LinkedIn Poll Validation

What is LinkedIn poll validation?

LinkedIn poll validation means posting a short poll on LinkedIn about a problem you want to solve, then using the votes to find and talk to the people who have that problem. The poll itself is cheap evidence. Its real job is to raise hands: on LinkedIn the author of a poll can see who voted and how each person voted, so every vote is a possible interview.

A LinkedIn poll has a question of up to 140 characters and 2 to 4 options of up to 30 characters each, and it runs for 1 day, 3 days, 1 week or 2 weeks. Plan the wording before you post, because LinkedIn’s own documentation says poll content can’t be updated once it’s live.

When should you use it?

It fits when:

• Your customers are professionals: recruiters, HR managers, accountants, marketers, consultants, founders, lawyers, teachers.
• You have at least a few hundred connections in or near that field, or a colleague with a strong network who will share the poll.
• You want to rank 3 versions of a problem or 3 current workarounds and find people to interview this week.

Skip it when:

• Your customers are consumers who don’t use LinkedIn for personal topics, like parents choosing a toy.
• The question touches health, politics or other sensitive data. LinkedIn’s rules don’t allow poll authors to ask for those, which femtech and health founders need to keep in mind.

How do you do it, step by step?

1. Ask about the past. “Would you use an app that…?” invites polite yeses. Ask what people DID last time: the last time X happened, what did you do? Past behaviour is evidence, while a promise about the future costs nothing to make.

2. Write a neutral question. Name the audience at the start so the right people stop scrolling (“Agency recruiters:” or “Freelance designers:”). Leave your solution out of it and avoid loaded words like “painful” or “broken” that push people toward one answer.

3. Use all 4 options. Three options describe real behaviour (the main workarounds you heard about), and the fourth is always Something else (comment). That last option pulls the most useful answers into the comments, in people’s own words.

4. Add 2 or 3 lines of context. In the post text above the poll, say why you’re asking and that this is research with nothing for sale. Keep it human: one sentence about what you noticed, one asking people to vote and comment.

5. Pick 1 week and post at a working hour. One week gives the post time to travel through your network. Post on a weekday morning in your audience’s time zone, then reply to every comment within the first hours to keep it visible.

6. Message the voters. After 2 or 3 days, open the vote list and message people who picked a workaround that signals frustration. Template: Hi [name], thanks for voting on my poll about reference checks. You picked “[option]”. Would you tell me more in a 15 minute call this week? I’m researching the problem and not selling anything. Send 20 to 30 messages and log replies in a sheet.

7. Record the results with context. When the poll closes, write down total votes, the split, how many were in your target group (check job titles in the vote list), comments and how many calls you booked. That last number counts most.

What does a worked example look like?

Say Olga is a former recruiter in Warsaw who is thinking about a fast reference-check service for small recruitment agencies. This is a hypothetical case with invented numbers.

Her post: Small agency recruiters, I’m researching how reference checks work in teams of 2 to 10. Vote and tell me in the comments what I’m missing.

Her poll question: Agency recruiters: the last time a reference check dragged on, what did you do?

Her 4 options:

• Chased referees by phone
• Placed the candidate anyway
• Asked the client to do it
• Something else (comment)

After a week she has 64 votes. 41 come from recruiters, the rest from HR people and friends. Among the recruiters, 22 chased by phone, 11 placed the candidate anyway, 5 handed it to the client and 3 chose something else. The comment thread surprises her: four people describe referees who never answer at all.

She messages the 33 recruiters who chased or skipped, gets 14 replies and books 8 calls. The poll told her where the frustration sits. The calls tell her whether agencies would pay, which a poll can never do.

Which free tools help?

What mistakes should you avoid?

• Asking “would you buy”. Hypothetical yes votes cost nothing and prove nothing. Ask about past behaviour.
• Pitching in the poll. Mention your product and the poll turns into an ad, and both reach and honesty drop.
• Counting every vote. Friends and former colleagues vote to support you. Filter by job title before you read the split.
• Leaving out “Something else”. Without it, people who don’t fit your options either skip or pick a wrong answer, and you lose the most surprising data.
• Stopping at the chart. A poll without follow-up messages is a vanity metric. The method only works when you talk to voters.

How do you know it worked?

A poll can show that a problem exists among people who bothered to vote, which workaround is most common, and who to talk to. It can’t show market size, willingness to pay, or anything about people outside your network.

Our rule of thumb for a go: at least 30 votes from your target group, at least a third of them choosing a workaround that costs time or money, and at least 5 calls booked from your messages. Under 15 relevant votes, or zero calls, means your network doesn’t reach these people yet, so try LinkedIn outreach or a community where they gather.

What should you do next?

Run the calls as problem interviews, then write a proper survey if you need numbers from outside your network. Keep posting about what you learn with LinkedIn founder posts, and if you run out of angles, a free LinkedIn post idea generator gives you a weekly plan. The F/MS startup game helps you plan polls and interviews with an AI co-founder. LinkedIn lists the poll limits in its Poll API documentation.

FAQ

Can you see who voted in a LinkedIn poll?

Yes, if you created the poll. The author can see who voted and how each person voted. Other people only see the overall percentages once they’ve voted.

How many options can a LinkedIn poll have?

At least 2 and at most 4, each up to 30 characters. The question can be up to 140 characters, so write short and test your options before posting.

How long should a LinkedIn poll run?

One week for validation. One day is too short for a network to see it, and two weeks lets the post go stale before you message voters.

Updated 26 October 2026 · Written by the Fe/male Switch team

Mark this method as done