We are all feeling the pressure to quantify our work. Donors want metrics, the board wants key performance indicators, our staff want clear and measurable goals. Increasingly, we are expected to be able to take the nuanced work we do in the community and package it up. The absolute focus on quantifiable measurements is harmful, but it is clear that in the right situation measurements are incredibly powerful tools.
Quantifying our work requires careful collection and manipulation of data. We need to track our clients, our programs, our donors, our grants, our funding restrictions, our phone calls, our funding prospects, our intakes, and a hundred other things that we do day-to-day.
This data doesn’t only support metrics, it allows us to manage the incredibly complex flow of our work. Keeping track of our interactions with a client from intake to discharge is much easier when each element is carefully recorded in a database.
The holy grail of these systems is a single unified database that our entire organization utilizes. In an ideal scenario, we can track all of our clients and everything they do and all of our donors and everything they do in one place. Every single call and interaction is stored. And we can use that data to gain insights into our work and satisfy our funder’s insatiable appetite for metrics.
But how do we get there? And where is “there?” And is it even worth it? In this article I am going to attempt to lay out four different “stages” of data maturity to help you understand your current state and the benefits and costs of improving. Not every organization should make these investments, and I’d like to help you figure out what is best for your specific needs.
From Head to Spreadsheet
The first step for any organization in moving to digital tracking is to get information out of staff members’ heads and into a spreadsheet. For most of us, this is almost instinctive! Need to call a bunch of organizations: “I know, I’ll make a spreadsheet!” Have to keep track of tasks for a small project: “a spreadsheet will help me do that!” Want to organize lunch for the office: “a spreadsheet would be just the thing!”
Spreadsheets are incredibly powerful and flexible tools for tracking data in our organizations. And they do a very good job tracking information for small and discrete projects. Spreadsheets are more or less free, and you don’t need to be an expert to create them. For some organizations – particularly small ones – tracking data in spreadsheets is all they need.
There are, of course, drawbacks. Spreadsheets are almost useless when multiple people are attempting to use them. They are hard to find, hard to use concurrently, and every one of your users are going to enter data slightly differently. As a result, it is very difficult to generate good measurements from a spreadsheet (I’m sure this is not news to you).
Spreadsheets should never go away, but if your organization has multiple staff and are having difficulty getting reliable measurements, you might want to think about the next step up…
Using a Relational Database
Relational Databases (think Microsoft Access, Salesforce, Dynamics CRM, and Raiser’s Edge) allow you to store information in a highly structured way. Contacts are stored in one place, donations in another, and the two are connected. With relational databases your organization gains the ability to enforce standards of data management. You can force users to record data in certain fields, create reminders and compliance reports, and create validation rules to make sure that users are meeting certain requirements.
Because the data is structured and standardized, relational databases allow you to create reports that give you easy access to critical information. Relational databases also allow you to distill a large number of transactional data points into simple measurements. These are the metrics and key performance indicators your board and funders are crying for.
Unfortunately, relational databases can be expensive and difficult to maintain. Unlike spreadsheets, they cannot be customized by just anyone and need to be carefully structured to reflect your organization’s specific processes (or you need to change your processes to reflect the database). As a result, implementing a relational database requires significant up-front investment. It also requires skilled staff members who can monitor the database, ensure data is being entered correctly, and get information out of the database in a useful format. Moving from spreadsheets to databases is a very large step, but it is one that most organizations will need to take once they grow beyond a few employees.
Un-Siloing your Data
Most organizations that already use a relational database actually use several. Over time, different programs find or develop their own databases. Fundraising uses something off the shelf, this program uses an Access database, and this other program found funding to develop something from scratch. The result is fragmented data.
The next step in data management is to combine all of these data sources into a single database. This allows organizations to understand the full extent of their interactions with individual constituents and funders. It also allows organizations to standardize information collection across multiple programs and ensure that metrics reflect the same information for every service the organization offers.
Creating a single centralized database is often prohibitively expensive. Since most nonprofits provide dozens of only marginally related services, there is very rarely a single database that will support the entire organization. As a result, organizations must invest in a custom database (either build from scratch or on top of Salesforce.com or Dynamics CRM). This is very expensive, and it also limits the flexibility of each individual program to react to their changing needs.
Creating this mater database is often not the right decision for organizations. Just because it feels tidy to have a single database doesn’t mean you should do it! The main benefit is for organizations that interact with the same people in multiple ways. If your donors are also receiving your services, then you will benefit from merging those databases together. If each of your programs services the same population then having everything in one place will allow you to see the true impact of your organization.
If, on the other hand, each of your programs interact with a different set of people and your donors never receive services, then you might want to save that money and keep your databases separate.
Data Mining, Visualization, Mapping, and Beyond
The final step in data management is to move beyond fulfilling specific information requests and to look for new insights using your data. This is known as data mining. With data mining you approach your data with general curiosity and poorly formed questions (“I wonder were my clients are located,” or “what kind of donors give us the most money”) and use specific tools and techniques to look for trends that will help you work more effectively.
Doing data mining does not require a single unified database. It does not even require a relational database (although it helps). Instead, it requires expertise with specific tools (and occasionally the field of statistics). And, above all, it requires good quality and comprehensive data.
A few years ago, performing data mining was not possible for a small nonprofit. Today, there are dozens of tools that make it relatively easy for someone with patience to make remarkable discovery. A few weeks ago, I used a plugin for Salesforce to help us figure out where to run a series of in-person workshops in California. This would be been impossible without a mapping tool of some kind:
A map with all of the organizations in our database in California