Start With a Good Problem to Be More Data Driven
If you’re reading this post, you want to be more data driven. How do I know this?
As I shift my business to social service agencies, I’ve been talking to many executives in those agencies. And almost every one of them tells me they want to be more data driven.
So, I’m guessing that—if you’re like most of your peers—you want to be more data driven too.
But the question remains—where do you start?
If you want to be more data driven, you need to start by finding a good problem to solve. But what makes a good problem?
In business intelligence, a good problem has three characteristics:
- You can see the bleeding
- You have the data to solve the problem
- You have the people to solve the problem.
1. You Can See the Bleeding
Apologies if I’m being too graphic, but I want to make my point clear. To get started in business intelligence, you need to find an obvious problem that is causing pain RIGHT NOW.
Most social service agencies have real challenges in some areas of their businesses. But it’s not enough to identify a challenge. You need to identify a challenge that has urgency.
Yes, we know that continuous improvement is important. And we know that Medicaid is moving to value based payments, which will bring its own set of problems. But if you’re just getting started with business intelligence, you want to start with a pain that’s present RIGHT NOW.
(If everything is going just great at your agency, you can stop reading now. Even better, tell me how everything became so great so I can write about it.)
2. You Have the Data to Solve the Problem
Next, think about the data you’ll need to solve the problem. Does it come from one system? Two systems? Five systems? Are those systems on premise or in the cloud? Can you connect to the appropriate database?
For your first project, you want to choose a problem that involves the smallest number—and the most accessible—data sources.
3. You Have the People to Solve the Problem
As a consulting firm, we have the technical and data expertise you’ll need to get your BI project going. But what we can’t guarantee is user commitment.
Without user commitment, your project will fail. Users need to recognize the problem as a problem and have the time and desire to work on solving it. Consultants and IT folks can do a lot of the work. But only end users can make sure the data makes sense in the end.
What You Don’t Need
You’ll notice I didn’t say you need to choose a BI or visualization tool. There’s a reason for that.
Almost every nonprofit I work with already has SQL Server (and its included SQL Server Reporting Services) and Excel. Or they can get them (and Power BI) pretty inexpensively.
They may not have the ultimate tools for every project, they certainly have enough to get started.
In future post, I’ll discuss the different types of BI projects and data that I work with for most of my clients.