Why 61% of card sorts get zero responses

Here's a number we're not proud of: 61% of card sorting studies created on ValidateThat never collect a single participant response.

Not "didn't reach statistical significance." Not "only got 5 responses." Zero. The study was created, published, and nobody took it.

We could hide this number. Instead, we dug into it — because understanding why studies fail tells you how to make yours succeed.

It's not a study design problem

The first thing we checked: are these poorly designed studies? Bad card labels? Too many items? Confusing instructions?

No. The distribution of card counts, study types, and configuration quality is nearly identical between zero-response studies and studies that collected 10+ responses. A well-designed study with 15 beautifully written cards will collect exactly zero responses if nobody sees the link.

The bottleneck is distribution, not design.

The five failure patterns

After analyzing the zero-response cohort, five distinct patterns emerged:

1. The "just exploring" study (≈25% of zeros)

These are template studies — created when a user signs up and gets a demo project seeded. The cards match our templates exactly. No welcome message. No recruitment configured. The user was exploring the tool, not running research.

This isn't a failure. It's exploration behavior. These users may come back when they have a real project, or they may have decided card sorting isn't the right method for them.

2. The "built it but didn't share it" study (≈30% of zeros)

Custom cards. Welcome message written. Study configured properly. Then... nothing. No shares, no recruitment, no responses. The study is sitting at a URL that nobody knows about.

This is the most preventable failure. The creator did the hard part (designing the study) but stalled at the distribution step. Common reasons:

• Didn't know where to share the link
• Felt awkward asking people to spend 5 minutes on a card sort
• Assumed participants would find it organically (they won't)
• Got busy with other work and forgot about it

Fix: Have a distribution plan before you publish. Even a simple list: "I'll post in our Slack, email 10 colleagues, and share on Twitter."

3. The "wrong audience" study (≈15% of zeros)

Study was shared — we can see referral data — but the link went to an audience that wasn't motivated to participate. Posting a B2B SaaS card sort to a general subreddit. Sharing an internal navigation study on Twitter where none of your followers are users of the product.

Fix: Match your recruitment channel to your target user. If you're testing e-commerce navigation, recruit shoppers (not designers). If you're testing a developer tool's docs structure, recruit developers (not product managers).

4. The "friction killed it" study (≈10% of zeros)

These studies technically received clicks (the study page loaded) but zero completed responses. Participants opened the study and bounced. Common causes:

• 40+ cards (too long, participants bail)
• No context — the welcome message says "sort these cards" with no explanation of why
• Mobile participants hitting a desktop-optimized study
• Unclear instructions

Fix: Keep it under 25 cards. Write a 2-sentence welcome message explaining who you are and what you'll do with the results. Test the study yourself on mobile before publishing.

5. The "abandoned project" study (≈20% of zeros)

The study was part of a project that got deprioritized. The creator's account shows no login within 7 days of creation. The study was a victim of shifting priorities, not poor execution.

This is normal. Research plans change. Not every study needs to be completed.

What the 39% did differently

Studies that collected responses share consistent traits:

They recruited within 24 hours of publishing

The most reliable predictor of getting responses is taking action within the first day. Studies where the first share happened within 24 hours of creation had an 82% chance of collecting at least one response. Studies where no action was taken in the first 48 hours had a 94% chance of getting zero responses forever.

Momentum matters. Don't let the study sit.

They used at least two distribution channels

Studies with responses typically appeared in 2–3 places: a direct link shared via email or Slack, plus one public channel (Twitter, a community forum, a Prolific recruitment). Single-channel studies had lower response rates across the board.

They set realistic expectations

Studies aiming for 10–15 responses (not 50) were more likely to hit their target. Lower goals meant the creator kept at it long enough for responses to trickle in. Overly ambitious targets led to discouragement and abandonment.

They used paid recruitment when their audience was small

Creators without an existing audience who used Prolific ($3.50/response) averaged 24 responses. Creators without an audience who relied purely on organic sharing averaged 2.1 responses.

If you don't have an audience, budget for participants. $50 on Prolific gets you a complete, statistically meaningful dataset. That's less than you'd spend on a team lunch.

The activation timeline

Here's the response accumulation curve for successful studies:

• Hour 1–4: ~30% of total responses (fast responders, Prolific participants)
• Hour 4–24: ~40% of total responses (email opens, timezone spread)
• Day 2–5: ~25% of total responses (reminders, secondary shares)
• Day 5+: ~5% of total responses (long tail, forwarded links)

Most studies that will succeed show signal within the first 24 hours. If you've shared your study and have zero responses after 48 hours, it's time to change tactics — not wait longer.

A practical recruitment playbook

Based on what works in our data:

Free tactics (expect 5–15 responses each)

1. Direct message 10 people individually. "Hey, I'm running a quick 5-minute card sort for [thing]. Would you mind giving it a go? Here's the link." Personal asks convert at 40–60%.

2. Post in 2–3 relevant communities. Slack groups, Discord servers, subreddits where your target users hang out. Frame it as helping improve something they care about, not as a favor to you.

3. Email your existing users or newsletter. If you have any existing audience — even 50 people — email them directly with the study link.

4. Swap with another researcher. UX communities have "participant swap" threads. You take their study, they take yours.

Paid tactics (expect 15–30 responses)

5. Prolific ($3.50/response). Pre-screened participants from your target demographic. Set filters for age, profession, location, etc. Responses typically arrive within 2–4 hours. How to use Prolific for UX research →

6. ValidateThat's built-in recruitment. Launch a Prolific study directly from the platform with demographic targeting. One click, responses within hours.

The math on participant recruitment

Here's the uncomfortable truth about "free" card sort responses:

For most founders and PMs without an audience, the fastest path to a complete study is: 5 personal DMs + a Prolific batch of 10. Total cost: $35. Total time: 15 minutes of recruitment work. Total responses: ~15.

That's your minimum viable study.

What we're doing about it

We're not just publishing this data and walking away. Based on these findings, we've been building:

1. Better onboarding. New users get a recruitment checklist as part of the study creation flow, not after.
2. Recruitment nudges. If a study has been live for 24 hours with zero responses, we prompt the creator with specific next steps.
3. Built-in Prolific recruitment. One-click participant recruitment with demographic targeting, directly from the study dashboard.
4. Done-for-you validation. For people who don't want to handle recruitment at all — we'll run the entire study for you and deliver a full report in 2–3 days.

Data from 491 card sorting studies on ValidateThat, January 2024–April 2026. Anonymized and aggregated. If you're planning your first card sort and want to avoid the zero-response trap, start here or let us handle it for $99.