Best Practice Tips

1. When to use a Dashboard level filter and when to use a Report level filter
2. When to Disconnect and/or Delete a connection
3. Creating Custom Metrics

When to use a Dashboard level filter and when to use a Report level filter

Sometimes, using a dashboard level filter instead of a report level filter, or vice versa, can cause some unwanted filtering or incompatibilities - causing your data to look different to what you would expect. Hence, it's important to know when to use the appropriate filter - here are a couple suggestions:

• If you have a dashboard for a specific brand, we'd recommend using a dashboard level brand filter which equals that specific brand. That means you won't have to create the same filter for every report within the dashboard. 
• If you are using a dashboard, there is a main date range that you can use. This main date range will affect any report (within the dashboard) which has its date range set to "Default". In most cases, we would recommend you set a date range for each individual report, e.g. if you want a report within the dashboard that is relevant to September, you can set a report-level custom date range for September - this will override the master date range. 

Disconnecting and/or Deleting a connection

Bearing in mind that connections are used to pull and store your brand's current & historical data, here's a few guidelines on when to disconnect and when to delete the connection. 

• If the brand has migrated to a new platform, you should only DISCONNECT the old connection. Do NOT DELETE it as you will want to keep the data from the original platform in Affluent. Affluent deletes all data permanently if you confirm a deletion.
• If the brand is migrating to a new platform, only DISCONNECT the connection once you are sure that the platform has ceased tracking - this will ensure you're not missing out on pulling any data. 
  
  
• i.e. only DELETE an account when you are sure that you no longer need/want the data. 

Custom Metrics

• When naming a custom metric, we would highly recommend you don't choose the same name as the native metrics - it might cause confusion when you are trying to select the metric that you want. 
  • Note that one way to tell a custom metric apart from a native metric is that they will have the following sign next to them: , i.e.
• Check that a custom metric is working as expected after you create it. We would recommend you create a very simple report using the custom metric, select a date range, and compare the value with a manual calculation on your end.