Day 9 — a review step that protects specialists from bad surveys before they're published
Designed two things today: the actual question builder (single
choice, long text, image upload, video response, plus screener
logic to filter respondents gracefully), and something I hadn't
planned on — an optional peer-review step before a survey publishes.
A company can send their draft to a colleague, who can approve it or
request changes with specific feedback ("Q3 wording is unclear,
screener should come before Q2"), all before a single specialist ever
sees it.
The reasoning: a confusing or poorly-screened survey wastes a
specialist's time even if the payout is fair. Catching that before
publish protects the supply side without the specialist ever knowing
there was a problem to catch.
Still zero app code, but this feels like the kind of detail that
either gets designed on purpose or never gets built at all.
The peer-review step is an interesting marketplace quality mechanism.
What would convince you that survey quality before publication is the biggest bottleneck, rather than other issues like specialist acquisition, demand generation, or payout incentives?
Honestly - nothing yet. I designed the review step because it felt
like a natural extension of the "protect specialist time" principle
I've been building around, not because I have evidence it's the
biggest bottleneck. I don't have real users yet, so I don't actually
know if survey quality, specialist acquisition, demand generation, or
payout incentives will turn out to be the constraint that matters
most.
If I'm honest with myself, specialist acquisition and demand
generation are probably bigger open risks right now than survey
quality - a badly-worded survey is a real but fixable problem, while
"nobody signs up" or "no company wants to pay for this" are existential
ones. I think I've been designing quality-of-experience features
partly because they're more fun and controllable to design than
solving acquisition, which is a fair thing to call out in myself.
What would actually convince me either way is getting to real users
faster - even a rough version with 10-20 specialists and 1-2 real
company surveys would tell me more in a week than another month of
Figma will.
That’s helpful context. I appreciate the honesty around what’s known versus what’s still a hypothesis.
I’d like to continue the conversation outside the thread. What’s the best email to reach you on?