Although the growth of programmatic advertising allowed Publishers to plug more and more SSPs through header bidding competition and other demand channels, this has brought them even more difficulties to get quick and comprehensive insights about their own programmatic activity.
SSPs usually provide the ability to collect a large amount of data from their servers, owned by the Publisher, generated across many sources with a large volume to process. Despite the fact that the amount of data available gets bigger and bigger, to understand its own programmatic sales activity has become more difficult than ever.
Naming conventions across all SSPs
At Manadge, we strongly believe in the strength of standards to make advertising a better and more transparent industry, for all.
This is why our software and data architecture are based on the following standards :
- Open RTB (Real Time Bidding)
- IAB Taxonomy (Advertiser categories, devices)
- International Standard for country codes (ISO 3166)
- International Standard for currencies (ISO 4217)
So what about dimensions that do not have any standard ?
Are preferred deals (GAM), Direct deals (Smart/Equativ), and private auction fixed price all the same ? Based on the OpenRTB documentation, they all align with the AuctionType=3.
After a long research and development in accurate mapping, we have built independent naming conventions across all SSPs with common dimensions (Placement, Ad Unit, Auction Type name, etc).
This helps our customers get a better understanding of their own data across their ad stack, and an easier onboarding for non-advanced programmatic people.
Mappings across all SSPs
Mappings must happen where Normalization cannot take place, and this is why we’ve built our Mapping Engine : we provide an advanced and powerful tool to the client to let them aggregate any dimension, automatically, in real-time.
As we have privileged insights into premium publishers’ programmatic ad stack, we can certify that there is no publisher’s stack similar to another : they may share similarities regarding integration and else, but everyone has a very different way of working.
It’s the reason why we successfully give our customers the possibility to aggregate and normalize its own data the way it sees it best, as many times as it wants.
Our mapping engine provides automated mappings for all the past and future data once the rule is created. This way, the client does not have to update its mapping every other day : skip the manual tasks !
Let’s take a look at an even more concrete use case : Programmatic Buyers & Agencies. It is one of the most difficult tasks to achieve regarding how SSPs name the programmatic seats buying inventory through their channels. For a publisher, it multiplies the volume by the number of SSP you have.
Below you can see how our generic buyers mapping aggregates dynamically more than 10 000 programmatic seats just for the main 7 french programmatic agencies : Dentsu, GroupM, Havas, JellyFish, OMG and Publicis.
However, we still give flexibility to our customers : Do they want to split Amnet from Dentsu, add Agence79 revenues into Havas, differentiate Starcom from Zenith from Blue449 ? We give the Publisher the choice and freedom to do so.
Who knows better how you want to handle and categorize your own business than yourself ?