Data warehouses have been around for a long time in the digital information universe, at least since the early 1990s. Early on, organizations recognized them as a means of storing and using data for planning and decision-making. In fact, information technology often utilizes a separate data warehouse for each area of a business. These are often called data marts, as opposed to enterprise data warehouses which contain data for every facet of a single entity.
Every association seeks to increase membership using various tools such as regular publications, meetings, advertising, mail outs, and more. Very often, these same associations have enormous amounts of data concerning their members and prospects socked away in a data warehouse; but it lies mostly unused. Why not use this dormant data to increase membership?
A data warehouse is a digital read-only storage facility for all manner of static, non-changing data. It differs from a database in that a data warehouse stores unchanging data. Databases store and rapidly update real-time data for immediate use, such as transactional data. Data warehouses store data for use in queries, research, and analytics.
When using a data warehouse, an organization must first determine what data will be stored in the warehouse and where it can be found. Some data will be internal, such as member information, donor information, subscribers, event attendees, and more. Other data will be gleaned from outside sources for comparison or research purposes. The data is cleansed from any errors and then stored in predefined locations in the data warehouse for later use.
Because a data warehouse is designed for query and analysis instead of transaction processing, it typically contains historical data from several sources. Historically, data warehouses have used structured repetitive data that has first been filtered or distilled. In recent years, data warehouses are using contextual information that can be attached to unstructured data. This is possible through the advent of artificial intelligence and contextualization.
Static, non-repetitive data, like emails, survey comments, and conversations, are handled differently than repetitive occurrences of data, like that from metering, a click-stream, or machine processing. With contextualization, organizations can now make sense out of such data. Establishing a context for the data is key.
Associations can use the dormant data in their data warehouse to increase member retention by using the data to identify those members who are least likely to renew their memberships. These at-risk members can then be targeted with retention efforts before they actually cancel their membership. Since it is much easier to keep existing members than reclaim canceled members, this can be an important part of your association retention efforts. .
Using predictive analytics, you can use the data in your data warehouse to create a list of members to target. Here is how it could work:
Now you have a list of members who show significant risk of leaving your organization, complete with a profile that shows behavior. You will also have a list of behaviors that demonstrate a high level of retention. With these lists, you can not only target those at the highest risk of leaving before they actually leave, but you can bolster your overall retention efforts by focusing on the high-retention behaviors.
At CIMATRI, we help association and non-profit leaders with digital strategy, workforce culture, service design, and association IT. Contact us for help by calling (571) 249-2719 or filling out the online contact form today.