Poll the People (pollthepeople.app) platform help — real-panel consumer creative/message/UX testing with ChatGPT-powered analysis: run an A/B, head-to-head, preference, ranked, or open-ended test past a 500,000+ human panel (100+ targeting options, US default) → vote % plus written rationale that ChatGPT synthesizes into an executive summary + themes, in minutes. A cheaper, UI-only PickFu alternative — no public API, webhooks, Zapier, or MCP. Use when setting up a Poll the People test, picking a test type, writing non-leading questions, split-testing a logo/ad/landing page/Google headline/brand name/design, targeting a consumer audience, budgeting per-response cost or Lite/Plus/Premium plans, reading the AI summary vs raw responses, or exporting without an API. Do NOT use for the validate-before-building method or comparing research tools (use /sales-idea-validation), an API/automation pipeline or larger international panel (use /sales-pickfu), or B2B-panel message testing (use /sales-wynter).
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
Poll the People (pollthepeople.app) platform help — real-panel consumer creative/message/UX testing with ChatGPT-powered analysis: run an A/B, head-to-head, preference, ranked, or open-ended test past a 500,000+ human panel (100+ targeting options, US default) → vote % plus written rationale that ChatGPT synthesizes into an executive summary + themes, in minutes. A cheaper, UI-only PickFu alternative — no public API, webhooks, Zapier, or MCP. Use when setting up a Poll the People test, picking a test type, writing non-leading questions, split-testing a logo/ad/landing page/Google headline/brand name/design, targeting a consumer audience, budgeting per-response cost or Lite/Plus/Premium plans, reading the AI summary vs raw responses, or exporting without an API. Do NOT use for the validate-before-building method or comparing research tools (use /sales-idea-validation), an API/automation pipeline or larger international panel (use /sales-pickfu), or B2B-panel message testing (use /sales-wynter).
argument-hint
[describe what you need help with in Poll the People]
license
MIT
version
1.0.0
tags
["sales","market-research","platform"]
Poll the People Platform Help
Poll the People (pollthepeople.app) is a real-panel consumer creative/message-testing tool with
ChatGPT-powered analysis. You create a test — A/B / head-to-head, preference, ranked, or
open-ended — pick a consumer audience from a 500,000+ human panel (100+ segmentation options,
US-based by default), and get back quantified preference (vote %) plus a written explanation
from every respondent that ChatGPT synthesizes into an executive summary + themes — usually in
minutes to an hour. Common jobs: logos, ads, landing pages, Google headlines, brand/product names,
Figma designs, videos, packaging, and copy — plus 20+ ready templates.
Its niche vs the rest of the research/validation cluster:
Real panel, consumer, self-serve — the direct PickFu competitor. Poll the People and
/sales-pickfu are the two real-panel consumer message/creative testers. Poll the People's
bet is a cheaper subscription entry (free Lite tier + $1/response) and ChatGPT-native
analysis front-and-center; PickFu's edge is a 15M+ international panel and a real API/MCP/CLI.
It is not synthetic — unlike Articos/Evelance/Societies it uses actual people.
Consumer, not B2B. For hard-to-reach B2B professionals (seniority/industry/company size),
/sales-wynter is the better instrument; Poll the People's panel is broad consumer, US-skewed.
Creative/message testing, not usability or idea validation. It tells you which option wins and
why with real people — not whether a built product is usable (/sales-lyssna, /sales-uxtweak)
and not build-or-not (/sales-idea-validation).
UI-only. No public REST API, webhooks, Zapier/Make, or MCP — for a programmatic pipeline PickFu
(API) is the pick; Poll the People is manual export.
Two things to say almost every time:
A test measures preference (stated), not demand. "Option B won / people said they'd buy" is
what a reader prefers when asked — not a purchase. Keep the winning option and the verbatim
responses; take the willingness-to-pay / conversion go/no-go to a real behavior test (smoke
test, pre-sale) via .
/sales-idea-validation
Read the raw responses, not just the ChatGPT summary. The AI executive summary is a genuine
time-saver on real answers, but it smooths over nuance and outliers. Trust the verbatim
responses + relative ranking over the polished summary or absolute scores.
