sales-skills/sales-evelance

Evelance Platform Help

Evelance (evelance.io) is a synthetic-persona (AI) consumer-research and message/creative-testing tool. You describe a target audience, and Evelance matches AI personas (a claimed 326M+ predictive personas across 23 countries; 800M+ on Enterprise) that "read" your messaging, copy, website, app, prototype, or design file and return 12 consumer-psychology scores (attention, trust, value, emotion, action-readiness, …) on a 10-point scale, written per-persona rationale (what they understood, what confused them, "the exact line where readers got lost"), and AI-synthesis reports — usually in ~10–30 minutes, with no recruiting. It runs A/B and competitive tests and is pitched as a fast, cheap PickFu alternative.

Its niche vs the rest of the research/validation cluster:

• Synthetic, not a real panel. The whole pitch is speed without recruiting — AI personas, not real people. That makes it the cheap/fast pre-flight counterpart to real-panel testers (/sales-pickfu consumer, /sales-wynter B2B), and a sibling of synthetic usability (/sales-uxia) — an AI opinion, not observed behavior.
• Message/creative/product testing, not idea validation or usability depth. Evelance tells you whether your words and creative land with a modeled audience — not build-or-not (/sales-idea-validation) and not deep usability metrics on a built flow (/sales-uxia / /sales-uxtweak).
• Self-serve, pay-as-you-go per persona. ~$2.99/persona (volume packs cheaper), a free 5-day trial (10 personas, no card), Enterprise custom. A solo maker can run a test today.

Two things to say almost every time (the double caveat):

• A synthetic score is an AI opinion, not real behavior. Personas are trained on averaged data, so they can't capture individual variation, they inherit bias amplification, and even Evelance says they don't replace real research. Trust relative comparisons (variant A vs B) over absolute scores, and validate the winner with real people. The vendor's 89.78% "thematic-accuracy" figure is a self-reported study (23 humans vs 7 personas) — treat it as best-effort.
• Even real resonance is not demand. "This message scored high / personas said they'd act" is what lands with a reader-model, not a purchase. Keep the winning message + the rationale; take the willingness-to-pay / conversion go/no-go to a real behavior test (smoke test, pre-sale) via /sales-idea-validation.

Step 1 — Gather context

If references/learnings.md exists, read it first for accumulated platform knowledge.

Ask only what you can't infer:

1. What do you want from Evelance?
   • A) Set up a test — pick what you're testing (messaging/copy, website, app, prototype, design file) and the test mode (single, A/B, competitive)
   • B) Describe the audience — the buyer/user you want personas matched to
   • C) Read results — the 12 psychology scores, per-persona written rationale, and the synthesis report
   • D) Budget — personas per test, pay-as-you-go vs packs, what a test costs
   • E) Get data out — export/automation (there is no API/webhooks/MCP)
   • F) Choose — Evelance (synthetic) vs a real panel (/sales-pickfu / /sales-wynter) or synthetic usability (/sales-uxia)
2. Is a decision riding on this, or is it a fast pre-flight? This decides how hard to push the "synthetic is directional; validate the winner with real people / real behavior" message.

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-native research pipeline (Evelance has none — it's UI-only) | /sales-userintuition {question} | | Running a real behavior demand test (smoke-test page, waitlist, pre-sale) after a synthetic test | /sales-idea-validation or /sales-funnel {question} | | Real-panel consumer creative/message split-testing (real people vote + explain) | /sales-pickfu {question} | | Real-panel B2B message/positioning testing (verified professionals in your ICP) | /sales-wynter {question} | | Synthetic usability testing on a built product/prototype (SUS/heatmaps, not message scores) | /sales-uxia {question} | | Analyzing existing NPS/CSAT/VoC/survey data into themes (post-collection) | /sales-customer-feedback or /sales-trill {question} |

When routing, give the exact command: "This is a {domain} question — run: /sales-idea-validation {original question}"

Otherwise, answer Evelance-specific questions using Step 3.

Step 3 — Evelance reference

Read references/platform-guide.md for the full reference — the test-type / automation-surface table (everything is UI-only), the 12 psychology-score model, best-effort per-persona pricing and plan gates, the Test → Audience → Persona-response data model, how to write a clean audience + variant set, and how to read relative scores + written rationale (not absolute numbers).

There is no evelance-api-reference.md — Evelance has no public API/webhooks/MCP (best-effort; confirm at evelance.io). Don't invent endpoints.

