Data Mart vs. a Data Warehouse: What’s the Difference?

Too often in advancement, a data mart and a data warehouse are used as interchanged terms. But they are, in fact, two very different things.

 

Here’s the difference between a data mart and data warehouse and best use scenarios for your advancement needs.

 

Data Mart

A data mart is a curated subset of data often generated for analytics and business intelligence users—typically limited to holding warehouse data for a single purpose, such as serving the needs of a single line of business or company department. Data marts are often created as a repository of pertinent information for a subgroup of workers or a particular use case.

 

Data marts are a subset of a data warehouse. Organizations may use data marts to provide user access to those who cannot otherwise access data. Data marts may also be less expensive for storage and faster for analysis, given their smaller and specialized designs.

 

Other significant differences between a data mart and a data warehouse include the size, range, and sources.

 

  • Size: Data marts are typically less than 100 GB, whereas a data warehouse is typically larger than 100 GB and often a terabyte or more.
  • Range: Data marts are limited to a single focus for one line of business, whereas a data warehouse is typically enterprise-wide and ranges across multiple areas.
  • Sources: Data marts include data from just a few sources, whereas a data warehouse stores data from multiple sources.

 

In the context of your advancement CRM, a data mart is beneficial for creating a suite of reports—either from a daily operational standpoint or issues that may require information you need to make decisions. In addition, these reports enable you to see your progress or implement a course correction if needed.

 

Another case where you may want to leverage a data mart is to maintain “supplemental information.” Supplemental information may include a report you use on an infrequent basis. You can use your data mart to create a table or a couple of tables to house supplementary information to leverage reports that you can refer to once or twice a year.

 

This is beneficial because you may not want to build something special in your advancement CRM to hold the report because you use it infrequently.

 

Another use for a data mart may be to store what’s considered “frozen data”–a snapshot in time of your transactional data. Data marts enable you to store those exact numbers and easily reference when needed.

 

For example, you created a Voluntary Support of Education (VSE) report. Each year, you reported specific numbers to an external entity on how your institution was doing. But your transactional system is constantly changing.

 

This is because new data is being input and updated daily in the CRM application. If you were questioned on a previous fundraising year, would you be able to back it up? You may get the same data if you run the report out of your current application in your advancement CRM. The “frozen data” that was stored is needed to be a reference. A data mart enables you to maintain those transactions and tag them appropriately.

 

Software Platforms for Data Marts

When it comes to which software platform to use for a data mart, you should select the one that best aligns with your advancement CRM. Avoid using multiple vendors, as this creates confusion.

 

For example, you have one vendor for your advancement CRM and one for your data mart, and they don’t get along. But you need the transition of information from your advancement CRM to your data mart to be seamless.

 

Opt to use something from the same software vendor or choose another platform that has a very close partnership with the software vendor of your advancement CRM.

 

From a usability perspective, you want something user-friendly. On the other hand, you don’t want an overly technical solution for your data. This is because the audience for your data mart may be tech-savvy, but they’re not programmers, and they shouldn’t have to be. So instead, select something for more of an analyst-level ability to interact with the data mart. Save your heavy technical resources for something else.

 

Data Warehouse

A data warehouse is a relational database designed for analytical rather than transactional work, capable of processing and transforming data sets from multiple sources. It draws in data from disparate systems, taking on more of an enterprise view of data.

 

Data warehouses receive information across your organization. This is not fundraising or advancement specific; the data warehouse should be feeding information from all areas of your organization. Best usages for a data warehouse include:

 

  • Healthcare: A data warehouse combines information from advancement with a patient information system.
  • Higher Education: A data warehouse combines information from an advancement CRM with a student information system like the registrar’s office.

 

Software Platforms for Data Warehouses

Now is the time to leverage your central information technology (IT) resources. When it comes to the software platforms for your data warehouses, you want the combined support of the IT resources or expertise of individuals for the other systems.

 

For example, your experts in your advancement area are usually not the same as those in your patient area. So instead, use your central IT who has colleagues and expertise in all these different systems—that’s whom you want working on your data warehouse projects.

 

Select a platform they know best and can manage the data from these different systems. For example, your data mart needs to communicate well with your advancement CRM, but your data warehouse needs flexibility with many other systems.

 

Slow and overloaded data warehouses are often the underlying reason for creating data marts and frequently serve as their underlying data source. Often, as data volumes and analytics use cases increase, organizations cannot fill all analytics use cases without degrading the performance of their data warehouse, so they export a subset of data to the mart for analytics.