How Embedding Works

Scalable Embedded Analytics with Pyramid

Contextual Analytics

To make business analytics more accessible and relevant, organizations have been striving heavily to put analytics in the context of their business applications and workflows rather than treat it as a separate set of tools and functions running in its own sandbox. The process of embedding analytics is now a top-tier demand, and it comes with its own set of problems and complexities. Pyramid’s embedded capabilities (separately licensed) is designed to meet those challenges—making scalable, high-end 3rd party analytic solutions a reality - without the complexities assocaited with development.

Injection vs IFrames

A key element of Pyramid's approach to embedding, is that all content is injected into the host page . This is in strong contrast to most other solutions that ultimately host the external analytic content in one or more iframes sitting on the page. While the difference may sound academic, there are significant downsides to iframe-based embedding:

  • The client engine hosting the content has to be instantiated multiple times with iframes (once per item). This ie EXTREMELY heavy and can debilitate any site and the user's browser - the memory requirements are greater, since the entire client engine and data model needs to be recreated per embedded element.
  • Building cross interactivity between elements, through custom HTML or JavaScript, is significantly harder to achieve with iframes.
  • Iframes are notoriously bad or difficult to use in single page apps built with frameworks like React or Angular - which don't expect to handle external, sandboxed content as part of their standard flow and operation.

Embedded Generative BI

Pyramid's embedded content inherently embeds all the AI functionality found in Pyramid's standard client experience. This includes access to Smart Insights and the power of the Gen-AI driven Chatbot. Through the Chatbot, users can even dictate questions or instructions, via speech to text, to orchestrate new analyses or effect adjustments to existng content. This works seamlessly with all embedded content, regardless of data source - without requiring further development or effort on behalf of the hosting application.

Choose your embedding approach

Pyramid supports two distinct types of embedding operation: ANONYMOUS and NAMED. They each have their distinct pros and cons and customers can pick the model that suits their use case best. Both types of embedding require additional licensing options in Pyramid.

Anonymous Embedding

Anonymous Embedding is designed for deployments where each user consuming the content is technically 'anonymous', operating without a distinct account on the Pyramid platform. This is a convenient mechanism when content is widely distributed to thousands of users who do not need personalized content or features. Importantly, there are mechanics to still secure the data for each anonymous user, but it needs to be programmatically injected into the authentication process of Pyramid.

Anonymous Embedding is licensed as a platform wide capability, and is not tied to users or seats. However, once enabled, it can be used by named users alongside anonymous users.

Named embedding

Named Embedding is designed for deployments where each user consuming the content is named in the system, with their own licensed seat on the Pyramid platform. By definition, the content and data is secured by user/role in named embedding scenarios unlike Anonymous Embedding deployments. Therefore, it is most relevant when users have personalized content, personalized data access or need features that require specific user tagging (like the Embed Hub).

Named Embedding is licensed as a platform wide capability AND by user seat. A special BASIC user type is also available to access embedded content under the named model (as well as any other user seat type: Viewer, Analyst, Pro). Users without a named seat CANNOT access embedded content under this model.