last payment. A free trial's checkout counts once its first charge goes through. A checkout that's never paid, such as a trial cancelled before its first charge, keeps its IP address 90 days.
Payment evidence: what you agreed to at checkout, the notices we sent, and a summary of your use for each paid period12 months after each charge. If a payment dispute opens, a copy of your brand's sign-in records is kept until it closes, plus 30 days.
Records of deletion requests3 years, so we can show the request was done. They hold the dates and what kind of data was deleted, never the deleted data itself.

Backups and copies.

  • Deleted data leaves our database backups within 7 days. Our hosting provider may also hold copies in its own platform backups, which we can't reach. They expire within 30 days.
  • Our background-job system keeps a record of each job, which can include your data, for 7 days.
  • Our email provider keeps sent emails and delivery logs for the periods its own terms set. Our emails don't contain your ad data.
  • Our servers keep logs for up to 30 days. We never write access tokens or your ad data to logs.
  • Stripe, which sends most of our billing emails, keeps its log of them for 60 days. Our free trial reminder goes through our email provider, like our other emails.

More detail is on our How to delete your data page.

How to delete your data

Anyone can ask us to delete their data, whether or not they use Parmenion.

By email. Email sam@chibbsmanagement.io from the address on your account, and tell us what to delete. We'll confirm the request is yours by replying to that address. You can also use Settings: Disconnect under Connections, Export everything and Delete my data under Your data, and Close my account under Your account. They do the same things as an email, on the same clocks.

  • Delete everything for a brand: only the team owner can ask for this. We delete our Meta access within 1 hour of acting on the request, disconnect Meta and end the brand's plan. The brand's ad data, uploaded reviews, goal, winner rules, strategies, board and uploaded files, and Parmenion AI history are deleted within 7 days of the request, and leave our backups 7 days after that. So are, when Parmenion offers them, its ad library, customer map, angles and concepts, the analyzer's reads and transcripts, subscriber counts, and its logo and colours.
  • Disconnect an ad account: we stop reading it, delete the access token within 1 hour of acting on the request, unless other ad accounts on the plan still use that access, and delete the ad account's data within 30 days. If you connect it again before then, with Connect again under Connections, nothing is deleted.
  • Delete your own account: if you're a team member, this deletes your account only, and your brand's data stays with its team. If the brand's Meta connection uses your Facebook login, that connection ends. If you're the brand's only owner, closing your account also deletes the brand's data.

Through Meta

  • If your business connected Parmenion: in Meta Business Settings, go to Integrations, then Connected apps. Choose Parmenion and remove it.
  • If you connected with your personal Facebook account: in Facebook, go to Settings & privacy, then Settings, then Security and login, then Business Integrations. Choose Parmenion and remove it.

If you remove Parmenion in Meta's settings, we stop reading at our next daily read, when Meta refuses our access, and the brand's ad data is deleted within 30 days after that, unless the brand reconnects.

If you commented on a brand's ad, or you're in a file a brand uploaded, see "People who comment on, review or appear in a brand's ads or files".

What deletion covers. We delete everything we got from Meta for that brand or person, everything we worked out from it (labels, summaries, scores, plans and Ask answers) and, when Parmenion offers them, our copies of ad images and video frames. For a brand, we also delete its board and the files its team uploaded. We don't keep summarised or "anonymised" versions. Afterwards we keep only these, and none of them hold the ad data we read from Meta:

  • billing records, for 7 years after the brand's newest billing record, because tax law requires them, with what we check to give each business one free trial;
  • payment evidence: the checkout's IP address and browser for 12 months after the last payment (a checkout that's never paid, 90 days), the rest for 12 months after each charge, and, during a payment dispute, a copy of the brand's sign-in records until it closes, plus 30 days;
  • a record of the request (the date and what kind of data was deleted), for 3 years;
  • security logs, until they age out after 12 months. They no longer link to your Meta account;
  • records that your team accepted our terms and agreed to what we read before connecting Meta, and any analytics choice you made, for 12 months after your account closes;
  • emails you sent us, for 24 months after the last one, unless you ask us to delete them sooner;
  • for a brand on the Parmenion Super Strategy, the list of switch-ins and the owner's switch records, for 12 months after the account closes, and our staff's own notes, for 12 months after the offer ends;
  • anything else the law makes us keep, only for as long as it must.

