Prebid Analytics dashboard

Last edited: 2023/09/25

Prebid Analytics (PA) gives you tools to analyse your performance with many metrics and dimensions, based on Prebid data.

Prebid Stack gives you 2 analytics tools for 2 different sources of data: - GAM reporting, based on GAM data - Prebid analytics, based on Prebid data For data from your Own bidders, Prebid Stack does not have an access and you need to connect to your Bidders account directly.

Analytics sub-modules:

• Dashboard.
• General dashboard
• Performance per bidders
• Performance by sites
• Performance by ad units
• Performance by ad sizes
• Performance by devices
• Performance by countries
• Ad type analysis
• Bidder analysis
• Bidder list
• General charts
• Incremental revenue
• Bidder response time
• Site analysis
• Auction reports
• Revenue reports
• A/B Testing

If you want to download data in .csv form, you can do so through the Auction reports or Revenue reports tabs.

How to use Prebid Analytics?

Activating Prebid Analytics (PA)

If you did not activate PA on domain level, your data will be sampled at 10%, hence not precise.

To activate PA:

1. Go to User Workspace > Add-on, and turn the PA toggle on.
2. For each of the domains you wish to activate PA, go to Configuration > Domains > Edit domain > Edit domain configuration, and turn the PA toggle on.

How to access Prebid Analytics?

Prebid Analytics is accessible from your left tab panel: Performance > Prebid Analytics

General metrics and dimensions

What are the metrics available?

General revenue

• Gross revenue: sum of impressions CPM / 1000
• Bid requests: number of bidder requests
• Requests: number of ad requests (1 ad request can have more bid requests upon the number of bidders connected to 1 ad unit)
• Impressions: number of impressions
• Bid price: average CPM for all bids
• eCPM: average CPM for all impressions
• Bid rate (%) = bids / bid requests
• Win rate (%) = wins / bids
• Timeout rate = bid timeouts / bid requests

Bidder information

All the above metrics are available on a bidder level… and more:

• Requests / bidder
• Bids / bidder
• Bid Rate / bidder
• Win Rate / bidder
• Bid Load Time / bidder
• Timeouts / bidder
• Timeout Rate / bidder
• Response Time / bidder
• Incremental revenue / bidder
• Unaffected revenue / bidder

What are the dimensions available?

You can look at your metrics filtering your data across many dimensions:

• At the top: filtering for an applicable time frame or currency
• On the right hand side:
• Sites: enter your domain(s)
• Bidders: enter the bidder alias(es)
• Ad units: enter the DIV Ids of the ad unit(s)
• Devices: select the type of device
• Countries
• All versions. Versions are mainly used for AB testing.

For more information: and A/B testing

Choose the versions you want to see. Note that the data will be the aggregated data for all the selected versions. If you want to compare several versions, go the the A/B testing tab.

• User consent. This is related to the feature “Ad serving for users without consent”. Choose between the 2 modes: User <Consent> and <No consent>
• Prebid analytics mode: if you are using prebid analytics, your data is 100%. If you are not using it, it is sampled 10%.

Dashboard

The dashboard provides you the most general Prebid performance data with the main metrics and dimensions.

Data sampling means that you get results from a portion of page views. The default setting is a 10% sample and this is sufficient to draw valuable conclusions, especially for domains with a lot of page views. If you want to see 100% data, contact your Customer Success Manager.

The Expected Gross Revenue is based on the revenue generated in this time frame, 1 week before.

Ad type analysis

Analyse the ad performance in terms of bids, wins, impressions, revenue, eCPM, bid prices, and win rates, and compare it to the type of ad (banner and video).

Bidder analysis

The bidder analysis is meant to analyse a bidder performance. It encompasses 4 sub-tabs:

• Bidders list
• General charts
• Incremental revenue chart
• Bidder responses times.

Bidder list

General charts

It includes the revenue, bid rate and timeout per bidder.

Incremental revenue

The incremental revenue chart is very useful to evaluate the impact of a bidder. It has to be put in parallel with the Unaffected revenue.

Incremental revenue = sum of all bid revenue won by the bidder - sum of all 2nd highest bids proposed by other bidders.

Unaffected revenue = sum of all 2nd highest bids proposed by other bidders.

In other words, the incremental revenue is the revenue that was generated exclusively thanks to the fact that the bidder was connected (otherwise, you would not have generated such revenue).

The unaffected revenue is the revenue generated by the bidder 1 which could have been generated by another bidder 2 should the bidder 1 not have been connected.

Site analysis

Auction reports

You can view the report within the interface or download the auction data in the form of a .csv file. You decide the grouping variable(s), and the metrics you would like to see.

Revenue reports

As for the above auctions, the revenue generates reports which you can either look at within the interface or download them as an .csv file. You decide the grouping variable(s), and the metrics you would like to see.

A/B Testing

The A/B testing analytics sub-module allows to compare various versions, and the desired metrics across the usual dimensions.

The A/B testing analysis is limited to up to 4 versions at a time. So, if you are running a test involving more than 4 (example 6 versions v1-v6)  then you have to choose 4 from any of these 6, (example v1, v2, v5, v6).

On this page:

• Prebid Analytics dashboard
• Analytics sub-modules:
• How to use Prebid Analytics?
• Activating Prebid Analytics (PA)
• How to access Prebid Analytics?
• General metrics and dimensions
• What are the metrics available?
• What are the dimensions available?
• Dashboard
• Ad type analysis
• Bidder analysis
• Bidder list
• General charts
• Incremental revenue
• Site analysis
• Auction reports
• Revenue reports
• A/B Testing