---

Ghost Bidding offers a testing framework that enables Beeswax customers to calculate the incremental impact of their media buys without having to pay for control group impressions. In this testing framework, certain bids are not returned to the auction, these are labeled as “Ghost Bids” and Beeswax system predicts if those bids would win an auction. If the Beeswax system predicts that a Ghost Bid would win, it generates a "Ghost impression" and these impressions form the control group. 

Beeswax uses the control group to analyze if there was a difference in conversion rates between users who were exposed to the Ad versus those who were not. It then generates an Incremental Lift analysis for the advertising run that can be seen in the Incrementality Report available under Reporting. 
Conversion is the desired action that a user takes after seeing your ad such as making an online purchase, signing up for a mailing list, subscribing to a service, and so on.  Conversion rate is the ratio of conversions to impressions delivered, shown as a percentage.

Additionally, Ghost Logs are available for users who wish to run elements of the Ghost Bidding methodology by themselves.

Benefits

Ghost Bidding helps advertisers to know if their ads are effective in driving up their conversion rates.

Prior to Ghost Bidding, advertisers used to split their customers into two groups, Treatment, and Control, where the treatment group was shown the advertiser’s ads and the control group was shown PSAs (Public Service Announcements). The testing method was used to evaluate the ads' effectiveness; however, advertisers were responsible for serving the PSAs. Ghost Bidding gives the opportunity to use a testing method evaluating ads effectiveness but without having to spend media dollars on PSA ads. 

How Ghost Bidding Works

Treatment and Control Group Assignments

• Ghost bidding assigns 80% of users in the Treatment Group (real bids) and 20% of users in the Control Group (ghost bids).
• Bid requests get assigned a user index (1 to 1000) using a hash based on the User ID type. User index 1 to 200 = Control Group and 201 to 1000 = Treatment Group.
• Beeswax uses Standard or IP Address for identifying Ghost Bidding User ID Type.
  
  • For Standard Ghost Bidding User ID Type, Beeswax uses cookies.
  • For IP Address Ghost Bidding User ID Type, Beeswax uses both IPv4 (IP version 4) and IPv6 (IP version 6), however, IPv6 takes precedence when both types of IPs are present.
    If a bid request contains obfuscated IPs, then the bid request is not eligible for line items using IP Address as a Ghost Bidding User ID Type.

Ghost Bidding

• All requests go through targeting, creative matching, pacing, and frequency capping to see if a line item bids on it.
• If a line item bids and the user has been assigned a user index between 1 to 200, then the bid gets dropped (not sent to the exchange) and logged as a Ghost Bid. 

Win Prediction

• Beeswax predicts if a Ghost Bid will win and become a Ghost Impression.

Treatment vs Control Group Conversion Attribution

• When Beeswax receives a conversion event, it runs attribution methodology across both Treatment (real) and Control (ghost) groups impressions to determine which group was responsible for driving a conversion.
• Note: This is not supported for in-app/postback/SKAd attribution methods because these attribution processes are run outside of Beeswax.

Reporting & Logs

Incrementality Report

In Report Builder, the Incrementality Report calculates the incremental lift and other key metrics across both treatment and control groups, including impressions, conversion, conversion value, and conversion rate. See Incrementality Report to learn more.

Incremental Lift is calculated using the following equation: (Treatment group conversion rate – Control group conversion rate) / Control group conversion rate 

Ghost Logs

Beeswax also offers Ghost Win Logs for users who would like to run their own attribution methods, and Ghost Attribution Logs for users to run their own incrementality analysis. Please speak to your Beeswax Account Representative to enable Ghost Logs. See Ghost Log Schema in GitHub.

Ghost Bidding Configuration

You can enable Ghost Bidding at the Campaign and Line-Item level, by selecting the ‘Ghost Bidding User ID Type’.
Note: Experiments and Ghost Bidding cannot be used in tandem on the same line items.   

Campaign

On the Campaign > Overview page, you can set the Default Ghost Bidding User ID Type. All line items created under the campaign will have Ghost Bidding enabled and will default to this Ghost Bidding User ID Type setting.

Line Item

On the Line Item > General page, you can enable Ghost Bidding by selecting the toggle to Yes. You can then select the Ghost Bidding User ID Type that will be used to bucket users for ghost bidding.

Ghost Bidding ID Types

Ghost bidding can be configured to use one of the two different ID types:

FAQs

Q: Does the user index for a particular user stay the same across all Ghost Bidding line items?
A: Yes, Beeswax does not change the hash to index a user, so the same User ID gets the same assigned user index. 

Q: Can a customer define the split between treatment and control groups?
A: Currently, all groups are split at an 80:20 (treatment and control) ratio.

Q: Is the Incrementality Report supported across all attribution types?
A: To generate a Incrementality Report, Beeswax needs to see all conversion events to run attribution against Ghost Impressions, therefore, the Incrementality Report does not support any campaigns or line items that are using post-back attribution (mobile in-app and custom attribution methods). For users who want to run their own attribution methods or incrementality calculations, Beeswax offers Ghost Logs.

Q: Can I use different Ghost Bidding User ID Types for different line items in the same campaign? 
A: Yes, but it is not recommended unless you are sure that the different line items have disjoint targeting.

Q: Where do Ghost Bids show up in Bid Decision Flow?
A: In the Post-Processing category under the sub-reason ‘Ghost Bid’.