How we keep your data safe
You trust us with your tracking setup. Here is what we ask for, what we keep, and how you stay in charge.
What we ask for, and when
We don't ask for everything at once. We ask for each kind of access only when you turn on the feature that needs it. Google shows you each request, and you can say no.
- When you sign in
- Your basic Google profile: your name, email and photo. Nothing else.
- When you connect containers
- Read-only access to Google Tag Manager. We can look at your setup, but we can't change it.
- When you watch leads and sales
- Read-only access to Google Analytics. We read hourly counts of your key events.
- When you ask us to add the monitor tag for you, or turn on AI drafts
- Edit access. The monitor tag is an optional tag that counts which of your tags run. An AI draft is a change your AI helper suggests. We put either one in a separate draft space. It does not go live from there. If you add the monitor tag yourself, we don't ask for this.
- When you go back to an old version, or approve an AI draft
- Access to make and publish versions. We use it only when a person on your team clicks to confirm.
What we keep, and how we guard it
What we keep
- Each version of the containers you watch, and what changed.
- The results of safety checks and browser checks, with screenshots.
- Hourly counts from Google Analytics and the monitor tag. Counts only, never people.
- Your Slack team and the channels you pick. We post there. We don't read your messages.
- A log of what your AI helper did.
How we guard it
- The keys to your Google and Slack accounts are encrypted before we save them. That means they are scrambled with strong code, called AES-256.
- Everything travels over HTTPS, the safe kind of web link.
- Your data is walled off by team. Every page, alert and AI helper key only shows your own team's data.
- We host on Cloudflare. Only the TagStrike team can reach the live system.
What the monitor tag sends
The monitor tag is optional. You choose to add it. We can put it in a draft for you, but we never publish it for you. It helps us see if your tags run for real visitors.
It sends
- Which tags ran on a page. It uses each tag's ID number, not its name.
- The names of events that happened, like a purchase, and how many.
- Your container ID and live version number.
- It sends from 1 in 10 page views by default. You can pick 1 in 100, 1 in 4, or every page view.
It never sends
- Names, emails or other personal info.
- Web addresses of the pages people visit.
- Cookies or what people type in forms.
- Any ID that could follow a visitor.
Like any web request, it reaches our host with the visitor's IP address and browser type. We use the IP address only to block abuse. We don't save either one with your counts.
What your AI helper can and can't do
You can let your own AI assistant connect to TagStrike through MCP. MCP is a common way for AI tools to link up with apps. You make a key for it in the app, and you can cancel that key at any time.
It can
- Read your containers, versions and alerts.
- Draft changes in a separate draft space.
- Send a draft to your team for review.
It can't
- Publish anything. A person must approve it in the app.
- Delete containers or versions.
- Change who has access.
We log every step it takes, so you can see what it did.
How to remove access and delete your data
- Stop watching a container at any time in the app.
- To cut us off from Google, remove our access on your Google account permissions page. We can no longer see your containers, so we stop looking for new publishes.
- To delete your workspace or your account and all your data, email jake@citamark.com. We will confirm within 30 days.
Our privacy policy says how long we keep each kind of data.
Questions?
Email us at jake@citamark.com. Think you found a security problem? Please tell us there too.