Open Data Starts with Informed Consent

What is open data, really?

We'll be honest—we get a little excited about open data. There's something beautiful about structured datasets that anyone can use, reuse, or redistribute freely. While it's often talked about in government circles as a way to boost transparency and accountability, open data shows up everywhere, quietly powering innovation across sectors.

One project that still makes us smile is our work with BC's Community Energy and Emissions Inventory team. Their dataset tracks greenhouse gas emissions across buildings, municipal solid waste, and on-road transportation for 161 municipalities, 28 regional districts, and one region. What started as guesswork around community emissions has transformed into insightful data that now helps provincial governments, academics, students, and private companies collaborate on reducing emissions. It's messy, important work—the kind that happens when data meets genuine purpose.

We've also had the privilege of using open datasets to host hackathons, puzzle through thorny policy challenges, and watch students discover their love for data analysis and tackling wicked problems. (Curious about what's out there? Statistics Canada, Data.gov, and NOAA's weather data are good places to start exploring.)

The collection question

But here's where things get interesting—and where we need to pause and think more carefully. How does data become "open" in the first place?

Usually, consent happens at the moment of collection. You check a box that says something like "I agree to provide this data for a specific purpose." The data gets scrubbed of personally identifiable information before going public—names, addresses, account details all removed so sensitive information stays protected.

Most places have privacy laws governing this process. Canada has PIPEDA (a delightful acronym for the Personal Information Protection and Electronic Documents Act), and Europe set a new standard with GDPR starting in 2018. If you're into policy frameworks, these are worth a read.

The consent gap we keep seeing

Through our work with public sector teams across Canada, we've been part of co-designing services that aim to be more human, accessible, and responsive. Data collection sits at the heart of transparent, accountable government services. But we've noticed something troubling.

There's been a quiet assumption that transparency will just happen. That people will share feedback willingly. That collecting data to "improve services" justifies itself.

Here's what we've learned: open data without informed consent isn't open—it's extraction. And no matter how elegant your interface design, trust doesn't follow code. It follows care.

Reframing consent as design

We don't see consent as a legal checkbox to tick. We see it as one of the most human design challenges we face. Real informed consent means people understand what they're sharing, who's asking, how it might be used, and—crucially—what choices they have, including the choice to say no or change their minds later.

It sounds straightforward. In practice, especially when timelines are tight and systems are already built, it gets complicated fast.

We've encountered feedback forms with vague language like "your responses may be used to improve services." We've seen pre-checked boxes buried in long submission flows. We've witnessed "opt out" mechanisms that require three clicks, scrolling through a privacy policy, and what feels like a law degree to understand.

In those moments, consent becomes friction—something transactional, bureaucratic, something people click through without really understanding.

That's not what trust looks like.

What good consent actually does

When teams get consent right, something interesting happens. It doesn't slow the work down—it makes it stronger.

Good consent might look like a short, clear statement about what the data will be used for and who's responsible. It offers specific choices—follow-up interviews, future contact, research analysis—rather than bundled all-or-nothing decisions. It includes honest language: "We'll use your response to help redesign X. Your answers will only be shared with Y team. You can opt out anytime."

Most importantly, it provides transparent exit paths that are visible and accessible, not buried or ambiguous.

These small shifts communicate something profound: "You're not just a data point. You're a participant. And we respect your choice."

Why this matters more than compliance

This isn't just about following regulations, though that matters too. We genuinely believe that good consent produces better data and better outcomes. Whether we're collecting employee feedback, health information, or satisfaction surveys about government services, people are most likely to provide honest, detailed responses when they understand what their data will be used for and how their privacy will be protected.

This is especially critical when working with equity-seeking communities, where institutional mistrust has been earned through years of experience. In those contexts, good consent isn't just ethical—it's strategic. It says: "You matter. Your choice matters. And we won't move forward without you."

The uncertainty principle

Teams often ask us: "Do we have to tell users exactly what we'll do with their data?"

The answer is no—but we do need to share what we know right now. Informed consent doesn't mean predicting every future use. It means naming the current purpose, identifying likely data users, and establishing the boundaries we'll hold ourselves to.

When things change (and they will), consent can evolve too. That's part of the agreement—a living element of how public trust gets built and rebuilt over time.

Where we see things go sideways

In our work across the public sector, we've noticed patterns in where consent tends to break down:

Consent gets assumed rather than clearly requested. Language becomes vague, technical, or defensive. Choices get bundled into all-or-nothing decisions. Exit paths become invisible or nonexistent. Nobody clearly owns the data relationship—people don't know who's actually responsible for what they've shared.

These aren't just usability problems. They're trust problems. When consent is weak, even the most thoughtfully designed service can feel like surveillance.

Getting started (without starting over)

We work with public sector teams to weave ethical consent practices into service delivery. Here's how we often begin:

Bring consent into the design conversation early—don't wait for legal review. Make it part of the journey from day one. Design layered, specific consent that offers opt-ins for different levels of engagement. Let people choose their own level of participation.

Test your consent language the same way you test your interface. If your grandmother or neighbor can't understand it clearly, it needs another draft. Make opting out as straightforward as opting in—build visible exits without shame or barriers.

And here's something we feel strongly about: share your consent design patterns with others. These solutions should be open-source. We all benefit when the models improve.

The bigger picture

Open data, at its core, is about giving people agency over what they share and clarity about what happens next. It's about building public digital services that people can believe in, not just use.

The systems we design today shape the trust relationships of tomorrow. We have the opportunity—and responsibility—to get this right.

Let's build something worth trusting together. Let’s chat.