(Illustration by iStock/Fahmi Ruddin Hidayat)
Many nonprofits are data-rich but insight-poor. That means they collect vast amounts of information—for compliance, in CRM systems, health records, or program logs—but lack the internal capability to turn that data into a road map for their mission.
In the social sector, the dominant conversation is funding, but funding alone will not fix an analytics problem. A nonprofit that gets money but doesn’t know how to look at its own data creates a “leaky bucket” effect: No matter how much money pours in, the impact leaks out because the organization is not learning or improving.
Boards and funders need to treat data analytics as essential infrastructure rather than a “check-the-box” reporting task. Building your own data skills can be more a powerful organizational investment than simply raising more money. Most nonprofits already collect more information than they use, but without analyzing it, they repeat the same operational mistakes year after year, turning dormant data into active learning. It also helps organizations tell more credible stories and marshal better evidence of impact. Finally, reducing dependency on outside experts keeps analytic capability in the organization long after any one grant cycle ends.
Diagnosing the ‘Leaking Funnel’
For example, take Hegira Health, a large community-based behavioral health provider in Michigan operating a co-response program with law enforcement. When our collaboration began, data was being diligently locked in raw Excel files. On paper, the program appeared to be a soaring success: Referrals were increasing by 35 percent annually. Yet leadership sensed the program was not reaching its full impact.
By applying a structured diagnostic to their raw referral logs, we visualized their program as a Referral Funnel, which was an immediate revelation for the board. The data, as we framed it, showed a high follow-up rate (76 percent), which showed that staff was working diligently to make contact. However, the engagement rate dropped to 42 percent. This 34-point gap was the leak: For every 100 people contacted, 58 individuals never became active participants.
This analysis made the conversion bottleneck visible, allowing the team to ask targeted questions about intake timing, communication methods, and process friction. It illustrates how headline growth metrics, like a 35 percent increase in referrals, can actually obscure operational weaknesses. By translating raw data into a structured strategy, Hegira Health moved from reactive reporting to proactive intervention, ensuring that more individuals in crisis received the care they were promised.
Data as a Sustainability Engine
Enterprise for Youth, a San Francisco-based organization providing internships to under-resourced youth, faced a paradox: record-breaking revenue but fluctuating donor counts. A donor segmentation analysis uncovered what I call the “Recruitment Treadmill”: Only 30–40 percent of donors each year were retained, meaning the organization had to work twice as hard to recruit new donors just to keep the total count stable.
Mapping the last year of giving across the organization’s entire history uncovered a list of ~9K lapsed donors who had not been engaged since 2020, all former supporters whose relationship with the organization had simply vanished. The data made clear that re-engaging even a small fraction of these existing donors would be far less costly than recruiting new ones from scratch. It also raised a more important question: Why had these donors stopped giving in the first place? That question, which the organization had never systematically asked before, became the foundation of a new stewardship strategy focused on relationships rather than acquisition.
In this way, headline metrics can actually mask retention fragility; investment in analytic capability may yield greater sustainability than continued marketing spend.
When Growth Misleads
A youth mental health technology app, AHADI, developed by an East African nonprofit TAHMEF, collected extensive data on downloads, active users, social media engagement, but lacked a system to analyze these metrics.
When we organized their dataset into a unified tracking framework, the gap between appearances and reality became clear. Monthly downloads had exploded from 2,000 in January to over 101,000 by June. However, the retention rate hovered between just 1 and 6 percent, meaning fewer than two of every 100 June downloaders remained active the following month. Without this analysis, the organization would have reported “100,000 downloads” as a headline success. Layering social media data against app usage revealed the same pattern: Instagram engagement looked strong on individual posts, yet follower counts stayed flat, and helpline conversations peaked at just 23. This was a clear sign that downloads were not translating into real support. The analysis gave leadership an honest picture to share with international funders, replacing an impressive-sounding headline with a meaningful story about where the app was falling short.
This case underscores the danger of equating reach with impact. It serves as a reminder that in the digital health space, a download is merely an invitation; true impact is measured not by how many people enter the room, but by how many find the value to stay.
Building a Data System
The pattern across all three case studies follows the same four steps:
- Consolidation: Nonprofits rarely lack data, but they lack organized data. The first step is pulling information from scattered spreadsheets, CRM systems, and program logs into a single, clean source.
- Automated Cleaning: Logic needs to be built directly into the workbook so errors surface immediately as new records are entered.
- The Three-KPI Rule: Many groups try to track too many metrics, which causes confusion. Instead, they should pick the three most important numbers that help them make decisions every week.
- Actionable Visuals: Translating those KPIs into visuals that update automatically and that a board member can interpret instantly.
What Funders and Boards Must Do Differently
Nonprofits cannot close the data-utilization gap alone. Funders decide what gets resourced. Boards decide what gets governed. If neither group treats analytics as a basic expectation, it will remain an afterthought regardless of how motivated the staff may be.
1. Treat Analytics as Infrastructure
Do not think of data tools as an extra cost to minimize, but as part of the building. A dashboard that updates itself, a clean list of donors, and a trained staff member should be things that funders want to pay for. Spending $15,000 to help a group understand their data can make a $150,000 program work much better.
For example, the Waterford Community Coalition, a youth-serving nonprofit in Michigan, invested approximately 20 hours in building a structured performance dashboard. The project was valued at $4,324 in pro bono analytics support. The result was not just improved visuals, but year-over-year comparisons that allowed leadership to see trends more clearly and communicate performance more effectively to stakeholders.
The scale of the investment was modest. The shift in visibility was meaningful. Relatively small investments in internal data structure, whether through grant dollars or supported pro bono work, can materially strengthen an organization’s ability to interpret and communicate its impact.
2. New Due Diligence
When funders only look at audited financials and a logic model—necessary but inadequate—they should also assess whether the organization has the capacity to learn from its own operations.
Three questions can quickly answer the question:
- What are your three governing KPIs, and who owns them? If leadership cannot answer this clearly, performance is likely to be tracked reactively rather than strategically.
- When did your data last cause you to change a program decision? If the answer is vague, analytics may exist only for reporting—not for management.
- If a consultant built your reporting system, can your staff maintain it without them? If not, the organization does not have internal capability—it has a dependency.
These questions shift the conversation from how much you spend to how much you understand, signaling to leaders that learning capacity is just as valuable as fundraising growth.
The Ability to See Clearly
Where data exists but it is not systematically translated into insight, neither advanced software nor specialized data science teams is required. What is required is leadership’s commitment to defining a small set of governing metrics, reviewing them consistently, and building internal ownership over time. Organizations that endure and improve are those that can answer, at any given moment: Are we reaching the intended population? Are participants or donors staying engaged? If not, where precisely is the breakdown occurring?
Treating analytics as governance discipline rather than as a reporting exercise may be one of the highest-leverage shifts available to boards and funders seeking durable impact.
Read more stories by Gaurav Mittal.