We regularly espouse the importance of using data to make decisions. In fact, it’s the basis of our organization. We use our advanced analytics to compare data and create concrete suggestions for how an organization can move forward. This process, as you’ve heard us discuss in the past, is called the 7 Steps to Data–Driven Decision Making. We apply it to our work, and we train our clients on how to incorporate it into their processes.

Recently, we were asked by a client to share a distilled version of the process. Our client, who was trying to instill this culture in his organization, needed a go-to guide for getting his staff on board. He knew that by getting his organization to think in a data-driven way, he could improve its outcomes and effectiveness. Fundamentally, he understood that by employing data-driven decision making, he could more successfully apply for grants, request funding, and direct his already stretched staff to achieve the highest impact with the lowest investment.

Here is what we told him:

1. Framing the issue

When concerns pass by the desk of nonprofit professionals, how do leaders know which ones need to be studied and which they need to devote their limited resources toward understanding better?

Ask the following questions about your issue:

• How did this issue come to my attention?
• How many people does it impact?
• What proof do I have that this problem exists?
• What else do I need to know in order to evaluate this issue?

2. Hypothesis development

Hypothesis development is a key step that organizations often skip. This complicates the remaining steps of the process and makes it difficult to draw clear conclusions to inform decision-making.

Follow these steps to develop an effective hypothesis:

• Be specific and measurable. Start by identifying the cause and effect of the issue you framed in step 1.
• Develop a logic model of inputs and outputs using “if…then” statements. Here is an example: "If the nonprofits who receive grant money use it to train their volunteers (input), then the volunteers for those organizations will feel more committed (output)."
• Generate as many hypotheses as possible with the right group of people. When you think you’re done, brainstorm more!
• Prioritize those hypotheses you’d like to test. Which relate to the mission? Which can you control?

3. Data Collection

Once you have a list of the prioritized hypotheses, you can then determine what data you need to test them. Collecting the right data is much easier when you know what to test.

Use these tips to collect the right data:

• Determine what data you need based on your hypotheses. Don’t just go after what’s easiest but create a plan to gather the data you truly need.
• Consider multiple methods of collection (surveys, tracking systems, internal databases).
• Be careful not to collect too much data or use sources that aren’t valid, like databases with incomplete or errant data.

4. Data analysis

You don’t need to be a statistician to successfully analyze and later interpret data. Basic statistical skills will suffice here.

You just need to know how to:

• Digest and assess statistical data provided
• Calculate summary statistics (average, median, standard deviation and distributions)
• Understand statistics well enough to explain the findings to key stakeholders.

5. Data interpretation

Data interpretation is straightforward for those who set up their “experiment” correctly in steps one through four. The analysis from step four allows you to test which of your initial hypotheses are true or false.

Compare your data in several ways:

• Longitudinally to see history and trends
• Demographically (zip code, income, gender, etc.)
• Across parallel survey questions
• Against peers

6. Decision making

After you test your hypotheses and determine which are confirmed or rejected by the data, there will be a variety of decisions that you can make. Since time is limited and you cannot accomplish everything, it is essential to prioritize your options.


Make sure your action items:

• Align with the mission and vision of your organization
• Have a mix between “quick wins” versus longer term initiatives
• Do not already exist in other parts of your organization (so that you do not mistakenly double efforts)
• Are incorporated into an action plan that outlines requisite resources (staff, budget, materials, etc.), sets measurable goals, details a timeline, and assigns ownership.

7. Communication

In the final stage of the process, you must focus on communicating the results of your analysis. Not everyone will be receptive to the information, especially if it highlights weaknesses or challenges anecdotal beliefs commonly held in the organization.


For each stakeholder group:

• Identify the priority or focus area that the group needs to understand.
• Create a “story” that helps your audience understand what the data says.
• Identify the hypotheses you tested, and then briefly describe your methodology for data collection and analysis.
• Make sure to highlight the recommendations you have developed based on the data and the relevant steps from your action plan.

As we told our client, your staff develops analysis skills, they will be able to know with confidence that they are making decisions that are in line with their vision. Staff, board members, volunteers and donors will be motivated and engaged because they will know they are investing in something that works. Their efforts will be aligned in a way that will allow the organization to focus on achieving its mission—and they will be able to measure that progress along the way.

We will dive further into each of these steps, giving you more concrete and specific tips in the future. If you have further questions about this approach or are interested in the Building Data Competency program for your organization, email info@measuring-success.com.

Check out our next installment Step 1: Framing the Issue