Online advertising in its most primitive form involves putting a creative (text, image, video) in a designated area, i.e. ad slot, on a publisher’s webpage.
The creative can come from the same server as the one that is serving the rest of the page content, or more likely, a dedicated ad server.
The fact that the creative can be chosen at runtime (i.e. when the webpage loads) takes online advertising to a whole new level. It gave birth to AdTech. AdTech is like eBay, AirBnB, and Uber. It is using technology innovation to provide better connectivity between providers and consumers. Better connectivity leads to a bigger market. A bigger market leads to more competition which is a win-win for both providers (i.e. publishers) and consumers (i.e. advertisers). With AdTech, publishers can maximally monetise their ad-slots by having a greater choice of advertisers bidding against each other. Advertisers can increase their cost-effectiveness by having a wider range of publishers to choose from.
Another aspect of AdTech is the research of using data to optimise the accuracy of targeting. It is like the study of weather prediction where new methods are being discovered all the time that increases accuracy of prediction. From the wind indicator, to barometers, to satellite imagery, to running supercomputer simulation. In AdTech there has been the use of tracking cookies, user profiling databases, and machine learning.
Significant developments in AdTech
Programmatic
Programmatic is the Uber-equivalent in advertising. Just like Uber matches a rider with a driver, in programmatic advertisers specify their requirements and they get matched to a publisher’s ad slot automagically.
Real-time bidding and prebid (aka header-bidding)
A publisher’s ad slots can always show the highest paying ad because at page load time an auction is run within seconds to find the highest bidders for ad slots. The auction platform is called an ad exchange. Prebid takes this one step further by running concurrent auctions with multiple exchanges in order to get more bidders.
Leveraging data
Data can help advertisers predict how likely a user is to commit to buy. This is an area of ongoing innovation. Some of the frequently used methods are:
- cookie tracking and cookie syncing — this is what results in the situation when you looked at a product you wanted to buy and suddenly you see ads on every website you visit.
- data platforms — as a web user goes about in her day-to-day browsing some of her browsing activities can be used to build a profile. Data platforms are specialised databases that stores the constantly updating user profiles. Advertisers can pay data platforms for the user profiles in order to increase ad targeting accuracy. Publishers can make money buy selling data to data platforms.