Skip to main content Skip to footer

An AdTech Stack for a Leading OTT Video-Streaming Publisher

A marketing-leading video-streaming service partnered with Qinshift to design and develop its proprietary AdTech stack.

About our client

Our client operates one of the largest video-streaming services in Asia.

Their streaming platform offers over 100,000 hours of video content, has 40+ million monthly active users (MAU), and serves over 190 countries. 

The challenge

The main challenge our client faced was that they had access to millions of active users but didn’t have a scalable system in place that would allow them to monetize this audience and create a new revenue stream.

Our client’s goal was to develop their own AdTech stack to build a walled garden and monetize their first-party data. The AdTech stack would be used to collect data, create audiences, and run advertising campaigns that showed targeted ads to those audiences.

We were responsible for the design and development of the essential components within the AdTech stack, including a self-serve ad platform, ad server, data lake, and customer data platform (CDP). 

Our client partnered with Qinshift as they didn’t have the required skills, knowledge and experience internally to design, develop and launch their AdTech stack.

The solution

We designed, developed and launched the various components of our client’s AdTech stack.

The above image illustrates our client’s proposed AdTech setup.

Although we designed and built the ad server, self-serve ad platform, and CDP, we haven’t yet integrated them together as this plan is scheduled for the future.

Self-serve ad platform

The self-serve ad platform allows advertisers, advertising agencies, and our client's internal AdOps team to effortlessly create, modify, and oversee their ad campaigns from a centralized user interface.

We established an API integration to facilitate seamless communication and data exchange between our self-serve ad platform and the external AdTech platform where the actual ad campaigns are carried out.

What We Did

  • Designed and built the self-serve ad platform’s UI.

  • Developed the necessary features for generating and overseeing advertising campaigns, including the ability to upload creative content, establish line items, and configure ad targeting. 

  • Integrated with an external AdTech platform to execute and update the campaigns created in the self-serve ad platform. 

  • Integrated with a payment system to deliver payment and billing-management functionalities. 

Ad server

The goal behind building an ad server was to replace our client’s current solution, Google Ad Manager, and enable them to deliver video and image ads seamlessly within their video-streaming service.

To begin the project, we focused on creating the minimum viable product (MVP) of the ad server. This MVP would later be integrated with our client's other AdTech platforms, such as the self-serve ad platform, data lake, and CDP. 

By doing so, we aimed to validate the company's business requirements and strategic objectives.

What We Did

Below are some details of the main components and areas of the ad server.

Created the Technical Requirements

We received general information about the ad server and how it should work, but we didn’t receive any specific technical requirements for us to follow. Our team thoroughly examined the entire project and developed a comprehensive set of technical requirements for the MVP using our expertise, knowledge, and extensive research.

Developed a Web SDK

We collaborated with another team from Qinshift to develop a web SDK that functions as an intermediary component connecting the ad slots with the ad server. This SDK enables the display of ads while also facilitating the transmission of user IDs to the ad server.

Used VAST 4.2

We utilized  Video Ad Serving Template (VAST) 4.2 to assist with the delivery of video ads between our client’s video player and the ad server.

Matched Audiences from the CDP

In order to facilitate targeted advertising, we developed a method to determine if a visitor was a member of a specific audience. In order to accomplish this, we obtained audiences from our client's customer data platform (CDP), which we also developed, using the SDK. We employed bloom filters to match visitors with the predetermined audiences.  

Data lake and CDP

Our objective was to design and build a data lake, which would collect data from different sources, and a customer data platform, which would allow users to create and modify audiences. The CDP would then integrate with the ad server and self-serve ad platform to enable advertisers to execute audience-targeted campaigns on our client's OTT video-streaming platform.

What we did

The data lake was built to retrieve data from multiple sources, including:

  • Our client’s OTT video-streaming platform: Information about movies, actors, etc. from our client’s content management system (CMS).

  • Our client’s analytics platform (Mixpanel): Event data about video views, watch time, sessions, as well as location, device type, and other types of data. 

  • External data sources (Axinom and Lotame): Interest-based audience segments created by third-party data.

  • Data from our client’s data science team: Data created by machine learning and data analysis. 

  • Device information: Data about the price of certain devices. This data was then connected with the device-level data collected by Mixpanel. 

After gathering the data, the data lake would proceed to arrange and refine it, such as eliminating duplicates and modifying the data schema.

Then, the CDP would obtain information from the data lake, generate audiences based on the data, and transfer them to our client's ad server for targeted advertising on their OTT video-streaming platform.

During our MVP Scoping phase, we created user story maps, learned about our client’s tools, and provided cost and time estimates. 

Subsequently, we started working on the MVP of the data lake and CDP, with this phase lasting 7 months.

The benefits of working with Qinshift

By working with Qinshift, our client was able to launch their AdTech stack and monetize their first-party data by allowing brands to run targeted advertising campaigns on their video-streaming platform.

The keys to the project’s success were:

  1. Qinshift’s experience, skills, and knowledge of designing and building custom advertising technology.

  2. Our incremental and agile approach to software development. 

  3. Our cooperation with our client and their internal teams.