Data warehousing is a key technology that many companies are pursuing today. A data warehouse can deliver enormous benefits and key data in a timely manner, however; they are also subject to a high failure rate. The reasons for failure vary but almost always two symptoms will surface, cost and user expectations. Notice I identified the symptoms of a failed DW project not the root causes. For that we need to delve a bit deeper into buy vs. build options around data warehousing.
“Build Your Own” Data Warehouse
Too often the commitments and aspects of building a data warehouse are not recognized by companies. Whether they know it or not, a company that decides to build their own data warehouse becomes a software development company. Once the warehouse is live the company must commit to the same actions a software development company does to maintain compatibility in a fast paced environment of change.
The following are just a few of the events that will trigger modifications and updates to the data warehouse:
- Users demand new functionality and data be added
- The company upgrades or changes ERP software
- The company switches database platforms or vendors
- Business rules evolve and change necessitating changes to ETL processes moving data from production systems and to data marts.
Since most small to medium companies cannot attract and maintain staff sufficiently skilled and all the required areas, the lack of resources eventually cause the data warehouse to implode under the load of unmet requirements.
Two key questions every company, large or small, needs to positively answer is:
- Do we want to become a software vendor?
- Can we attract and retain the key skill sets?
Most companies do not write their own ERP systems. Why then would they consider creating and maintaining their own data warehouse? Simple answer is the lack of products to address this need. Fortunately, that is no longer a road block. Many companies have decided to productize their offerings. As a result, there are robust flexible products that can deliver the majority of needed functionality out of the box.
Among these packages is one aptly named BI360. Completely based on the Microsoft BI stack, using familiar and easy to use Microsoft Office tools for user interaction, solid proven data transfer and storage technologies in the back end, BI360 delivers many commonly used financial reporting aspects out of the box as well. Extensibility is achieved by using meta data tools to map and connect to most major financial systems while being flexible enough to allow total customization for those unique challenges.
Whatever route you choose, you should evaluate out-of-the box data warehouse solutions like the BI360 Data Warehouse before deciding to build your own data warehouse.