Share your Meta click history

AI Ads Audit reviews the Meta ad clicks you pay for and flags invalid traffic. The pixel only sees clicks from the day it goes live. Your past clicks already sit in tools you run, mostly Google Analytics. Start any export you send us from 1 Jan 2025.

What we need

3 fields per click, and nothing else:

Field	Example
Landing URL with `fbclid`	`https://yourstore.com/products/linen-shirt?utm_source=facebook&fbclid=IwAR2fX9kQ7sLmP4vB8n`
Timestamp	`2025-01-14 09:12:44 UTC`
Visitor IP	`203.0.113.7`

No names, emails or order data.

We download these records to build evidence files for your Meta clicks. The access you give us can’t change or delete anything on your side, and you can remove it at any time.

Two routes: pick one or both

Share your GA4 click records

Landing URLs and timestamps from your GA4 BigQuery export. Start here.

Add visitor IPs from other tools

GA4 doesn’t keep IPs. Amplitude, PostHog, your CDN or server logs often do.

Either one is enough to start.

How to send it

Read-only access is enough. Use whichever route fits where the data lives.

BigQuery dataset

Dataset > Sharing > Permissions > Add principal. Add the Google account your Dash.fi contact sends you with the BigQuery Data Viewer role. We run queries from our own Google Cloud project, so this adds nothing to your bill.

Google Cloud Storage bucket

Bucket > Permissions > Grant access. Add the same Google account with the Storage Object Viewer role.

Amazon S3 bucket

Bucket > Permissions > Bucket policy, then add this statement. Replace YOUR_BUCKET with your bucket name. Your Dash.fi contact sends our AWS account ID.

{
  "Version": "2012-10-17",
  "Statement": [{
    "Sid": "DashfiLogReadAccess",
    "Effect": "Allow",
    "Principal": { "AWS": "arn:aws:iam::DASHFI_AWS_ACCOUNT_ID:root" },
    "Action": ["s3:GetObject", "s3:ListBucket"],
    "Resource": ["arn:aws:s3:::YOUR_BUCKET", "arn:aws:s3:::YOUR_BUCKET/*"]
  }]
}

If the bucket uses SSE-KMS encryption, also allow our account kms:Decrypt on that key.

Snowflake

Create a read-only role with SELECT on the schema that holds the data, and a user with that role. If both accounts sit in the same cloud region, a secure share to our account works too.

Cloudflare R2

R2 > Manage API tokens > Create API token. Permission: Object Read only, applied to that one bucket. Send the access key ID, secret and S3 endpoint through a password manager share.

Azure Blob Storage

Container > Shared access tokens. Permissions: Read and List. Set an expiry date, then generate the SAS URL.

Files (CSV, logs, exports)

Share them from Google Drive with the Google account above, or ask your Dash.fi contact for a secure upload link. Keep original .gz and JSON files with their names unchanged.

Never send keys, tokens or passwords by email. Use a password manager share.

Then reply to your Dash.fi contact with what you shared and where. We confirm once we can read the data.

Questions

What will you use this data for?

Only to build evidence files for the Meta clicks you paid for, as part of AI Ads Audit. See data collection and privacy.

Can we remove access later?

Yes. Remove our account from the dataset or bucket at any time. That stops new downloads. Records we already downloaded stay in the evidence files we build for you.

Do you need a DPA?

Visitor IPs count as personal information under laws like the CCPA. If your team needs a data processing agreement before sharing, ask your Dash.fi contact.

We don't use any of these tools

AI Ads Audit still covers every click from the day the pixel goes live. Link GA4 to BigQuery today so new clicks are also kept on your side, and tell your Dash.fi contact.

​What we need

​Two routes: pick one or both

Share your GA4 click records

Add visitor IPs from other tools

​How to send it

​Questions

What we need

Two routes: pick one or both

How to send it

Questions