Business Intelligence Implementation: Plan, Software, Costs, and Required Skills
ScienceSoft has been providing a full range of BI services, including consulting, implementation, support, and BI as a service since 2005.
Business intelligence (BI) implementation involves designing, developing, deploying a BI solution to integrate, process, and analyze historical and real-time data. The aim of BI implementation is to support decision-making at strategic, tactical and operational levels.
Sample project scope
Business domain: Market analysis, financial analytics.
Data sources: 2 big data sources (data in the data sources has clear mapping rules, so no data cleaning is needed).
Storage layer: Data warehouse (cloud deployment).
Analytics layer: an OLAP cube.
Data output layer: Power BI reports and dashboards (dashboard levels: CEO, department, personal).
Data security model: Role-based security model.
Analyzing the existing business needs and analytics environment; eliciting the requirements for the future BI solution.
Defining functional and non-functional requirements for a BI solution. These requirements can be further classified into two groups – mandatory and optional.
Note: Interview all stakeholders to collect their needs, goals, and vision of the successful BI project implementation.
Conceptualization and platform selection
Defining the desired BI solution features, technology stack and skills required to fulfill the project. Mapping the proposed solution to the requirements. Defining data sources and ETL procedures, data quality assurance processes, BI implementation and user adoption strategies. As a result, the project team draws up the solution architecture with a detailed feature list.
Note: The processes of conceptualization and platform selection are built on the established iterative communication among a BI analyst, a technical team and business stakeholders.
Defining deliverables, assessing risks, estimating BI implementation costs, TCO and ROI. After fulfilling this step, you get a detailed project plan, a project schedule and a communication plan.
Note: Effective BI implementation project planning can help you reduce project time and budget by up to 30%. To achieve that, carefully elaborate on the findings of the preceding stages.
Delivering the back end and the front end of the BI solution, implementing ETL processes for each of the data sources, setting up data quality management and data security. Quality assurance procedures are run to avoid such problems as wrongly calculated KPIs, slow BI solution response or low-quality UX.
Note: To deliver a BI solution in the shortest time possible, opt for DevOps-driven iterative development to assure the quickness and frequency of releases without sacrificing the solution’s quality.
~ 2-4 weeks
Providing end users with user manuals and training sessions, adjusting common workflows for each user group, etc.
Note: It’s recommended to schedule manuals creation after the development step is finished to have valid user screens.
Pre-launch user acceptance testing to check the BI solution in real-world scenarios. Then, deploying the solution in production, ready for end users to employ.
Note: Depending on the number of users, you may use a phased approach to solution launch, where you release reports and dashboards to different user groups one by one over time.
Solution support and evolution
Throughout the entire BI solution lifetime
Further, in the course of BI evolution, the team can upgrade the solution with self-service capabilities, advanced business analytics and data science capabilities, etc.
– manages scoping, planning, executing, and monitoring all aspects of the BI implementation project.
– interprets business needs, spells out the BI solution requirements, BI solution functionality, user roles, content and modules, integrations of the future BI solution.
– models BI infrastructure components (DWH, ETL processes, BI reports, etc.) and their integration points, ensures design feasibility.
Business intelligence developer
– develops a data model, sets up ETL processes, implements and maintains data modeling tools, query tools, data visualization and dashboadring tools, ad hoc reporting tools, etc.
– transforms data into a format suitable for analysis by analyzing raw data, developing and maintaining datasets, improving data quality and efficiency.
Quality assurance engineer
– validates the BI solution: designs and implements a test strategy, a test plan and test cases for BI components, validates SQL queries related to test cases and produces test summary reports.
– sets up the BI software development infrastructure, automates and streamlines development and release processes by introducing CI/CD pipelines, monitors system security, performance, and availability, etc.
The sample BI implementation project involving building a DWH, an OLAP cube, custom-built reports and dashboards would require the following team composition:
The company has full control over the BI implementation project.
Caution: The implementation project can be delayed or compromised due to the lack of the required resources.
Technical activities are partially outsourced
Hiring a vendor to outsource the design, implementation or support of the BI solution. The model presupposes your high control over the implementation project.
Caution: High requirement for in-house competencies.
Technical activities are fully outsourced
Among the major pros – no risk of the resource overprovisioning after the project completion.
Caution: High requirements for in-house PM and BA competencies.
Everything is outsourced, except a project sponsor
This model promises no idle costs and no delays due to resource unavailability.
Caution: Increased vendor risks due to increased vendor dependency.
Best for: big data warehousing
Leader in Gartner’s Magic Quadrant for Data Management Solutions for Analytics, this DWH stores and processes business data of (extra-) large volumes.
- Ingestion of all data types.
- Integrations with AWS services (including S3) and third-party tools.
- Federated queries support.
- ML capabilities.
- Separate scaling of compute and storage, etc.
- On-demand pricing: $0.25/hour (dc2.large) - $13.04/hour (ra3.16xlarge).
- Reserved instance pricing offers saving up to 75% over the on-demand option (in a 3-year term).
- Data storage (RA3 node types): $0.024/GB/month.
Azure Synapse Analytics
Best for: enterprise DWH
Leader in Gartner’s Data Management Solutions for Analytics, this DWH unifies enterprise data warehousing and big data analysis.
- Ingestion of all data types.
- Multilanguage support.
- Embedded Spark engine.
- Native integrations with an Azure data lake and operational database.
- Built-in integrations with Microsoft’s BI and ML software and third-party solutions.
- Advanced security, etc.
- Compute on-demand pricing: $1.20/hour (DW100c) - $360/hour (DW30000c).
- Compute reserved instance pricing allows saving up to 65% over the on-demand option (in a 3-year term).
- Data storage: $122.88/TB/month.
Best for: data visualization
BI and data visualization software.
The leading Analytics and BI platform (Gartner’s 2020 Magic Quadrant) and enterprise BI platform (the Forrester’s Wave).
- Native connectors with 100+ data sources.
- Multilanguage support.
- Integrations with a data lake and operational databases.
- Self-service data preparation and analysis.
- Scheduled and ad hoc reporting.
- Interactive dashboarding.
- Pre-built and customizable visuals.
- Row- and workspace-level security.
- Data encryption.
DEMO: Watch our Power BI demo.
- Power BI Desktop – free.
- Power BI Pro – $9.99/user/month.
- Power BI Premium –$4,995/dedicated storage and compute resources/month.
The approximate cost of a BI implementation project, which involves developing a data warehouse, an analytical cube, and Power BI dashboards and reports to consolidate and analyze data from 2 complex data sources is $400,000 (monthly software license fee or other regular fees are NOT included).
The cost of a BI implementation project may vary significantly from project to project. We present the major cost factors:
BI implementation consulting
- BI implementation feasibility study.
- BI solution concept.
- BI needs business analysis.
- BI solution launch strategy.
- Optimal BI implementation sourcing model.
- BI software selection.
We help you implement a full-fledged solution in accordance with the defined strategy by:
- Analyzing your BI needs.
- Developing components of a BI solution (a data lake, DWH, OLAP cubes, reports and dashboards).
- Setting up ETL processes.
- Designing reports and dashboards.
- Adding data science capabilities, if needed.
- Conducting management activities (master data/metadata management, data quality assurance, data security, etc.).
- Running quality assurance procedures, etc
ScienceSoft is a global IT consulting and IT service company headquartered in McKinney, TX, US. Since 2005, we’ve been providing our customers with BI services, including BI consulting, implementation, supporting and upgrading a BI solution, or getting regular data analysis and reporting on a subscription-fee basis.