Architecture and Structure


This guide introduces how Connected-Stories fits in the advertising ecosystem, and how to configure it to seamlessly interact with Demand-Side Platforms (DSPs) and Supply-Side Platforms (SSPs). It presents an overview of the Connected-Stories conceptual architecture as well as a few implementation examples. This page goes into depth on:

  1. How Connected-Stories fits in the advertising ecosystem
  2. Integrating Connected-Stories and distributing stories onto publisher's properties
  3. Sequencing Personalized Stories
  4. Optimizing media buying
  5. Integrating real-time signals from external platforms

HOW CONNECTED-STORIES FITS IN THE ADVERTISING ECOSYSTEM

Publisher Implementation Scenario

In the typical distribution scenario, the Connected-Stories cloud interacts with  Publisher Digital Properties and Supply Side Platform:

INTEGRATING CONNECTED-STORIES AND DISTRIBUTING STORIES ON PUBLISHER'S PROPERTIES

In the advertising ecosystem, Connected-Stories acts as third-party ad server so it can be easily integrated in the standard distribution workflow. Connected-Stories uses the most-common standards to reach a high compatibility level with the most-used ad distribution mechanisms. In fact, Connected-Stories uses technology such as HTML5–instead of old SWF files that are no longer compatible especially in the mobile environment–or pure JavaScript tag. Connected-Stories is also committed to remaining compliant with all standards and best practices developed by the Interactive Advertising Bureau (IAB). This means that with Connected-Stories you can distribute standard creative formats such as masthead, leaderboard, half-page, along with video formats such as VAST and VPAID, and mobile-centric formats like MRAID.

Stories are published by trafficking their respective Ad Tags. An Ad Tag is automatically generated in Ad Export tab in the Story Manager section of the Connected-Stories administration panel. The generated code can then be downloaded from the Connected-Stories cloud and then uploaded onto publisher's distribution system (generic ad server, DFP, AppNexus, etc.). A Connected-Stories Ad Unit must appear in an area of the publisher page designed to host the creatives.

Connected-Stories creatives can be hosted in:

  • Standard Placement:
    • Standard iAB display placements (billboard, half-page, etc.) 
    • Video (VAST, VPAID)
  • Native Placement: to match the form and function of the platform upon which the Ad Unit appears, the creatives must be designed with look and feel of the publisher's properties. A tag must be set up in a designated area to simulate native content.
  • Social Media Canvas: Connected-Stories offers free form canvases that can be triggered by sponsored posts published on the major social media platforms. Social networks supported includes:
    • Facebook
    • Instagram
    • Linkedin
    • Twitter
    • Pinterest

To serve a Connected-Stories Ad Unit you can export the Ad Tag as:

  • HTML5: a Zip package that contains an HTML file and minimal assets needed to display the Ad Unit 
  • JavaScript: a JavaScript code that loads the Ad Unit.
  • VAST/VPAID XML: a url that represents the XML used to interpreter the VAST or VPAID standards.

See also:


Ticketing System

SEQUENCING PERSONALIZED STORIES

Connected-Stories has the unique capability to use different advertising placements as if they were a virtual canvas, carefully preserving narratives conceived to be a sequence. The platform makes this possible because it is able to:

  • Recognize the same user across different digital touchpoints, across media and across devices
  • Infer the user's current status within the campaign-specific customer journey.

No customized development work is needed to use these features because the system functionalities are based on deploying cookies (available in each Connected-Stories service plan) and probabilistic building of user device graphs (activated in the highest-tier service plan). The information collected is anonymous and compliant with leading industry standards.

OPTIMIZING MEDIA BUYING

One key advantage of using Connected-Stories to boost the ad serving effectiveness is its ability to bundle placements which implements the overall content experience as designed by creatives, thereby reaching the campaign objectives. The platform is intuitive, meaning it can help advertisers optimize their media spending by selecting the placement sequences more precisely. Enabling this process in every Connected-Stories deployment project is the inclusion of an extension to the ad server tag used by the chosen ad server technology vendor.

The process of optimizing media buying deploys the same Agile methodologies used for developing and improving the core Connected-Stories platform. Agile calls for a rapid, iterative cycle of buying, measuring and applying learning. The Connected-Stories platform collects extensive metrics on placements using the ad server tag extension. The platform then analyzes campaign performance and effectiveness against our knowledge base of aggregated past campaigns as well as objectives set for the current campaign. This leads to interim outcomes from each Agile iteration, or "sprint", that can be immediately applied to buying decisions and parameters. Each sprint builds on previous rounds of decisions to reach an optimal buying profile within a short period of time, that also remains adaptable to changes in user behavior that might vary in the course of a campaign.

Interaction data are collected by each widget used inside an ad unit. Data are keyed to each user by an anonymous identifier in order to construct a picture of the overall customer experience and rate it against the campaign KPIs. We recommend cycles of two weeks in order to work with a sufficient number of data points, at least 10 million impressions per cycle. More detail can be found at /wiki/spaces/CS/pages/2949181.

Optimization actions are executed by interacting with and modifying the decision tree visualizations, accessible from the administration panel, that define the campaign's consumer journeys and story personalization rules.

INTEGRATING REAL-TIME SIGNALS

Connected-Stories' cognitive engine provides several parameters that can be used for adapting the storytelling to the user's context. The system supports information arriving from mobile devices when its SDK is linked with publishers' apps. More signals can be imported from external platforms able to push real time information, i.e. trending posts, breaking news.

To know more about supported signals see the Rules Engine section.

Detail about Real-Time signals

Decision trees forming the cognitive engine of a given campaign can be freely customized for any variable passed to the system by using the cognitive engine API. The design for the number of real-time inputs to process should be limited to avoid creating too-complex structures that impact the number of stories that must be created as well as the ability to measure the cross-media attribution of KPI values.


Related pages


NEXT STEP: Ticketing System >>