Step 1 — Gather context
If references/learnings.md exists, read it first for accumulated platform knowledge.
Ask only what you can't infer:
What do you want from Poll the People?
A) Set up a test — pick a type (A/B / head-to-head / preference / ranked / open-ended) and
write neutral questions
B) Target an audience — build a consumer panel by demographics/segmentation (US default;
advanced targeting is +$0.50/criteria)
C) Read results — vote %, written rationale, and the ChatGPT executive summary / themes
D) Budget — per-response cost, Lite vs Plus vs Premium, response caps per plan
E) Get data out — export/automation (there is no API/webhooks/Zapier/MCP)
F) Choose — Poll the People vs PickFu (API + bigger panel) / Wynter (B2B) / synthetic simulators
Consumer audience, or B2B/professional? If B2B, the panel may be thin — consider /sales-wynter.
Skip-ahead: if the user wants the validate-before-building method or a cross-tool comparison,
that's a /sales-idea-validation question — route in Step 2.
Step 2 — Route or answer directly
If the user's question is about…
Route to
The validate-before-building method, or comparing research/validation tools across the market
/sales-idea-validation {question}
A REST API / webhook / MCP / CLI automation pipeline, or a larger international consumer panel
/sales-pickfu {question}
Running a real behavior demand test (smoke-test page, waitlist, pre-sale) after a test
/sales-idea-validation or /sales-funnel{question}
B2B message/positioning testing on a professional panel (seniority/industry/company size)
/sales-wynter {question}
Synthetic-audience message simulation (AI personas instead of a real panel)
/sales-evelance or /sales-articos{question}
Usability / prototype / IA testing on a built product or design (not preference testing)
/sales-lyssna or /sales-uxtweak{question}
When routing, give the exact command: "This is a {domain} question — run: /sales-idea-validation {original question}"
Otherwise, answer Poll-the-People-specific questions using Step 3.
Step 3 — Poll the People reference
Read references/platform-guide.md for the full reference — the test-type table, best-effort
per-response + plan pricing (Lite/Plus/Premium/Enterprise) and response caps, the Test → Question →
Response data model, audience targeting and the +$0.50/criteria advanced-targeting cost, how the
ChatGPT analysis (executive summary, themes, scoring) works and how to read it vs the raw responses,
and the UI-only automation surface (no API — manual export).
There is no pollthepeople-api-reference.md — Poll the People has no public API/webhooks/MCP
(best-effort; confirm at pollthepeople.app). Don't invent endpoints.
Answer using only the relevant section — don't dump the full reference.
Step 4 — Actionable guidance
Say the caveat: a test is preference, not demand. Make explicit that a winning option (or "I'd
buy this") is what a reader prefers when asked — a real signal, but stated preference, not
observed demand or a purchase. Point the user to the verbatim responses (the "why", objections,
confusion) and relative ranking over the top-line %, then take the willingness-to-pay / conversion
go/no-go to a real behavior test (smoke test, pre-sale) via /sales-idea-validation.
Read the raw responses, not just the ChatGPT summary. The AI executive summary + themes save
time, but the summary is a synthesis of real answers that smooths over nuance, outliers, and
dissent — tell the user to open the verbatim responses, especially those disagreeing with the
winner, and treat the summary as a first pass, not the verdict.
Insist on neutral, goal-based questions. Have the user write non-leading questions (never "do
you like this?" / "isn't X better?") and let the panel explain why — framing bias is the #1 cause of
misleading results. Match audience segmentation to the real buyer; shallow answers are a
question/audience-design fix, and bot/fraud detection auto-rejects poor responses (unbilled).
Size a test by responses; call pricing best-effort. Billing is ~$1/response on top of a plan:
Lite ($0/mo, 3 tests, ≤250), Plus ($50/mo, 10 tests, ≤500), Premium ($100/mo, unlimited
tests, ≤2,000, priority support), Enterprise (custom, ~$10k+/mo); advanced targeting
+$0.50/criteria, US geo free. Present every number as best-effort — point to pollthepeople.app/pricing.
