Surveys are too easy... and too hard.
You can launch a survey faster than you can microwave popcorn.
Instagram, Slack, LinkedIn, Google—they all let you toss out a question to the world in seconds. No training. No vetting. Just hit send and go.
Surveys have become too accessible. Really. It’s become way too easy for anyone to click a button and collect data.
Surveys are so easy — so common — that we’ve started treating them like they’re meaningless. Forgettable. Background noise.
“It’s just a survey, no big deal.”
But I’d argue it’s a huge deal.
People treat surveys like they’re harmless. Like, “It’s only data,” or “It’s just a few questions.” Or my least favorite: “Done is better than perfect.”
But let’s think about what surveys actually drive:
What products companies develop
What initiatives philanthropies fund
What priorities governments take on
What the media reports as “public opinion”
Survey data shapes the world around us. We are making high-stakes decisions based on low-quality data, every single day.
In fact, we often demand numbers to justify strategies, funding, product launches, political messaging—everything.
Data is currency. But somewhere along the way, we’ve completely devalued it.
Surveys are one of the most common ways people and organizations collect data—and yet, we’ve made them so easy to deploy that we’ve eroded the respect for the skill it takes to do them well.
That’s created a dangerous gap.
For a society obsessed with data-driven decisions, we seemed to have dropped the ball on quality control. We’ve made data so accessible, it’s become worthless.
Meanwhile, good survey design—the kind that leads to useful, trustworthy data—is almost completely inaccessible.
You won’t learn it in school. You probably won’t learn it on the job. And unless you know exactly what you’re looking for, you’re not going to stumble across it online.
Survey design is one of those rare things where you can be bad at it and never know—or realize only after the damage is done.
I’ve seen it again and again. People use data from surveys with:
Confusing or vague questions
Missing or overlapping answer choices
Samples that only reflect a fraction of their audience
And then they treat the results like gospel. Not necessarily because they’re lazy or irresponsible—but because they don’t know better.
We’ve made the tools accessible, but not the skill.
It’s like giving everyone a stethoscope and saying, “Okay, monitor your heart.”
Most people can figure out how to put the earpieces in. Some might even hear a sound. But they don’t know what they’re listening for—or what to do with what they hear.
Here’s the thing, though: we don’t give everyone a stethoscope! If someone needs to check their heart, we send them to a doctor.
Even in the age of the Apple Watch—where you can run an EKG from your wrist—you’re still prompted to talk to a professional. The tool doesn’t diagnose. It flags issues. It tells you, “Something might be off. Get this checked.”
Most tools work this way:
Microsoft Word has been underlining your spelling mistakes for you in real-time for decades.
QuickBooks flags errors if your books don’t balance.
Google Sheets pops up with #ERROR when your formulas are broken.
These tools don’t just make creation easy—they build in feedback. They help prevent common mistakes. But somehow… surveys platforms missed the memo.
We’ve made it normal—expected, even—for people to collect survey data without any training, tools, or feedback at all.
And that’s where things get twisted.
We made data collection more accessible—but didn’t put in any guardrails for quality. And look, some survey practices are objectively bad. Missing answer choices? Overlapping answer choices? Complicated language? Those aren’t stylistic preferences. They are mistakes.
Platforms removed the barriers to launching surveys, but neglected to add the guidance to design them well. They made it easy to collect data—and just as easy to get it wrong.
And the worst part? We’ve built a system that makes people think any data is good data. There’s no warning when you mess up. No spell check. No error message. No prompt to stop and ask, “Is this the best way to ask this?”
So of course people trust what they collect. Of course, they use it to make decisions.
They’ve been told that data equals truth—without ever being told how to tell good data from bad.
We’ve made it possible to be completely confident and completely wrong at the same time.
When people say something is “data-driven,” it gives the decision this aura of truth. It makes it sound objective. Unbiased. Credible. And it doesn’t deserve that label if the data was garbage.
I don’t want survey design to be more accessible. I want good survey design to be accessible.
Survey design is a skill. It’s teachable. It’s learnable. And honestly? It’s one most people should probably have.
That’s why I talk about it constantly. That’s why I wrote Ask Better Questions. That’s why I built bisque.
Because someone needs to stand up and call bullshit. Someone needs to say: hey, maybe it’s irresponsible for every platform to have a poll features when people don’t know how to write good questions or interpret their sample.
We need better defaults.
We need a way to make quality the norm—not the exception.
Credibility shouldn’t be this easy to fake. And, the fact that it is, should scare us all way more than AI…
If you’re realizing your last survey might’ve had some issues—don’t panic. That’s what happens when the system gives you tools but no training.
The good news? It doesn’t have to stay that way.
If you want to learn what good survey design actually looks like:
Read Ask Better Questions: It breaks down everything I wish people knew about survey design—without the academic fluff. (non-Amazon version here).
Join the waitlist for bisque: My platform will walk you through common survey mistakes and flag issues before you hit send.
Take my pre-recorded Survey Design Essentials workshop: This workshop will introduce you to practical survey design techniques and include real-time feedback, so you leave with confidence and a clear path forward.
For a limited time, you can work with me 1:1 or get a question makeover.