What are common use cases for analytical applications?

Analytical applications tend to fill gaps where either data within an operational application required for a decision is lacking, the operational application does not have the required workflows to take appropriate action, or the data within the application needs additional processing to enable effective decision making.

These applications tend to support highly functional decision making or specific operational workflows, and as such have a lot of domain expertise baked into them. They tend not to be generic business intelligence type applications.

If data is required from multiple sources systems to enable action then analytical apps can help. For example sales forecasting may need data from Salesforce as well as your finance system. By merging the two data sets together you can provide a complete picture to the end user.

When the primary application does not provide the data in a format that makes decision making easy, or does not have the internal workflows that allow end users to take action. In this case you will extract the data from the source system and build operational workflows into your application. Using the salesforce example, although you may be able to build a dashboard – the dashboard does not facilitate the ability to action the opportunity directly from the dashboard. The user must first navigate to the opportunity to do so. This creates friction in the operational workflow that can easily be addressed by building actions directly into a dashboard.

Sometimes the data in an application needs additional processing – for example you want to run a data science model over the data to predict an outcome. In this case delivering the outcome back to end users can be difficult via the originating application. Once again you can solve this problem via an app. So if you want to predict which customers are likely to buy an upgrade – you can extract your salesforce data, run your model. Deliver the results via a dashboard to the sales organization, who can then take action on the most highly likely buyers and make them an offer directly from an analytical application.

What types of enterprise transactions systems do data driven apps make sense for?

Analytical apps can be applied to many types of transactional systems across multiple industries or functions. Primarily they make sense when combined with deep domain knowledge so that they address gaps in workflows or data needed by the organisation to take action.

What types of Analytic Applications are a good fit for Yellowfin?

The best way to explain this is to use Gartner’s Pace-Layered Model. Based on this model, an organization’s application landscape is made up of three distinct layers, each with a corresponding rate of change. This rate of change is dictated by the uniqueness of the solutions and how concrete the requirements are:

The system of record, at the bottom of the model, is the foundation of the business. This a structurally solid system with a slow rate of change and well-defined requirements. In the middle is the system of differentiation, which focuses on fostering outside-in and customer-centric thinking, accelerating the rate of change, and developing unique approaches to sustain differentiation. And at the top is the system of innovation, which represents brand new ideas for the organization and thus has fuzzy requirements and a high rate of change.

Yellowfin helps most in projects where the requirements are not completely defined up front and in projects that have a high rate of change. In line with Gartner’s Pace-Layered Application Strategy, a system of differentiation or innovation matches these criteria. For a system of record, Yellowfin is not a good fit.