Metrics, Dimensions and Datasets
Whenever you connect to a Google Analytics or Paid Advertising data source, you get access to many predefined datasets made up of metrics and dimensions.
It is not possible to have just one super dataset that combines all metrics and dimensions. This is because programmatic data interfaces (APIs) do not allow us to retrieve all combinations of dimensions and metrics in one query. Usually, it is only possible to retrieve a limited subset of compatible dimensions and metrics.
The following image shows Bing Ads datasets (a Custom Data Stream is being created based on one of them):
In that example above, there are 3 datasets each with a subset of compatible dimensions and metrics:
- Campaign, Device dimensions
- Campaign, Ad Group
- Campaign, Ad Group Ad
Octoboard PPC Data Analytics Add-on offers a collection datasets for all integrations that we support. When you create a Data Stream, you always base it on a dataset and “inherit” its metrics and dimensions.
Datasets are important because they help you group compatible metrics and dimensions. They are also useful when you create a new metric based on an existing data field. For example, when creating a new CPC metric in a Custom Data Stream, you can base it on the standard Cost Per Click formula stored in the default dataset.
The process of creating new metrics and dimensions is covered here
Data metrics are numeric data fields retrieved via programmatic interfaces (APIs). They usually represent time and date, number or financial data values. Every data source such as Google Ads, Bing Ads and others has different metrics but the following metrics can be expected in all PPC platforms:
- Impressions
- Clicks
- Cost
- CTR
- CPC
- CPM
Here they are in a default Data Stream of the Bing Ads data source:
Octoboard allows you to edit any rule of any metric in PPC Data Analytics Add-on. Click the EDIT icon next to the metric and you will be presented with an interface that would allow you to change the metric definition. Here you can see the metric definition of the Bing Ads’s Click-Through Rate (CTR):
For more information on how to modify existing metrics in PPC Analytics datasets, please this tutorial
PPC platforms such as Bing Ads have dozens of metrics but since the common metrics can be found in all platforms, they are automatically mapped by Octoboard in the Combined Multi-Channel dataset.
PPC Data Analytics Add-on ensures, for example, that ad spend values from Facebook Ads, Google Ads, LinkedIn Ads all end up in the Cost field of the Multi-Channel Data Stream. Since this stream is always created by Octoboard it always automatically collects data from all connected PPC platforms to create a cross-platform view of the advertising expenditure.
The following image shows ad spend per country across all connected paid advertising channels.
For more information on how metrics are mapped to the Multi-Channel Data Stream, please this tutorial
Dimensions are text-based data values such as Country, Campaign or Ad Group that are used to aggregate data and offer powerful analytics capabilities and data breakdowns.
Here is an example of Google Analytics and PPC metrics broken down by Country and PPC Channel:
Every dimension definition can be modified. Here is an example of the Bing Ads’ Campaign dimension definition. In this case it simply uses a data source value from the Bing Ads.Campaign field.
Similar to metrics, PPC platforms such as Bing Ads have dozens of dimensions but since the common dimensions can be found in all platforms, they are automatically mapped by Octoboard in the Combined Multi-Channel dataset.
These are the main common dimensions that are used to build global PPC breakdowns across data sources and accounts:
By using common dimensions, such as Country or Campaign, Octoboard will try to combine Google Analytics Session data with standard PPC metrics. The following widget shows metrics from Google Analytics (Sessions) and Paid Advertising platforms (Transactions, Impressions, Clicks, Cost) broken down by Markets (Country) and Marketing Channels (Channel).
Values for the common dimensions in the Multi-Channel DataStream are mapped from PPC and Google Analytics datasets. Here is an example of the Country dimension in the Combined DataStream that uses Country fields in PPC datasets of Google Analytics, Google Ads and Facebook Ads.
For more information on how dimensions are mapped to the Multi-Channel Data Stream, please this tutorial
