How Does The Mobile RTB Auction Process Work?

By Diksha Sahni | April 13th, 2016

Even as Real Time Bidding (RTB) continues its exponential growth, there is still curiosity among mobile advertisers to fully understand what programmatic advertising really is. One of the most misunderstood concepts in programmatic is the RTB process, which is often confused to being synonymous with what programmatic stands for.

For the uninitiated, RTB is a type of programmatic buying where the buying and selling of online ad impressions is done in real time for the transacting of one single impression at a time. This happens in milliseconds, on average around 200 milliseconds to be precise, before a page loads.

Now, let us deep dive into how the auction process works and the different price mechanisms that work within RTB.


  • DEMAND SIDE PLATFORM (DSP): DSPs are technology platforms that advertisers use to automate the buying of ad space and monitor their campaigns.
  • SUPPLY SIDE PLATFORM (SSP)/PUBLISHERS: Publishers provide the inventory to serve ads. In a programmatic setting, they go through SSPs to help them sell and better manage their inventory. SSPs are DSPs' counterparts on the supply side.
  • AD EXCHANGES: These can be understood as stock exchanges, in the sense that they facilitate the buying and selling of inventory through real-time auctions. Some of the biggest and most popular ad exchanges include DoubleClick by Google, among many others.


The life of an RTB ad impression is approximately under 200 milliseconds, that is from the start when a bid request is placed to finally when an ad is served. Now compare that with the average time it takes for an human eye to blink (300 milliseconds) and you will realize all that complex mechanism is happening even before a blink of an eye!

Let us start with a user X who visits a publisher’s mobile web page or app. As soon as X visits this web page or app, a bid request is triggered, and unique information about the user such as the device type, IP address, operating system etc. is made available when the latter accesses a publisher’s app or web page. Publishers make their impressions available through an ad exchange, which then pings all the participating ad vendors for bids on the ad impressions that are about to be served. The advertisers use DSPs to acquire information about the visiting user and to look at the bid request. First- and third-party data help advertisers evaluate whether the user is part of their target audience, based on which the DSPs bid to buy ad impressions in real time. Depending on the information about the user, the advertisers can manage their campaigns and decide the value of bid -- they may choose to set a higher bid value if they think that user is highly relevant to their ad. Once a bid value is determined, the DSPs look into all the campaigns the advertisers are running to see which one best matches the user. The DSPs, on behalf of the bidders, place sealed bids against other bidders, and the highest bidder wins the auction, getting to place their ad on the publisher’s mobile web page or app.

This process starts each time a user opens a new mobile page or an app.


Unlike the traditional model of media buying, which works on bulk buying ad impressions, in RTB, an auction defines the inventory prices for every single impression, in real time.

Typically, there are two types of price mechanisms -- the first-price auction and the second-price auction. However, Open RTB auctions generally work on the second-price model.

In a first-price auction, the winner is the bidder offering the highest price and auction is cleared based on the highest price offered. For instance, advertiser A bids $1, advertiser B bids $1.50 and advertiser C bids $2.00. The highest bid not only determines the winner (in this case advertiser C), but also the price at which the auction is cleared - $2.00. While a first price auction mechanism gives publishers the highest CPMs for impressions won, in an open auction environment it can create a price inflation, forcing bidders to guesstimate if the competition will bid and what they will bid. Hence, it is suitable for a low demand environment, more commonly used for a private auction.

In case of second price auctions, the winner is the bidder offering the highest price, but the auction is cleared on the basis of the highest bidder paying fractionally more than the second-highest price offered, the exact value of which depends on the platform. To explain this, again let us take the example of three advertisers A, B and C and their bids. Just as above, advertiser C is the winning bidder, but the auction will be cleared at the second highest bidder price plus a small fee, generally one cent - $1.50+fee.

In an Open RTB system, which is a highly competitive environment, a second price auction model lets buyers bid aggressively and honestly for what they are willing to give for a particular inventory, knowing they won’t overpay. It also equips them with a comfort to bid with confidence without the fear of overshooting the mark or paying double to what the competition is offering. By only having to pay a few cents over the closest competitor's bid, it helps maintain a pricing efficiency, transparency, and an efficient marketplace for yield optimization.

However, when the gap between a bid and the second bid is significant, it may create a gap between potential revenues and actual revenues for a publisher. Often, to create an equilibrium, publishers may set floor pricing, known as hard floor price and soft floor price. A hard floor price is the minimum price that the publisher will accept, whereas soft floor price let’s publishers have greater flexibility for their inventory. Bids above the soft floor price are treated as second price auction, and those below it are treated as first price auction. The advertisers or DSPs normally do not know what the floor price is, unless they directly contact the SSPs or the publishers. It is critical for publishers to be aware of what the value of their inventory is and setting floor prices as either high or low can ultimately have an impact on the revenue.


For advertisers, RTB is a great way to optimize their campaigns and realizing its potential on mobile can help in maximizing ROIs, while allowing you to monitor as well as track the performance of your campaign. RTB is becoming a clear winner for mobile app advertisers and is a future to look forward to. When done right, mobile RTB campaigns can help advertisers not only improve their campaign performance, but also help expand the reach of their campaigns to the right audience, eliminating waste and inefficiencies.

Diksha Sahni
Diksha is an Associate Content Marketing Manager at AppLift and is based out of our Bangalore office. When she is not behind her computer writing, you can find her binge watching Hindi movies from the 90s, finding her happy place at a dance studio, and checking off places on her bucket list.

12063761 1030943930291273 1184227129633473729 n