Treat automation as UI-only; don't invent endpoints. On any API/webhook/export question, state
Poll the People has no public REST API, webhooks, Zapier/Make, or MCP (best-effort — confirm at
pollthepeople.app); data-out is a manual dashboard export. Don't design a pipeline or guess
endpoints; for a programmatic pipeline route to /sales-pickfu (real API/MCP/CLI).
Choose Poll the People vs PickFu, and match the panel. Say the split: its edge is a cheaper
subscription entry (free Lite) + ChatGPT-native analysis, PickFu's a 15M+ international panel +
real API/MCP/CLI. It's consumer, US-default — route B2B to /sales-wynter, API/international to /sales-pickfu.
If you discover a gotcha or tip not in references/learnings.md, append it there with today's date.
Gotchas
Best-effort from research (2026-07) — Poll the People's pricing (per-response cost, Lite/Plus/Premium
rates + response caps, advanced-targeting fee), panel size/composition, test-type lineup, ChatGPT
analysis features, and lack of an API change; verify at pollthepeople.app and pollthepeople.app/pricing.
Preference ≠ demand. A winning option or "I'd buy this" is what a reader prefers when asked,
not a purchase. The verbatim responses (the "why") and relative ranking are the value — take the
go/no-go to a real behavior test.
The ChatGPT summary smooths over nuance — read the raw responses. The AI executive summary +
themes save time on real answers, but they hide outliers and dissent. Open the verbatim responses
(especially the ones disagreeing with the winner); treat the summary as a first pass, not the verdict.
Leading questions poison results. Vague or leading wording ("do you like this?", "isn't X
better?") is the top cause of misleading data. Write neutral, goal-based questions; the panel
explains why automatically. Shallow answers are a question/audience-design fix, not a dead panel.
Priced per response on top of a plan; caps scale with tier.$1/response, plus Lite ($0, ≤250)
/ Plus ($50/mo, ≤500) / Premium (~$100/mo, ≤2,000) / Enterprise (custom); advanced targeting
+$0.50/criteria, US geo free. Size it before launching; treat all figures as best-effort.
No public API/webhooks/MCP/Zapier — automation is manual export. Don't invent endpoints or design
a webhook pipeline; for a programmatic pipeline use /sales-pickfu (real API/MCP/CLI). Confirm at pollthepeople.app.
Consumer, US-skewed panel. Route hard-to-reach B2B/professional audiences to /sales-wynter
and a larger international panel or an API to /sales-pickfu; verify the audience builds out first.
It's real-panel testing, not synthetic and not usability. For a fast/cheap synthetic pre-flight
use /sales-evelance / /sales-articos; for a built product use /sales-lyssna / /sales-uxtweak;
for build-or-not use /sales-idea-validation.
Related skills
/sales-pickfu — The direct real-panel consumer competitor: same job (vote % + written rationale in minutes), but a 15M+ international panel and a real REST API / MCP / CLI. Pick PickFu for automation or a bigger/international panel; Poll the People for a cheaper US-consumer entry with built-in ChatGPT analysis. Install: npx skills add sales-skills/sales --skill sales-pickfu -a claude-code
/sales-wynter — The real-panel B2B message-testing counterpart (80k+ LinkedIn-verified professionals in your ICP, scores + quotes, UI-only/no API). Pick Wynter for hard-to-reach B2B positioning; Poll the People for consumer creative/message testing. Install: npx skills add sales-skills/sales --skill sales-wynter -a claude-code
/sales-evelance — The synthetic consumer message/creative counterpart (AI personas score copy/creative on 12 psychology metrics, UI-only/no API) — a fast/cheap pre-flight to triage options before deciding with Poll the People's real panel. Install: npx skills add sales-skills/sales --skill sales-evelance -a claude-code
/sales-articos — The synthetic user-research cousin (bias-diverse AI personas run interviews / concept / A/B tests) — synthetic, not a real panel; use it to pre-flight, then decide with real people. Install: npx skills add sales-skills/sales --skill sales-articos -a claude-code
/sales-idea-validation — The tool-agnostic validate-before-building method + the full research/validator tool landscape (use to decide build-or-not; a test "win" is preference, not demand). Install: npx skills add sales-skills/sales --skill sales-idea-validation -a claude-code
/sales-lyssna — Self-serve usability/IA testing on a built artifact (five-second/first-click/preference, card sort, tree test) — the right tool when the question is "is my product usable", not "which option wins". Install: npx skills add sales-skills/sales --skill sales-lyssna -a claude-code
/sales-do — Not sure which skill to use? The router matches any sales objective to the right skill. Install: npx skills add sales-skills/sales --skill sales-do -a claude-code
Examples
Example 1: "My logo B won the Poll the People test and the AI summary loved it — ship it?"
