Amidst a myriad of visualization tools available off the shelf, it is important that BI folks make a right choice in the selection of the BI tools for better utility and results. This article makes an attempt to explain how one should select the right tool depending upon the requirements and not the tool’s sales pitch.
Listing of requirements and expectations from Visualisation tool?
It is important to first understand the method of hosting the data i.e. if it is hosted on the premises or on a cloud. It is also important to quantify the volume of data because it will help in identifying as to what extent visualization can be achieved. Visualisations are on top of a deep structured system and therefore a company’s readiness for visualization must be identified beforehand.
Check for some of the basic features like Export to PDF, Multi browser support, Device responsiveness, etc.
Keep aside a budget on a monthly or on a yearly basis. It is a good practice to first find out the maximum limit of money that a company can shell out. A tool can then be found out in that budget.
Identify if the user would be personal, departmental or enterprise wide.
In the case of a multi-user application, list down all the functionalities essential for smooth operations and co-ordination and check if the tool serves the purpose correctly.
Any reservations on technology like open source or proprietary should also be listed.
Ensure that the tool supports multiple Operating Systems or is at least compatible with the existing OS on the company’s systems.
Find out the inbuilt features of the visualization tool.
What are important features from any Visualization Tool?
- Wide Database Connectors Support
- Live Data Connection for Real Time Analytics
- Multi Device support to view Dashboards
- Security Features
Since visualization helps in understanding the data and facilitates decision making, the tool should have strong security features:
a. Application Level Security – Application level security lets one control the BI application access on a per-user role or per-user basis. This means the user profile defines the levels of accessibility.
b. Data Level Security – It is a critical feature of BI and reporting applications. The multi-tenant security lets one control data access within the application at the row level.
c. Active Directory Integration – A session/user authentication process, single sign-on (SSO) lets users enter their name and password in only one place and access multiple related applications. It authenticates the user for all authorized applications and eliminates login prompts when switching between applications in a single session.
d. User Privilege Parameters – User privilege parameters let you personalize features and security to individual users or user roles. Saved to a user’s profile, these user privilege parameters control user-specific features throughout every BI application.
- Reports Type
Visualization tool should help in creating different types of reports as follows:
a. Ad-hoc Report
b. Executive Dashboards
c. Interactive reports
d. What-if analysis
e. Geospatial/mapping applications
- Advanced Features
a. Intelligent Alerts – Intelligent alerts let a BI application automatically send an email or SMS message to the appropriate party when data reaches a pre-defined threshold. For instance, intelligent alerts can instantly notify the CEO any time a customer cancels their account, or any time sales numbers reach abnormal levels.
b. Collaboration – As we learned from the rise of social media, the internet providers the perfect collaboration platform. This concept translates seamlessly to Business Intelligence. In the future, BI application will incorporate commenting and let you interact with other co-workers directly within the application.
c. Cloud-ready BI – Most analysts and researchers agree on one point: Cloud computing is the future. BI on the cloud promises near 100% uptime and scalability while avoiding the effort/expense of in-house hardware. As illustrated below, advanced BI software creates cloud-ready apps that deploy anywhere—on premise or in the cloud.
d. Built-in ETL – ETL tools let you extract data from multiple source systems, transform it into a single format, and load that data into a target database. They give end users a simple way to include data from multiple locations in a single report. However, many BI solutions don’t offer ETL capabilities as a built-in feature, but as a costly add-on.
- Chart Gallery
Charts are a graphic and pointed representation of data, which can then be converted into decision impacting information. The different types of charts are Bar Chart, Line Chart, Area Chart, Pie/Doughnut chart, Funnel Chart, Gauge Chart, Radar Chart, Gantt chart, Maps, Matrix/Tables, Scatter/Bubble chart, Tree map, Waterfall chart and Box Plot Chart.
- Integration with R and Python
BI is not only about past and present, but also about the future. So integration with tools like R and Python is necessary as it enables Forecasting, Segmentation and Decision Tree like functionalities.
- Developer Community for rich collection of Use Cases
This is crowd sourcing the intelligence in the business.
I think this extensive list of checks will take significant time and efforts, but it is indeed worth as Visualization Tools helps us seeing the data patterns & health of Organization. Also it involves cost for buying licenses and thereby use stakeholders.