It’s impossible to imagine leading mobile and enterprise apps without garnering important insights from embedded analytics at least on some level. Integrating reporting, analytic and BI capabilities into existing business applications can really help enterprises achieve business objectives in record times. However, embedding has to be implemented correctly and at the right depth for your company to reap most benefits. Today we look at different 4 integration levels, based on the depth of analytics embedding (from Stage 0 of embedding to the deepest integration level):
Standalone app separate from the process, the core of traditional BI.
Integration Type: there are two apps: the main data generating app and the analytic app
Access Type: via data export
User Experience: users work with two apps that look and operate differently
Drawbacks: even though data becomes outdated no matter how you embed, in this case it has to be updated manually, putting a strain on resources; no syncing
Example: exporting data from the app for analysis in Excel, and creating a new copy of data along the way.
When to use: The right answer would be never. The only exception is lack of opportunity to embed (for various reasons). When you are just at the beginning of a product lifecycle, it does happen that analytics are part of a separate app. Still, if that’s the only way you are going to utilize analytics, it’s best to do it this way than not do it at all. But you have to aim at integrating analytics deeper into the core app as soon as your situation allows for that.
Itransition’s customer from Australia aimed to launch a nationwide online pay-as-you-use recruitment platform providing full cycle process support in order to cut recruiting costs by up to 80%. Itransition designed and developed a scalable and secure media content processing system supporting metadata management and media distribution mechanism involving video streaming solutions.
Since the platform supported BI-based mapping of recruiters and candidates, the analytics was initially planned to be embedded in the app but because the client was a startup, it was decided to use manual embedding on the MVP stage. It was a costly tedious process, so as soon as the client scaled up, scheduled analytics features were embedded into the app, increasing the effectiveness of the app hundreds of times.
Single sign-on integration. Security is integrated with the core app.
User Experience: more seamless than Stage 0 as users need only one set of login credentials, but it’s still a separate experience
When to use: Use this method when your solution contains multiple apps but a single analytics app to access data from one or more of core apps or when the analytics app is in the cloud extracting data from on-premise or cloud apps. Another example is the situation where the analytic app is offered to be purchased separately from the core app or if you are using the embedding method as an intermediate step in development, to be integrated later.
Tips: Remember about the roles and rights of users and tailor authentication features accordingly while also making sure users only see information they are allowed to see. Ensuring synchronization with user profiles and the core app and enhancing security in multi-tenancy environments will help you maximize the potential of this embedding stage.
One of the leading US brand building consulting agencies needed BI software to help their clients monitor, collect, analyze and influence brands online.
Project objectives included minimizing time loss from transferring and merging stats accrued by different apps, cutting efforts spent on locating and accessing distributed data, boosting analytics efficiency with rich visualizations. To achieve these goals Itransition developed a secure BI portal with multiple modules, realized independently from each other with an independent analytics module, but united by a single SSO for users to maintain the illusion of working within one app.
The BI portal delivered by Itransition consists of 4 major solutions providing an end-to-end data processing capability: a central BI storage with processing kernel/a service-oriented integration point; app integration & ETL solution to attach to the data collected by third-party software and directly from other available data sources to the processing kernel; report designer to allow analysts to write custom queries and a presentation layer for easy visualization.
As a result, the analytical kernel offers a 360-degree brand behavior view to configure and customize reports and match the monitored brand specifics. The software solution became a core business-critical instrument for the customer, increasing operational efficiency and providing clients with value-adding services, helping to identify weak points in business processes and strengthen those areas by making data-driven investment decisions.
Analytics included in the core app. This is the most common analytics embedding model.
Integration Type: Functionality of the analytics part appears inside the UI of the core app in the form of a report module or tab or as a dashboard visible to users on the landing page immediately on log-in.
User Experience: the look and feel of analytics components match the main app
When to use: When analytics has to be accessed often and in the easiest way possible and when you need convenient reporting modules
Tips: Make the embedded app customizable and easy to complement your products for a cohesive user experience. It’s best to choose API that is stress-free not only to implement but also to maintain, update and service.
A UK-based provider of reporting, budgeting and reconciliation software needed an upgrade to its ERP reporting/query app for business users. Itransition’s team added vital analytics features extending system functionality and introducing new supported platforms. The results exceeded everyone’s expectations, with more than 50,000 corporate end-users and an app outperforming market competitors.
Itransition’s team enhanced existing functionality by developing an advanced commenting engine and creating a report variables framework to define a set of report variables of different types
The Pixel perfect reporting feature included: rich text formatting in report titles; advanced reusable formatting in report headers; custom style sheets defined when exporting reports to Word with users enabled to customize field layout or add objects to the exported report (text, company’s logo, additional tables, charts, etc).
Analytics are part of the main app. This type should be your ultimate goal and end destination of embedding.
User Experience: analytics is the core of the experience, embedded within major workflows
Depth of embedding: analytics are provided in the moment or in existing app screens or by interacting with analytics in order to make analysis-based actions and transactions
Advantage: you can bring the most value to your organization, realizing strategic goals and objectives
When to use: If you are offering personalized experiences and support or when analytics features are the core capability where insight and action are united in the same context.
Tips: The embedded analytics app and the main app should be contextualized accordingly. When transacting and updating data from the analytics part of the app, it should be implemented with the help of calling a backend API of the core app while enforcing business rules or via direct data call-backs to the back database.
The Customer is one of the leading oil&gas enterprises that required an update to the existing reporting system.
Performance data on key business dimensions was scattered across documents and reports created by different employees and departments, with information shared via email monthly slowing down decision making. To increase data analysis efficiency, a centralized solution for corporate data storage, processing, visualization and transfer became a necessity.
Itransition’s team developed the solution with the following objectives in mind:
As a result, 10 toolbars containing KPIs presented as diagrams were created, united in a single user-friendly interface that allows working with toolbars in English and Russian while filtering data that certain department analysts process. For in-depth analytics, users can quickly navigate to a detailed report from any diagram point.
What types of embedded analytics have you tried in your company? Please share your thoughts below and comment if you have any special tips and tricks on the topic.