Advanced Considerations in ad server Cookie Level Data


Ad Server cookie level data offers unrivaled granularity when it comes to machines digesting the ad server output as rows and columns provide the metric insights occurring on a cookie level and for each action related to that cookie, be it impression, click, interaction or conversion. These reports are large databases outputted daily or more frequently from the ad server and cookie level reporting (or CLD) is a common feature.

The following section looks into detail about some of the the more interesting nuances of cookie level data reports.


Reporting on Conversion tag Actions

Granular reporting beyond aggregated reports are referred to by some ad serving providers as “Raw Data Reports” or “Data Pass Back” reports, and offer a granularity such as an aggregated by conversion tag extended data custom parameters or a list of conversion tag “Actions”; showing the number of times that a conversion tag has fired in addition to conversions. A report of this nature can be crucial to troubleshooting the behavior of conversion tags as it shows the number of times a conversion tag is loading and not just the total sales or counts attributed to traffic to the conversion tag, coming from an ad.


Advertisers that look at this data are able to make a better inference about the impact of online tracked activity because of how many non-conversion Actions are taking place on the conversion tag. If the number is very small then the Advertiser-side analyst can be assured that the majority of sales are being tracked and attributed to paid-for activity online. If the number of Actions is very high then it becomes tougher to identify the sale attribution for untracked channels and may indicate a need to extend tracking to other channels or marketing activity.


Conversion-based extended data at the Cookie Level

Cookie level reporting is also one of the few places where extended data parameters from the conversion tags can be captured. This would include the OrderID, Quantity of items purchased, and revenue generated, as well as customised parameters such as the names of the products purchased or their properties. These properties are presented on each row of the cookie-level data within additional columns featured after the conversion tag name and type, where a cookieID has performed a conversion activity.


To enrich the cookie level data further, an Analyst at the Advertiser or Agency side can marry the OrderID field between the third party ad serving data and the Advertiser POS or CRM. This is best achieved bringing the two data sets together in a Data Management Platform. Bringing these data sets together allows an Advertiser to build a full map of specific customer behaviour, since the cookieID becomes enriched with data that might offer the chance for the analyst to perform a method of profiling.

For example, the OrderID could indicate from the data in the CRM that the Order belongs to an account ID which had undertaken purchase actions over the last few years to indicate a compact buying pattern in the lead up the Christmas holiday with the same Advertiser. Using the ad serving data, the cookieID may have undertaken a certain pattern of site visits displaying the Advertisers ads before purchase. The analyst could see if there was a trend between users of look-a-like profiles and decide whether or not to inform the planners and buyers to consider buying more inventory on such sites closer to the end of the year to attract the buying behaviour from these users and similar users. Alternatively creative messaging could be adapted using DCO to perform a targeting function across inventory that had already been purchased on such sites.

Impression and Click based token data at the Cookie Level

Publishers can pass information into the impression call or the click pathway of the ad by populating a token field in the ad tag code when the ad tag code is fired. This is executed by a Publisher created script that sits on the Publisher page. Any extra information passed through in either of these dynamic token fields can be used to populate additional columns in cookie level data. Both of these fields are highly favoured by marketers that want to get a better idea about exactly where the ad appeared on the page. Since the Publisher ad ops team is the one implementing the ad, even page scraping the domain or page level URL where the ad was displayed is not enough information to convey where about on the page the ad sat. These fields can be populated by placement position information such as ‘position 4’ so that when analysing the cookie-level data, the analyst can determine if ad performance or verification are affected by the position of the ad spot.

These advanced strategies take careful planning analysis. Some Agencies or marketing teams hire Data Analysts who process such data sets full time, for the purposes of campaign optimization. Analytics come with an important cost consideration when purchasing the media in the first place, so the full budget should not be blown on ad space but also on interpreting and presenting the results. Even bolting on other technologies like third party data providers instead of physical analysts to get this data insight, can add significant cost and therefore affect the bottom line.

To find out more about cookie level data and more simplified forms of reporting get a copy of: Ad Serving Technology – Understand the Marketing revelation that commercialized the Internet – available now from..

….Amazon UK:

Or Amazon USA:

No comments.

Leave a Reply

%d bloggers like this: