How to report data integrity (mismatch) issues in Windsor.ai

In case you are facing data integrity issues when using a Windsor.ai connector, please start with this troubleshooting guide: https://windsor.ai/documentation/how-to-fix-data-discrepancy-issues/.

If the issue persists, contact our customer support for further assistance by preparing a thorough report and sending it to [email protected].

Please, follow this quick guide to effectively report on your data integrity problem. In this format, our team will be able to identify the reason much quicker.

Before reporting any issue, please make sure that the metrics and dimensions you compare in our connectors versus your platform are exactly the same ones (e.g., some users compare URL clicks against clicks).

The combination of fields also matters: Do make sure that, for example, you do not compare source/medium, users, or date in our connector with sources/medium and users in Google Analytics 4, as the data will be different.

When reporting an issue it helps us a lot if you submit it in the following format:

• Description of the issue

For example, My Looker Studio Facebook Ads dashboard (insert a link to dashboard) is showing me incorrect numbers: I’m expecting 1025 clicks for the date range from 1/10/2023 to 7/10/2023, but I see 1200 instead.

• Actual result (how the issue looks on your side)

In this dashboard, on page 2, the number of clicks is 1200 for the date range 1/10/2023 to 7/10/2023.

• Expected result (data from the original source)

See the attached screenshot from the Facebook Ads business manager for the same date range. (Important: Include the time range in the screenshot, the account ID or account name, and the relevant metrics).

Sample screenshot:

In case you face the issue in Looker Studio or Google Sheets, please share the report with [email protected] and provide edit access so that we can quickly investigate it.

In case you face the problem in other destinations like Power BI, Excel, BigQuery, Snowflake, Azure SQL, Amazon S3, MySQL, PostgreSQL, Databricks, or Snowflake, make sure to include the connector URL in the description of the issue so that we can quickly reproduce it and solving the problem for you.

Usually, investigation of data integrity issues takes time (1-3 days), so it’s better to organize the report and send it to email [email protected]. We’ll take a detailed look into it and reply to you asap as soon the resolution is found.