Why Nametags renders on your device, not in our cloud

When a security team first looks at Nametags, the question is always the same: this thing touches our meeting video. Where does the video go?

The answer is nowhere. Your video is processed on the employee’s own computer and goes from there straight to the meeting app, the same way it would from a webcam. It never reaches us. That was the most important decision we made when we designed the product, and this post explains why.

Two ways to put something on a video call

There are broadly two ways to add something to a person’s video on a call.

Join the meeting. A bot joins as a participant, or a cloud service processes the video stream and sends it back. This is how recorders and note-takers work. It means a third party receives the audio and video of your calls, and it shows up in the participant list.

Change the video before it leaves the device. Software on the person’s computer adds to the camera image locally, and the meeting app receives the finished video as if it came straight from the webcam. Nothing joins the meeting, and nothing extra leaves the machine.

Nametags only ever does the second.

How it works

Nametags reaches your calls in one of two ways, and neither sends video to us.

The desktop app reads your real camera, combines it with your tag, background blur and meeting background on your computer, and offers the result as a camera called Nametags. You pick that camera in your meeting app, and the finished video goes from Nametags to the meeting app and nowhere else. It’s 1280 × 720 video at 30 frames per second, and all of it is produced locally.

The Zoom app runs in a panel inside the Zoom client. Nametags draws your tag as an image and hands it to Zoom, and Zoom adds it to your video using its own video effects, inside the Zoom client.

In neither case is there a bot, a participant, or a stream of your video heading to Nametags.

What we do store

A nametag needs some data, and we’re specific about what that is. The full list is in Privacy and data. In short:

• Nametag fields: name, title, company, department, logo and where the tag sits. These come from what the person types, their Zoom profile, or your identity provider.
• Account details: email address, how the person signs in, and their organization, group and role.
• Templates and brand assets your admins set up.
• A meeting background and a saved frame, only if the person sets them. Save frame is the one feature that uploads a still image, and only when the person clicks it.
• Usage events, such as app opened or call started and ended with its length and frame rate.

And what we don’t store: no video, no audio, no recordings, no transcripts, no participant lists.

Usage events never contain free text. When someone edits their tag, the event records which fields changed, not what they changed them to. An automated test enforces this on every change to the code, so it can’t slip in by accident.

Why we built it this way

The best way to protect data is not to collect it. Meeting video is about as sensitive as company data gets. Board calls, customer negotiations, performance reviews and interviews all happen on video. A design that never receives that video doesn’t need to be trusted with it.

It makes the security review short. Much of a typical vendor questionnaire is about how a vendor stores, encrypts, retains and deletes the data it processes. When the vendor never receives meeting content, most of those questions don’t apply. That’s the reason our security page leads with what Nametags doesn’t do.

It works everywhere. Because the result is a camera, Nametags doesn’t need an integration with each meeting app to work in it. Zoom, Teams, Google Meet, Webex, Slack huddles, and recording tools such as OBS all see a camera and use it.

It’s fast. There’s no round trip to a server, so there’s no added delay and nothing to fall over when the connection is poor.

What it means for your rollout

For IT and security teams, the practical result is a smaller set of decisions:

• Approve the app. The desktop app is signed, and the macOS camera extension can be pre-approved by policy so people skip the System Settings step.
• Decide how people sign in. Single sign-on can fill nametag fields from your identity provider at each sign-in. See SSO field mapping.
• Decide what admins control. Policy decides what each field may show. A field your organization hides is never sent to the app at all.

If you’re running a review now, send us your questionnaire. We’ll return it completed with the SOC 2 report and DPA attached.