Security

  • We encrypt data in transit and at rest.
  • We encrypt Meta access tokens with a key kept outside our database, and never write them to logs.
  • We keep each brand's data separate, and check every request against the brand it belongs to.
  • People on our team who can reach your data sign in with multi-factor authentication, and get only the access they need.
  • We keep security logs of important actions.
  • To report a security problem, email sam@chibbsmanagement.io.

No system is perfectly secure. If a breach affects your brand's data, we'll tell your team's owner without undue delay after we become aware of it. If it affects your own account details, we'll tell you as the law requires. If it involves data from Meta, we'll also report it to Meta.

Your rights

What you can ask for

Wherever you live, you can ask us to:

  • tell you what personal data we hold about you, and give you a copy;
  • correct it if it's wrong;
  • delete it;
  • give it to you in a file you can take elsewhere. For your brand's data, the team owner or an admin can use Export everything in Settings, under Your data, which holds the ad data we read from Meta, your brand's settings and answers, your strategies, tags, tag rules, custom metrics, customer map and angles, and, when Parmenion offers the board, your boards and a list of the files on them, or ask us for an export of any of your brand's data;
  • stop using it for analytics, for offering you our larger plans or for marketing emails, and withdraw any consent you gave. This doesn't change what we did before you asked. Every marketing email also has a one-click unsubscribe.

You can ask us to change your account details, or change them yourself where Settings lets you.

How to make a request

  • Where: email sam@chibbsmanagement.io. Say what you want, and which account or brand it's about.
  • Checking it's you: we'll confirm your identity, usually by replying to the email on your account. We only ask for what we need, and use it only to check. If you're a commenter, we may ask for details that show the comment is yours.
  • Someone acting for you: an authorized agent can make a request for you with your signed permission, or a power of attorney. We may still ask you to confirm your identity with us directly.
  • Team data: some requests about a brand's data, such as deleting all of it, need the team owner's OK.
  • Data a brand gave us: if we hold your data only because a brand shared it with us (for example, your comment on its ad, or a file it uploaded that names you), the brand decides most requests. We'll pass your request on and help it answer. If you ask us to delete your comment, or a review the brand uploaded as a review, we'll do it within 30 days and tell the brand.
  • Timing: we reply within 7 days, and answer in full within 30 days. If the law allows more time and we need it, we'll tell you why within those 30 days.
  • Cost: it's free.
  • If we can't do it: we'll tell you why. For example, the law may require us to keep billing records.
  • No penalty: we won't charge you more, give you a worse service or refuse you service because you used your rights.

California

If you live in California, the California Consumer Privacy Act (as amended by the California Privacy Rights Act) gives you the rights to know, access, correct and delete your personal information, to opt out of its sale or sharing, and to limit the use of sensitive personal information. It also protects you from being treated differently for using them.

What we collect, where it comes from and who gets it:

CategoryWhat it includesWhere it comes fromWho we disclose it to, for business purposes
IdentifiersName, email, phone number (for founding access), IP address, account and device IDs, and, to give each business one free trial, website and email domains and Stripe's code for your cardYou, your browser, Stripe, Meta (the ID for your business)Our hosting, database, email, analytics and payment providers; Meta (business name and contact, if it asks for our customer list); our professional advisers
Customer recordsName, email, phone number (for founding access), billing addressYou, StripeOur hosting, database and payment providers; our professional advisers
Commercial informationYour plan, invoices and payment statusYou, StripeOur database and payment providers; our analytics provider (plan and amounts, never names or card details); our professional advisers
Internet or other network activityPages, clicks, scrolls, features used, error reports, server logs, survey answers, your country, Parmenion AI's counts and cost, what you type to Parmenion AI, device and browser detailsYour browser and our appOur hosting and database providers, our analytics provider (pages, clicks, scrolls, features used, error reports, server logs, survey answers, country and Parmenion AI's counts and cost, not what you type to Parmenion AI), and our AI provider (what you type to Parmenion AI)
Professional informationYour company and your set-up answers, and, if you join our list on our website, whether you run a brand or work as a strategist and your monthly ad spend range, if you give themYouOur hosting and database providers; our email provider (for the list)
Sensitive personal informationThe access token that lets us read your Meta ad account, which counts as account log-in credentialsMeta, when you connectOur hosting and database providers, stored encrypted
  • We use each category for the purposes in "How we use it", and keep it as "How long we keep data" says.
  • We don't collect precise location or biometric data, and we don't use a face or voice to identify anyone.
  • We haven't sold or shared personal information collected through Parmenion or its website, and we don't know of selling or sharing the data of anyone under 16.
  • We use sensitive personal information only to provide Parmenion, as the law allows. We don't use it to infer anything about you.
  • This table covers data we collect for ourselves. For data we process for a brand (its ad data, comments, reviews and the files and ads it uploads), the brand is the "business" and we are its "service provider". Please ask the brand, and we'll help it answer.
  • To use your rights, see "How to make a request".

