Understanding how programmatic advertisements work can be challenging if you are not very familiar with the ad tech ecosystem. I keep getting questions on how programmatic ads work in real life, who the players in the space are, and why every player is important. I thought I would create something very simple to make this complex topic easy to comprehend.
In this section, I’ll add some more explanation to every piece to make it simple. These are the important players in the Programmatic AdTech ecosystem:
Publishers are the sites or apps that generate content like news or sports. They attract a specific type of audience. They sell the ad spaces to the advertisers who want to reach that audience. They are integrated with various supply-side platforms that help them with populating those ads.
Supply-side platforms integrate with the publishers and send that inventory to the ad exchanges. These ad exchanges are integrated with multiple demand-side platforms and send them those requests. They conduct an auction by receiving the bids from the various demand-side platforms.
Some well known supply-side platforms are Magnite (previously Rubicon Project + Telaria), PubMatic, Index Exchange, Xandr, Smaato, OpenX, MoPub.
Demand-side platforms are where all the agencies go and run the campaigns on behalf of the advertisers. Demand-side platforms are also integrated with multiple ad exchanges and receive traffic from them. They have what they call a ‘Bidding Engine’ which helps match the inventory and the campaigns and places a bid on behalf of the agencies. The bidding engine is one of the most complex parts on the demand-side platforms.
Some well known demand-side platforms are The Trade Desk, DataXu (acquired by Roku, rebranded as OneView), MediaMath.
Agencies run the campaigns on behalf of the advertisers on the demand side platforms.
Diving a little bit deeper, this is how an auction process works in the programmatic AdTech ecosystem.
When a user opens a publisher website, along with the actual content load, a request is made to the SSP’s ad server to fill the ad space on their website.
The SSP forwards that request to the exchange (within the SSP) which is responsible to conduct an auction. The SSP might have various integrations with some third-party services that help augment data coming from the publisher.
The exchange will send the same request to multiple DSP’s (via OpenRTB spec) with all possible information it can gather on the user.
The DSP’s look at the incoming request and check if any of the campaign setup within their platform is wanting to target that particular user. If yes, the bid is sent back at a price the advertiser is willing to pay for that space on the publisher’s website. If the DSP doesn’t want to bid, that request is dropped and a no-bid is sent.
The highest bidder within the timeframe is deemed as a winner and their ad (sent in the bid response) is sent to the publisher’s website and shown to the user.
The SSP sends the bid request to the DSP. This bid request consists of a lot of information that is required by the DSP to make a decision whether to bid on that request or not.
This bid request is evaluated against the campaigns that the agency has set up within the DSP. If all the parameters of the bid request match what the campaign wants to target, the DSP would send a bid for that particular campaign along with the bid price.
The SSP gets multiple bids for the same request and the winning bid gets a spot on the user’s device.
SSP is responsible to send the notification back to the winning DSP to indicate that it has won the auction along with the winning price.
Win price can be less than or equal to the bid price.
DSP has servers that capture the win notification and sends that information to the Pacing server, which is responsible to ensure that the campaigns are evenly delivered across the dates of the campaign. It syncs with the bidders to notify them to stop bidding for the campaign if it’s already delivered in full.
All these agents send the data to data and analytics which in turn enriches the platform and helps make the agencies better target the end-users with their campaigns.
Disclosure: I work for AdTheorent (Nasdaq:ADTH) and the views in this post are solely my own and are for educational purposes only. None of this should be construed upon as financial advice.