Why More Data Doesn't Mean More Clarity

The difference between information and insight, and what gets in the way of using data well in collaborative work.

Here's something I've noticed after years of doing evaluation work with food systems organizations, nonprofits, and public agencies: the organizations that struggle most with evaluation rarely have a data shortfall. They have an interpretation shortfall.

They're swimming in numbers. Surveys, dashboards, program reports, funder metrics. What they're missing isn't more information — it's a way of making sense of the information they already have. And in complex, collaborative work where outcomes are shaped by relationships, community context, shifting conditions, and multiple actors doing interconnected things, that distinction matters enormously.

More data doesn't automatically create more clarity. Sometimes it creates more noise.

The Assumption Hidden in Every Dataset

A few years ago, I was part of a public health planning effort focused on folic acid intake among Latino communities. The science was clear: folic acid taken during the childbearing years significantly reduces the risk of neural tube defects, including spina bifida — a condition that occurs at higher rates in Latino communities, partly because folic acid is commonly fortified in foods that don't feature prominently in traditional Latino diets.

The question on the table wasn't "why isn't our program working?" There was no program yet. The question was: what should we do? One idea that came up early was whether making free folic acid supplements available in common spaces (grocery stores, pharmacies) might be enough to meaningfully increase uptake on its own.

It seemed reasonable. The supplements were free. The need was documented. If access was the barrier, then increasing access should solve it.

Before recommending anything, we went into the community first. We held focus groups — not to validate an approach we'd already chosen, but to genuinely understand what was shaping people's lives and decisions.

What we heard reframed everything.

The barrier wasn't awareness. It wasn't access in the way we'd been thinking about it. It was time. It was transportation. It was working two jobs and not owning a car and already navigating the compounding weight of economic inequality in a system that wasn't designed with your daily reality in mind. The idea of making a dedicated trip, even to pick up something free and important, was just one more ask layered onto a life with very little margin.

Simply placing supplements in more locations might help at the edges. But without understanding what was actually shaping people's capacity to act, any intervention designed from the outside would be incomplete at best, and patronizing at worst.

The data that prompted the planning effort hadn't changed. The epidemiological picture was the same. But what needed to happen next looked entirely different once we stopped assuming we knew what the barrier was and started listening for it.

Data Is Never Neutral

This is the thing that's easy to forget: data doesn't arrive pre-interpreted. Every dataset comes embedded with assumptions — about what's worth measuring, who gets to define success, what counts as a problem, and whose experience is treated as the norm.

In the folic acid example, the assumption embedded in the early planning conversation was that low uptake would reflect a behavior gap — something to be fixed through better access or stronger communication. That assumption was invisible because it felt like common sense. And because it was invisible, it nearly produced the wrong solution before anything was even built.

The planning team could have moved forward with a distribution model, a marketing campaign, a reminder text. More data collection. Instead, because we interrogated the interpretation before committing to an intervention, the response became something more useful: a genuine reckoning with structural context, and a set of recommendations that could actually meet people where they were.

In food systems work, this pattern shows up constantly. A food access program measures how many people visited a farmers market but doesn't ask whether the produce was culturally familiar, affordable after accounting for transportation, or accessible without a car. A supply chain initiative tracks tons of food distributed but doesn't capture whether the farmers at the beginning of that chain are earning a living wage. A nutrition education program measures knowledge gained and then wonders why behavior didn't change.

The data isn't wrong. The frame is.

The Interpretation Gap in Collaborative Work

The problem gets more complicated when multiple organizations are working together — which is almost always the case in food systems work. Collaboratives, coalitions, and backbone-supported networks are the dominant structure for this kind of change. And when you bring multiple organizations into the same effort, you also bring multiple interpretive frameworks.

One partner looks at the same dataset and sees a capacity problem. Another sees a systems problem. A third sees a communication failure. These aren't arbitrary disagreements — they reflect genuinely different theories about how change happens, shaped by each organization's history, expertise, and community relationships.

In that environment, adding more data often doesn't resolve the tension. It amplifies it. Everyone finds what they're looking for.

What actually moves the needle is creating space for those divergent interpretations to surface and be examined together. Not to vote on the "right" answer, but to ask: what is each of these perspectives seeing that the others might be missing? What does it tell us that we're reading this differently? What question would help us get underneath the disagreement?

This is a sensemaking problem, and it requires a completely different set of tools.

What to Do Instead

If you're leading or funding collaborative food systems work, here are three reorientations worth considering:

Slow down at the interpretation stage. Most organizations move too quickly from data collection to conclusions. Build deliberate pause points where you ask not just "what does this show?" but "what are we assuming this means?" and "who isn't in this conversation who should be?"

Treat community expertise as an interpretive resource, not a validation step. In the folic acid project, community focus groups weren't a way to explain the data — they were a way to understand it. That's a fundamental difference. Community members aren't there to confirm what you've already decided; they're there to help you see what you couldn't see on your own.

Get curious about disagreement. When partners or stakeholders interpret the same data differently, resist the urge to reconcile it quickly. Different readings are often diagnostic. They tell you something important about what the data isn't capturing and where the real questions live.

The Deeper Point

The organizations doing the most thoughtful evaluation work aren't necessarily the ones with the best dashboards. They're the ones that have developed the discipline to ask: what are we assuming here, and what might we be missing?

That discipline is harder to build than a new data system. It requires humility, time, and a genuine commitment to being wrong in service of being useful. It also requires recognizing that in complex systems — where food, health, economics, culture, policy, and relationships all interact — data is always a partial picture.

Your job isn't to collect the whole picture. Your job is to stay curious about what you can't yet see.

That's what evaluation in complex systems is really for. Not to prove. Not to perform accountability. But to keep learning in the direction of what's actually true.

Bridgepoint Evaluation partners with food systems, nonprofit, and public-sector organizations navigating complex impact and collaborative change. If this resonated, subscribe to our newsletter for ongoing thinking on systems-oriented evaluation, strategic learning, and collaborative sensemaking.