User says: "I ran an A/B of two logos on Poll the People, B won 61% and the ChatGPT summary said it's more trustworthy and modern. Good enough to make it my brand and expect more signups?"
Skill does: Credits the result as a real signal — actual consumers preferred B and explained
why. Then draws two lines: (1) that's preference, not demand — a preferred logo is not a proven
conversion lift; and (2) the ChatGPT summary smooths over nuance — directs the user to open the
verbatim responses, especially the minority who preferred A, for the real payload. Suggests using
those to sharpen the logo and confirming any lift with a real behavior test (A/B the live
site / smoke test) via /sales-idea-validation. Notes per-response pricing is best-effort.
Example 2: "How do I create a test and pull results into my dashboard via the Poll the People API?" (developer/automation)
User says: "I want to script Poll the People — create an A/B test, wait for it to finish, and POST the vote % and responses into our internal dashboard. What's the endpoint and how do I authenticate?"
Skill does: States plainly that Poll the People has no public REST API, webhooks, Zapier/Make,
or MCP server (best-effort — confirm at pollthepeople.app), so there's no endpoint or key to build
against, and does not invent one — the only data-out is a manual dashboard export. Routes an
automated create-and-collect pipeline to /sales-pickfu (real REST API base https://api.pickfu.com/v1,
MCP server, @pickfu/cli), which does the same consumer-panel job programmatically.
Example 3: "My Poll the People results feel random / responses are shallow — is the panel bad?"
User says: "Half my respondents just say 'looks good' and the votes swing every time I run it. Is the panel junk?"
Skill does: Explains it's usually a question/audience-design problem, not a dead panel: vague or
leading questions invite low-effort answers, and a too-broad audience adds noise. Prescribes
neutral, goal-based questions, tighter audience segmentation to match the real buyer, and a
larger sample size for stability — and notes the platform's bot/fraud detection auto-rejects
poor-quality responses (you aren't charged for rejected ones). For B2B targets the consumer panel may be
thin — consider /sales-wynter; for a larger international panel, /sales-pickfu.
Troubleshooting
"Where's the Poll the People API / webhook to fire when a test finishes?"
Symptom: You're looking for an API key and a webhook to trigger a pipeline when a test completes.
Cause: Poll the People is UI-only — no public REST API, webhooks, Zapier/Make, or MCP server (best-effort; confirm at pollthepeople.app). Data-out is a manual dashboard export.
Solution: Export results from the dashboard. For a programmatic create-and-collect pipeline, use an API-native consumer-panel tool — route to /sales-pickfu (REST API + MCP + CLI). Don't build against guessed endpoints.
"My Poll the People results swing or feel shallow — is the panel bad?"
Symptom: Votes change between runs, or respondents give one-line generic answers.
Cause: Usually a question/audience-design problem — vague/leading questions and too-broad targeting, not a dead panel.
Solution: Write neutral, goal-based questions (the panel explains "why" automatically), tighten audience segmentation to the real buyer, and increase sample size for stability. The platform's bot/fraud detection auto-rejects poor-quality responses (unbilled). For B2B targets the consumer panel may be thin — consider /sales-wynter.
"Can I trust the ChatGPT summary, or do I need to read every response?"
Symptom: The AI executive summary gives a clean verdict and you want to act on it directly.
Cause: The summary is a synthesis of real responses — a genuine time-saver, but it smooths over outliers, dissent, and nuance, and it's a synthesis, not a vote count.
Solution: Use the summary as a first pass, then open the verbatim responses — especially the minority disagreeing with the winner — and trust the relative ranking + the raw "why" over the polished summary or absolute scores. Remember a winner is preference, not demand — take the go/no-go to a real behavior test via /sales-idea-validation.