EU, EEA and UK

If you're in the European Economic Area or the UK, this section also applies to you.

  • Who's responsible: CHIBBS MANAGEMENT LLC is the controller for your account data and our website. For data a brand connects or uploads, the brand is the controller and we're its processor.
  • What you must give us: to use Parmenion, you need to give an email address, your brand's name and the country your business is based in. To see your own brand's numbers, you also need to connect a Meta ad account. A paid plan also needs billing details. Everything else, including the set-up answers, is optional.
  • Automated decisions: we don't make decisions about you based only on automated processing that have legal or similarly significant effects.

Why we're allowed to use your data (legal basis):

Why we use itLegal basis
Running your account, connecting Meta and giving you ParmenionLegitimate interests: giving the business you work for the service it signed up for. If you signed up as a sole trader, contract
Billing, tax and accounting recordsOur legal obligations, and legitimate interests: billing the business that signed up
Giving each business one free trial, by matching website and email domains and Stripe's card code against earlier trialsLegitimate interests: preventing repeat trials
Keeping payment evidence to answer disputesLegitimate interests: defending payments against disputes, and legal claims
Security, bot checks and stopping abuseLegitimate interests: keeping Parmenion and its users safe
Product analytics and improving ParmenionYour consent where the law requires it (such as for analytics cookies). Otherwise legitimate interests: knowing how the app is used
Deciding who might want to hear about our larger plans, from your set-up answers and how you use ParmenionLegitimate interests: offering relevant plans to business customers
Emailing you after you join our list on our website, and choosing which of our emails fit you from the answers you giveYour consent, which you give by joining and can withdraw at any time with the one-click unsubscribe
Sending marketing emails about launches, guides and offersLegitimate interests: telling people who signed up about our own product, or your consent where the law requires it
Answering support messagesLegitimate interests: answering you and supporting the business you work for. If you signed up as a sole trader, contract
Answering lawful requests, and making or defending legal claimsLegal obligation, or legitimate interests

For data brands connect or upload, we act on the brand's instructions, and the brand decides the legal basis.

Your extra rights. As well as the rights above, you can:

  • ask us to restrict how we use your data;
  • object to our use of it based on legitimate interests;
  • object at any time to our using your data to offer you our larger plans, including deciding who might want to hear about them, or to send you marketing emails. We'll stop;
  • withdraw consent at any time;
  • complain to the data protection authority where you live or work. In the UK, that's the Information Commissioner's Office (ico.org.uk). We'd like the chance to help first.

Complaints (UK). Send data protection complaints to sam@chibbsmanagement.io. We'll acknowledge each one within 30 days, look into it without undue delay, and tell you the outcome.

Data transfers outside your country

  • We're based in the US. Our providers store your data mainly in the US. Some may process it in other countries under their own safeguards, and Anthropic may process AI requests in other countries. Our subprocessors page says where each one works.
  • If you're outside the US, what you give us yourself, such as when you sign up, comes straight to us in the US.
  • For now, we only accept brands based in the United States, apart from founding members' brands. Before we accept brands in the EU, EEA, UK or Switzerland, or a founding member's brand there connects, we'll protect the transfer of their data with our data processing addendum, using the European Commission's Standard Contractual Clauses, the UK Addendum and the Swiss changes. Our providers use the same kinds of safeguards.
  • Ask us at sam@chibbsmanagement.io for a copy of the safeguards we use.

Children

Parmenion is for businesses, and for people 18 or older. It isn't aimed at children, and we don't knowingly collect children's data. If we learn that someone under 18 has an account, or that we hold a child's data, we'll delete it within 7 days.

Comments on a brand's ads can come from anyone. We handle them as described above, and don't try to find out who wrote them.

Brands may not upload files that hold children's details (our terms of service, section 5). If we learn that a file does, we'll delete it within 7 days.

Changes to this policy

  • When we change this policy, we'll post the new version here with a new date.
  • For an important change, we'll tell you by email or in the app at least 30 days before it takes effect.
  • We won't use data we already hold in a materially different way, such as training AI on your brand's data or using it for other brands, unless you agree first.

Contact us