Privacy Policy

This Privacy Policy explains how Drove Group, LLC, a California limited liability company (“Drove IQ,” “we,” “us”) handles personal information in the Drove IQ service and on droveiq.com. Drove IQ is a product of Drove Group, LLC.

For a company portal, the company is the customer. We process workplace records — roster, scores, plans — to provide that company’s service. For a person who starts on their own, we hold the account they created until they join a company or ask us to delete it.

Questions or requests: hello@droveiq.com.

Assessment scores, levels and reports are for learning and development only. They are not designed, validated or intended for hiring, compensation, promotion or other employment decisions, and customers agree not to use them that way.

We collect and show scores so people can learn and so a company can run a development programme. We do not offer, and the product is not built as, an employment-decision or credit-decision tool.

• Account information. Name, work email, and the one-time codes used to prove that inbox. Stewards and people taking the assessment both sign in this way. There is no password.
• Assessment responses. The answers a person submits, including free-text reflections, and the scores and level we compute from them.
• Activity. Plan progress, level-exam attempts (a fail is stored as an attempt, not as a new score), and product events we write ourselves so we can see whether a flow was used. We do not use a third-party analytics product.
• Roster imports. Names, work emails, job titles and teams a steward uploads or types so the company can invite people.
• Company setup. The company’s name, branding, teams, sanctioned tools, written rules, and — if they upload one — an AI-policy file and a logo.
• Ideas and examples a person writes. Those land on the company’s Opportunity Map and Realized wins.
• Billing. If the company pays, Stripe holds the card and the invoice. We store Stripe customer and subscription identifiers, seat quantity, and plan status — not the full card number.
• Technical data. Browser and device information the host sees in ordinary request logs, and first-party cookies needed to keep a signed-in session.

We do not ask for government ID, date of birth, payment card numbers, or health information. We do not run advertising pixels or a third-party analytics SDK in the product.

• To score the assessment, show a person their result and plan, and show a company its portal.
• To send the emails the product actually sends: sign-in codes, invites, personal-plan links, exam-ready notices, and plan nudges.
• To bill a paying company and keep seat counts in line with Stripe.
• To operate, secure and debug the service, and to meet a legal request we cannot refuse.

We do not sell personal information. We do not use it for advertising. We do not train our own models on customer answers.

This is what the product shows a company about the people it assesses.

• Leaders always see the aggregate picture and who has finished.
• On some plans (Pro), leaders can also open an individual’s readiness scores, level and plan progress, with that person’s name. On the free plan the portal shows the company average only.
• On the same plans, what a person writes about where AI could help and where it has already changed their work appears on the company’s Opportunity Map and Realized wins, with that person’s name.
• Role-task picks stay private. The app does not collect working notes from a personal plan.
• Marking a role briefing as read is the one Role fact that crosses to the company’s Progress view. It records the reading, not the picks.
• Drove operators (platform admins) can open a portal to run the service. That access is limited to a short, named list.

These are the companies that actually receive customer data from this product. We list only what the code calls.

• Supabase — database, authentication, and file storage (logos and uploaded AI-policy files). Data is encrypted at rest (AES-256) and in transit (TLS).
• Vercel — hosts the application and the marketing site, and therefore sees ordinary request logs.
• OpenAI API — scores free-text answers, groups ideas and examples into themes, and generates plans, briefs, suggested rules, leader messages and on-demand role research. Customer content sent to the API is not used to train OpenAI models by default, and OpenAI retains API content for up to 30 days for abuse monitoring. We do not send a person’s name or company name in the scoring prompt when the code can avoid it. Grouping ideas into themes sends each person’s name and team with what they wrote; roster context may identify a workplace in other features.
• Stripe — payment, invoices, the customer portal, and subscription status. The app never stores a card number.
• Resend — sends the product’s email.

The marketing site (droveiq.com) also loads fonts from Google Fonts. That is a request from the browser to Google, not a copy of assessment answers. The application itself hosts its fonts.

When Role research runs, our servers fetch a public page to check a figure and a quote. Those hosts see a request from us, not the person who typed the job title. The cache is keyed by the job title, not by a person or a company.

We do not use PostHog, Google Analytics, Mixpanel, Segment, Sentry, or Vercel Analytics in this codebase.

• A signed-in session is a first-party cookie set by Supabase Auth.
• A short-lived cookie marks when a steward is taking their own assessment from their portal, so the header can offer a way back.
• The assessment in progress is held in the browser’s session storage so a refresh does not lose answers. It is not a tracker.
• A few first-party local-storage flags remember that a person has already seen a one-time celebration on their plan. They do not leave the device.

We do not set advertising cookies.

We keep a company’s records for as long as the portal exists and the company uses the service. There is no automatic deletion job in the product today.

• A steward can erase one person’s assessment and plan. The roster row can remain, with no assessment attached. A log that the deletion happened is kept, with the email, so the act can be attributed.
• Marking someone departed keeps the record and ends their access.
• Deleting a company is an operator action. It is refused while Stripe is still billing that company.
• OpenAI may retain API content for up to 30 days. Stripe and Resend keep the records they need to bill and to deliver mail.
• Role-research findings are cached by job title for up to 90 days (or 14 days when a search found nothing).

You can ask for access to, correction of, or deletion of personal information we hold. Email hello@droveiq.com.

If your company invited you, that company can erase your assessment from its portal. The product’s own “remove my record” path writes to your company, not to us, because they hold the workplace record. You may still email us; we will need the company’s help to finish a workplace deletion.

If you started on your own and have not joined a company, email us at the address above and we will delete the account we created for you, unless we must keep a record to meet the law or resolve a dispute.

Depending on where you live, you may have further rights (including to object, to restrict processing, or to complain to a regulator). We will honour a request the law requires.

Access to another company’s data is denied by database policies keyed on the company. The browser never holds a database key. File uploads (logos, AI-policy documents) go to Supabase storage on the company’s path. Supabase encrypts that data at rest (AES-256) and in transit (TLS). We use HTTPS in production. No method of transmission or storage is perfectly secure.

Drove IQ is a workplace product. It is not directed at children, and we do not knowingly collect personal information from anyone under 16.

We may update this policy. The “last updated” date at the top will change. The current policy will be posted at /privacy.

This policy is governed by the laws of California.

Privacy questions and deletion requests: hello@droveiq.com.