Data Maturity Part II: How Grown Up Are You?
Data Maturity Stages
In Part I of this blog series, I introduced the concept of data maturity, its importance to organizations, and steps for becoming data mature. Today, we’ll tackle the ways in which organizations can be thought of as existing in different stages of data maturity and how you may move through them. Particularly relative to reporting, knowing where you are on the data maturity scale will help you identify the right approach and the right tools to move forward.
Early Phases: Piloting and Building
Phase I: Piloting (Infancy)
Your data is new and cute, but you can’t really leave it alone when you go out for dinner. You may be collecting some basic data, but you haven’t created a data strategy and made clear decisions about what you are collecting and why. There’s a bit of chaos in the process, such as random addition of indicators when funders ask for them or keeping old indicators you don’t need anymore. Chances are your technology is pretty simple, and you are just trying to get some data out of it for required reports. In this phase, people tend to be using Microsoft products like Excel to review information ad hoc, or Google Sheets if they are a Google shop. No one is in charge of data hygiene and you don’t have an internal data champion.
Phase II: Building (Early Childhood)
Your data can walk and talk, it’s pretty entertaining, and growing like a weed. But you still wouldn’t give it the car keys. You’ve done some thinking about your metrics, what you collect, and why. You’ve got more data to work with, and you can easily get the data out of your existing technology. You might use flat files or even simple databases, like Access or AirTable. At this phase, organizations tend to be using simple low-cost tools like Microsoft BI or Google Data Studio to visualize their information mostly with snapshots. However, data strategy work is inconsistent and there are significant ebbs and flows, causing gaps in data collection, thinking, and strategy. You are still limited in both expertise and technology to connect data across systems and in doing long-term trend analysis.
What Can You Do Here?
At this level you can do basic performance measurement (How much? How many?), process evaluation, and some basic (often short-term) outcome measurement.
What Do You Need to Level Up?
You need two (technically three) kinds of resources to get from the piloting and/or hanging on for dear life stage, to a more functional data mature stage with users who are really good at using the tools you do have:
- Expert consultation on data strategy, in the form of evaluation/research expertise and
- technology expertise, either in-house from the folks managing the current simple systems, or outside expertise to advise and help support a transition to a more user-friendly, comprehensive, and integrated data system.
But even more than that (and this is key) you need a data champion. As I mentioned in Part I, a data champion is someone who believes in the idea of using data to improve programs and make better decisions (not just to fulfill requirements and check boxes) and pushes the organization and its leaders to buy in to the idea and implement it. The higher up in the organization the data champion is, the more successful the move towards data maturity will be.
More Advanced Stages of Data Maturity: What Does That Look Like?
Phase III: Data Adolescence and Beyond
At this point your data is pretty grown up. It can do most of the things it needs to, but may still make bad decisions without proper guidance. The most advanced level of data maturity involves: having clearly defined metrics for all your programs, validation by external third parties or based on credible research linking the organization’s outputs to outcomes, and a process for staying on top of new research. The organization’s Board and the Executive Director regularly use data to assess performance and make improvements, and the organization has a right-sized staff to handle data collection, strategy, cleaning, and visualization. Hypothesis testing (internal or external) is happening at this phase as well.
At this point the organization most likely has a custom data warehouse that integrates data across systems and constituents to produce the necessary metrics for decision-making.The system is straightforward for those entering data, and staff can see reports that are actually informed by their data.
- What can you do here?
- Now you can test different hypotheses, incorporate data from other sources, look at long-term trend analysis, and do more real-time reporting. You can go beyond how much and how many and start to answer questions about what, why and when more effectively.
- What are your needs here?
- External evaluation and validation, software developers/data warehouse managers, and ALWAYS data champions as high in the organization as possible. You will also need (probably at every stage) sufficient staff and resources.
Resources: You Can’t Level Up Without Them
- Internal: Data champion, data cleaning and data hygiene practices (and staff to ensure them), staff to maintain data strategy and ensure alignment as changes take place. At this level in particular, these practices can’t continually be added onto current programmatic and organizational work. Staff dedicated to data strategy and data hygiene become necessary.
- External: Third party research partners to validate program results, and most likely a software vendor or team to develop and maintain the data warehouse and reporting system.
If you use Salesforce, you need both an internal staff person to maintain, and an external consultant to build and make changes, if you use ETO you can have an internal person do the building but it takes significant expertise and you have to re-train with turnover. With our system, Incite, you only need an internal data champion as the contact person, because all maintenance and building happens with us. When thinking about the cost of a system, you have to think about what internal capacity you want to maintain, versus what you want an external party to handle.
No matter where you are on the maturity scale, the most important thing is to recognize where you are, know that stage’s limitations (it’s OK to need training wheels) and to know what it takes to get to the next level.
If you want to talk more with us about how to be more data mature, contact me at email@example.com!