Answer using only the relevant section — don't dump the full reference.

Step 4 — Actionable guidance

• Lead with the double caveat: synthetic = AI opinion, and resonance ≠ demand. Whatever the user asks, make explicit that (1) Evelance personas are synthetic — an AI opinion trained on averaged data (bias amplification, no individual variation), so a high score is not real behavior; and (2) even real message resonance is stated preference, not a purchase. Both caveats stack: trust the written per-persona rationale and relative rankings, validate the winner with real people, and take the willingness-to-pay / conversion go/no-go to a real behavior test via /sales-idea-validation.
• Trust relative comparisons over absolute scores. Steer the user to read Evelance as a ranking engine — which variant/message the personas prefer and why — not as a calibrated absolute score. Use the 12 psychology scores to compare A vs B or vs a competitor, and mine the per-persona rationale for the objection/confusion language. Do not report an absolute "5.8/10" as if it were a validated metric.
• Treat automation as UI-only, and don't invent endpoints. When any API/webhook/export/integration comes up, state that Evelance has no public REST API, webhooks, Zapier/Make, or MCP server (best-effort — confirm at evelance.io); data-out is manual export from the dashboard. Don't design a REST/webhook pipeline or guess endpoints; if a programmatic pipeline is a hard requirement, route to /sales-userintuition.
• Size a test by persona count, and call pricing best-effort. Cost scales with how many personas you run (~$2.99/persona pay-as-you-go, cheaper in volume packs; a 10-persona test ≈ $23.90–$29.90). Give a directional figure but present every number as best-effort and point to evelance.io/pricing; note the free 5-day trial (10 personas, no card) for a first run. More personas = tighter averages, not more truth — a synthetic sample doesn't become demand by getting bigger.
• Match the test to the job, and describe the audience well. Recommend messaging/copy testing for "does my headline/value-prop land", A/B for picking between variants, competitive for reading against a rival's message, and website/app/prototype/design-file tests for creative — and reserve Evelance for the fast, cheap synthetic pre-flight, routing a real-panel decision to /sales-pickfu / /sales-wynter and build-or-not to /sales-idea-validation. A vague audience description yields generic personas — describe the buyer/user specifically.

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) — Evelance's pricing (per-persona rate, packs), persona-pool size, score model, test-type lineup, and export options change; verify at evelance.io and evelance.io/pricing.

• Synthetic ≠ real behavior. Personas are an AI opinion trained on averaged data — they inherit bias amplification, can't surface a new individual reaction, and don't replace real research (the vendor agrees). Trust relative rankings + the written rationale; validate the winner with real people.
• Resonance ≠ demand. A high score or a persona "I'd act on this" is what lands with a reader-model, not a purchase. Take the go/no-go to a real behavior test (smoke test, pre-sale) via /sales-idea-validation.
• The 89.78% accuracy figure is a vendor claim. It's a self-reported study (23 real people vs 7 personas) — directional evidence the tool tracks some real reactions, not a guarantee for your audience.
• No public API/webhooks/MCP/Zapier — automation is manual export. Data-out is dashboard export; don't invent endpoints or design a webhook pipeline (that's a /sales-userintuition job). Best-effort — confirm at evelance.io.
• Priced per persona; more personas isn't more truth. ~$2.99/persona PAYG (cheaper in packs); cost scales with persona count. A bigger synthetic sample tightens the average, it doesn't turn an AI opinion into demand.
• It's synthetic message/creative testing, not usability or idea validation. For a built product/ prototype use /sales-uxia (synthetic) or /sales-uxtweak (real); for real-panel message testing use /sales-pickfu (consumer) / /sales-wynter (B2B); for build-or-not use /sales-idea-validation.
• Absolute scores aren't calibrated. Read Evelance as a comparison/ranking tool — an absolute "5.8/10" is not a validated metric; the value is A-vs-B and the per-persona why.

