Written by Garrett Carey
When it comes to business intelligence strategy, companies can choose to either buy or build a software solution, and the decision is a complex one. There are numerous build or buy criteria that an organization must consider: ROI, total cost of ownership, time to deploy, resources available, skill sets, and most importantly, how do the tools chosen integrate into the current technology environment.
There are industry estimates that a custom BI solution can extend the time-to-value by as much as 60% to 80%. There are many reasons why this is the case but the main reason for that extension is the basic fact that when implementing a prebuilt application, 60%-80% of the work has been completed before the project even begins.
The primary criteria that an organization usually considers when opting for a prebuilt versus a custom BI solution is implementation time and quicker return on investment. Some of the other reasons are to leverage the expertise of the vendor used for other applications within the organization, integration with other systems, application scalability, and higher adoption rates by the users.
Another equally compelling reason to buy is the availability of upgrade paths, especially when purchasing from one of the BI vendors. For example, when a CRM system is upgraded, the BI vendor will provide clients with the new adaptors as part of the support agreement.
All BI solutions, whether custom or prebuilt, will all contain some of the same moving parts; however, prebuilt applications usually go beyond that standard data model and will include delivered ETL code and BI tool metadata, along with prebuilt reporting content.
The above figure shows an overview of the major pieces of a BI solution. Looking at the individual sections, in particular the ETL section, one can quickly begin to see some of the complexities involved in developing a data warehouse and BI solution.
In addition to rapid deployment and the lower cost of implementation, there are other supporting arguments to buying a pre-built BI application, including:
- Enterprise Data Model using best practices and industry standards.
- Views for the Business Model metadata.
- Pre-built ETL framework containing mappings/routines to load and refresh the application.
- ETL routines have built-in Change Data Capture logic.
- Open Interfaces to other data sources.
- Industry standard Metadata for Common Business Functions (i.e. AP, AR, GL, Sales Forecasting, Pipeline, etc.).
- Faster time to value.
- Smaller deployment team.
- New features available via upgrades.
- Savings on requirements gathering and analysis.
- BI Tool configuration.
- Pre-built dashboard and report content.
- Pre-built alerting capability.
Conversely, there are also substantial risks to building a complete custom solution, such as:
- Increased time to value.
- Requirements change and are not easily implemented.
- Developmental efforts may lack reusability and not meet changing requirements.
- Lack of native integration with CRM and ERP systems
- 100 % custom ETL’s.
- 100% development of reports and dashboards.
- 100% development of metadata.
- 100% configuration of the BI tool.
- Increased cost and overall time related to custom build.
- Smaller ROI
eVerge Group is an Oracle Platinum Partner with extensive business intelligence strategy experience. Contact us to let eVerge be your Oracle implementation guide.
About the author: Garrett Carey is a Senior BI/EPM Solution Architect who has worked with eVerge Group since 2005.