Frequently Asked Questions
Q5. How much data do I need in order for DCI to help me?
;)
We’ve had success on projects with as few as 100 rows of data and with more than a million. Although more data is often better, the quality of the data is at least as important as the quantity. The amount of data needed also depends on the complexity of the relationship you are trying to understand or model. The more “noise” there is in the data, the more it helps to have additional rows from which a model can learn how to filter out that noise. Marketing data tends to be quite noisy, since it reflects the complexities of human behavior. Scientific data, on the other hand, is often more clean because it deals with inherent, reproducible relationships. If you don’t have enough data for us to give you a good result, we will let you know. In addition, we have partners who can enhance your data with additional attributes (i.e. adding more demographic, geographic, lifestyle, and buying information about your current customers).
Q6. How much do I have to prepare my data before DCI can work on it?
;)
The most efficient and economical scenario is one in which you have created a rectangular table of data that includes all of the attributes you want us to study or model. Alternatively, we can assemble the data for you, but this can be more expensive and time-consuming. Whenever possible, we like to collaborate with your people in bringing the data together. This way we leverage your staff’s expertise on how the data fits together with our expertise on what factors/elements are important for analysis, mining, and modeling.
Q7. How much does DCI charge for a typical project?
Since Discovery Corps does custom work for our clients, there really is no such thing as a typical project. But we can tell you some important things about our approach to pricing. When we can clearly define the requirements of your project, we can propose a fixed price. When you need advice or consulting, we charge an hourly rate. If you want us to function as your outsourced analytics department, we draw up a retainer agreement. Occasionally a client is at a very exploratory stage in working with their data and they’re not sure what form the final results will take. For that kind of ad hoc work, we work on a “time and materials” basis. Finally, in other contexts (such as new product development), we can include a royalty arrangement. To borrow a couple of phrases, our philosophy is “Win-win” or “No Deal.”
Q8. What kind of Return on Investment can I expect when I hire DCI?
;)
While we cannot guarantee the ROI of a project, we can say that clients have achieved good multiples (3:1, 5:1, 10:1 and up) for their project investments. And the payback time has been as little as a couple of weeks to a couple of months.