Related skills

• /sales-idea-validation — The tool-agnostic validate-before-building method + the full research/validator tool landscape (use to decide build-or-not; a synthetic "yes" is an AI opinion, not demand). Install: npx skills add sales-skills/sales --skill sales-idea-validation -a claude-code
• /sales-pickfu — PickFu, the real-panel consumer counterpart (15M+ real people vote and explain why, with a real REST API/MCP/CLI). Pick PickFu for a real-panel decision; pick Evelance for a fast/cheap synthetic pre-flight. Install: npx skills add sales-skills/sales --skill sales-pickfu -a claude-code
• /sales-wynter — Wynter, the real-panel B2B message-testing counterpart (80k+ LinkedIn-verified professionals in your ICP). Pick Wynter for real B2B positioning signal; Evelance for a synthetic pre-flight. Install: npx skills add sales-skills/sales --skill sales-wynter -a claude-code
• /sales-uxia — Uxia, the synthetic usability-testing sibling (AI testers on a built design/prototype → SUS/heatmaps) — the right synthetic tool when the question is "is my product usable", not "does my message land". Install: npx skills add sales-skills/sales --skill sales-uxia -a claude-code
• /sales-userintuition — The REST-API/webhook/MCP research peer — the pick when you need a programmatic pipeline Evelance's UI-only surface can't offer. Install: npx skills add sales-skills/sales --skill sales-userintuition -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: "Evelance scored my new headline 8.2/10 — should I launch on that?"

User says: "I ran my homepage headline through Evelance and 10 personas gave it 8.2/10 on trust and value, and most said they'd sign up. Green light?" Skill does: Names the double caveat — the personas are synthetic (an AI opinion trained on averaged data), so 8.2/10 is not real behavior, and even if it were, resonance is not demand. Reframes it as a relative ranking + rationale, not a calibrated score, and routes the user to validate the winner with real people (/sales-pickfu) and a real behavior test (smoke-test/pre-sale) via /sales-idea-validation. Result: The user keeps the winning message + rationale and earns the go/no-go from real behavior, not a synthetic score.

Example 2: "How do I pull Evelance results into my dashboard automatically?" (developer/automation)

User says: "I want a webhook to fire when an Evelance test finishes and POST the psychology scores into our BigQuery. What's the endpoint?" Skill does: States plainly that Evelance has no public REST API, webhooks, Zapier/Make, or MCP server (best-effort — confirm at evelance.io), so there's no endpoint or key to build against, and does not invent one — the only data-out is manual export. Routes an automated completion-to-warehouse pipeline to /sales-userintuition (API + webhooks + MCP). Result: The user stops hunting for an API that doesn't exist and either exports manually or moves the pipeline to an API-native tool.

Example 3: "Evelance or PickFu for testing my Amazon main image?"

User says: "I've got two product main-image options. Should I use Evelance or PickFu to pick the winner?" Skill does: Frames the trade-off — Evelance is synthetic (AI personas, ~10–30 min, an AI opinion) and PickFu is a real consumer panel (15M+ real people vote and explain). Recommends Evelance as a fast/cheap pre-flight to triage options, then a real-panel PickFu poll (/sales-pickfu) for the actual pick, and reminds that even a PickFu winner is preference, not demand — the real proof is click-through/conversion. Result: The user uses synthetic to triage cheaply and a real panel to decide, keeping demand as a separate question.

Troubleshooting

"Can I trust the Evelance score — is it accurate?"

Symptom: The user wants to know whether the 12 psychology scores reflect how real people will react. Cause: Personas are synthetic — LLM models trained on averaged audience data. The 89.78% "thematic-accuracy" is a self-reported study (23 humans vs 7 personas); averaged models can't capture individual variation and inherit bias amplification. Solution: Use the scores as a relative ranking (A vs B vs competitor) and mine the per-persona written rationale — not as a calibrated absolute. Validate the winner with real people (/sales-pickfu / /sales-wynter) and take the go/no-go to a real behavior test (/sales-idea-validation).

"Where's the Evelance API / webhook to export results?"

Symptom: You're looking for an API key and a webhook to fire when a test completes. Cause: Evelance is UI-only — no public REST API, webhooks, Zapier/Make, or MCP server (best-effort; confirm at evelance.io). Data-out is manual export. Solution: Export from the dashboard. For a programmatic warehouse/CRM pipeline, use an API-native tool — route to /sales-userintuition. Don't build against guessed endpoints.

"How much will an Evelance test cost?"

Symptom: The user wants to budget a test. Cause: Evelance bills per persona (pay-as-you-go, cheaper in volume packs), so cost scales with how many personas you run — a 10-persona test is roughly the price of ten personas (best-effort). Solution: Size the test by persona count; start with the free 5-day trial (10 personas, no card) and confirm current rates at evelance.io/pricing. A bigger synthetic sample tightens the average but does not make an AI opinion into demand.