There has been a lot of talk about measuring Return on Investment (ROI) in the data warehousing area. The best study that I have seen was published by IDC a number of years back. This study consisted of 62 companies from Europe and North America, running the gamut
- 62% had revenues of $1B or more
- Data warehouse sizes ranged from megabytes to over 1 terabyte
- All data warehouses were in use for more than 6 months
- Usage ranged from 3 to 1300 users
This shows that there was not a stacking of the study (this is always good). If you are a CFO, you will understand this next section really well. There are assumptions used within this study that help understand the calculations of ROI. They are:
- Measurements taken over 3 years
- Discount rate = 15%
- Tax rate = 50%
- Inflation rate = 3.5%
- No debt to equity limit
- Cost of capital = 12%
- S/W and H/W expensed up to $100K
- Straight line depreciation
The range of the results of ROI ran from -1,857% to +16,000%. When you exclude the 8 lower outliers and 9 higher outliers (the range then becomes +3% to +1,838% ROI), you will have:
- Average 3 yr. ROI (AROI) = 401%
- Median 3 yr. ROI (MROI) = 167%
- Average Payback (APAY) = 2.3 years
- Median Payback (MPAY) = 1.67 years
- Average Cost = $2.3M
- 14 companies had an ROI of over 1,000%
Some might view these statistics as the value of a data warehouse. More reflective, I think that it shows the potential of the data warehouse. It also shows something of what I consider the utmost importance: it takes time to build a data warehouse. The average payback is 2.3 years and the median payback is 1.67 years. So when you say that you want to build a data warehouse in 90 days, you might be kidding yourself. And when we want to do it for $50,000, we should reconsider due to the average cost during this study ($2.3 million).
Some important aspects of this study provide us with a good understanding of the issues involving the ROI of a data warehouse. There are significant variances in the ROI that exists between the different companies in this study. Some conclusions that could be raised are 1) organizations get impressive results from their data warehousing efforts, 2) discrete applications (data marts) had a higher ROI than enterprise applications (533% vs. 322% - both which are quite impressive!), 3) ongoing support of users is the key differentiator between pedestrian payback levels and truly impressive results, and 4) it is possible to have a negative ROI. Some reasons for the negative ROI are 1) extraordinarily high costs, 2) low usage of the data warehouse, and 3) having a too large scope which will cause a greater than 3 year payback.
So, what can we learn about ROI from this study? The potential is indeed quite high (although your mileage will vary). We need to reframe from attacking the big vision with a big project plan. As my friend Bill Inmon has always said, you need to start small and grow, but with a vision in mind.
We need to look at why there may be low usage of the data warehouse. It could be a number of reasons: inconsistent results, poor performance, lack of training, wrong selection of user tools (or more succinct, non-involvement of users during the selection process!).
We need to look at how we are building our data warehouse. Are we using a non-incremental development approach? Are we focusing on the business drivers? Are we building the data warehouse environment like we build the operational environment?
The data warehouse can bring substantial ROI into our organizations. We need to build the data warehouse with user support and build the data warehouse in an iterative process.
About the Author
Chuck Kelley is president and founder of Excellence In Data, Inc. and an internationally known expert in database technology. He has more than 20 years of experience in designing and implementing operational/production systems and data warehouses. Kelley has worked in some facet of the design and implementation phase of more than 35 data warehouses and data marts. He also teaches seminars, co-authored a book with W. H. Inmon on data warehousing and has been published in many trade magazines on database technology, data warehousing and enterprise data strategies. Please feel free to email him at firstname.lastname@example.org with comments (negative or positive) about this column or ideas for future columns.
For More Information
- The best return on investment web links: tips, tutorials, and much more.
- Do you have any technical questions about data warehousing or BI ROI? Post them--or help out your peers by answering them--in our live discussion forums
- Have a data warehousing or BI tip to offer your fellow gurus? The best tips submitted will receive a cool prize--submit your tip today!
- Ask the Experts: Our ETL, Data Warehouse Design, and Data Analysis experts are waiting to answer your toughest questions.
This was first published in May 2001