Field Notes: Who is missing from your data?
The question.
How do we understand audiences when some people can't even reach our surveys because of blocked apps, poor internet, or security risks?
Why this matters.
Traditional survey methodologies assume everyone has an equal chance of responding. But what happens when entire groups can't participate because Instagram is blocked, their phone can't handle the app, or responding puts them at risk?
We end up with precise-looking statistics about people we actually reached, while the people we most need to understand remain invisible. This leads to strategies built on incomplete pictures.
What we're exploring.
Instead of trying to estimate how many people exist in a population, we're mapping the actual barriers that prevent people from reaching information—and sizing the groups affected by each barrier.
Here's how:
- Start with what we can verify: App store logs, survey responses, and partner reports about who tried to access what and what happened.
- Group by barriers, not demographics: Instead of "women ages 18-35," we look at "people blocked from app stores" or "people with unreliable connectivity."
- Cross-check across sources: When we see the same pattern in server logs, surveys, and field reports, we know we're onto something real.
- Report rough ranges, not false precision: We say "between 10,000-50,000 people can't access this" rather than "23,847 people" when we don't actually know.
This process tells us exactly where people get stuck, making it clear what needs fixing.
More questions for you.
- What stops your intended users from reaching you? Blocked platforms? Cost? Safety concerns?
- What data do you already have that shows where people leave your services?
- How do you talk about uncertainty with partners or funders who want definitive numbers?
- What's the clearest way you've seen someone visualize "we're confident about this, less confident about that"?
Read the full article on this topic here: Constraint-based segmentation: A research alternative for systematic exclusion
We'd love to hear your thoughts and feedback. Don't hesitate to get in touch at hello@gazzetta.xyz.