Sensory Panel Planner

How to use it

1. 1Paste the prompt into ChatGPT, Claude or Gemini.
2. 2Fill the bracketed inputs (product, decision, samples, panel resources).
3. 3Answer its clarifying questions about the decision at stake.
4. 4Iterate: ask it to simplify the protocol for an untrained office panel.

See it in action

Panel objective: Determine whether Brewline NA IPA is perceived as more 'craft-IPA bitter' than Competitor A and Competitor B — and if not, which attribute gap explains it. This informs the hop-dosage reformulation decision.

Test format: blind descriptive intensity rating across three samples, 9-point scales. Ranking alone would tell you who wins, not why — you need the why.

Attributes to measure:
1. Bitterness intensity — overall perceived bitterness strength.
2. Bitterness quality — harsh/metallic vs rounded.
3. Hop aroma — citrus/pine character on the nose.
4. Body — thin vs full mouthfeel.
5. Sweetness — residual malt sweetness.
6. Aftertaste length — how long bitterness lingers.

Sample plan: 60 ml pours at 6-8 C, 3-digit blind codes, serving order rotated across panelists, water and plain crackers between samples, 2-minute gaps.

Panelist screening: exclude anyone who worked on the recipe; screen for basic bitterness sensitivity. Twelve staff panelists gives a directional read, not a claim.

Reading results: if Brewline trails on intensity but matches quality, adjust hop dosage; if quality lags, revisit hop variety or process.

Want me to turn this into a printable score sheet and serving schedule?

The prompt

Here's the start of the prompt. Download the free bundle for the full, ready-to-paste version — plus the installable Claude Skill and Custom-GPT instructions.

# Role
You are a senior sensory scientist who has designed hundreds of panels for food and beverage manufacturers. You refuse to run a tasting until the decision it informs is written down, and you design protocols that non-specialist teams can execute credibly with the people and time they actually have.

# Context I'll provide
- Product and category: [PRODUCT]
- The decision this panel informs: [DECISION e.g. reformulation choice, competitive benchmark, launch go/no-go]
- Samples being compared: [SAMPLES e.g. our prototype vs two competitors]
- Panel resources: [RESOURCES e.g. number of panelists, staff or consumers, session time]
- Known sensory concerns (optional): [CONCERNS]
- Facility or budget constraints (optional): [CONSTRAINTS]

# Your task

Frequently asked questions

What is a sensory panel in food and beverage development?

A sensory panel is a structured tasting in which screened panelists evaluate products against defined attributes and scales under controlled conditions — blind samples, randomized serving order, palate cleansing. Teams use panels to compare recipes, benchmark competitors, or approve changes. The discipline matters more than the size: a well-run panel of twelve produces decisions, while a casual tasting of fifty produces opinions.

Does this prompt require a specific AI model?

No. Any capable chat model — ChatGPT, Claude, or Google Gemini — runs it as written, because it relies on structure rather than model-specific features. Many teams save it as a Custom GPT, a Claude Skill, or reusable instructions so every panel across the company is scoped the same way and results stay comparable over time.

Can I run this with untrained office panelists?

Yes — that is the most common use. The skill sizes the method to your real resources: with untrained staff it chooses simpler formats, adds plain-language attribute definitions, and states clearly that results are directional reads rather than statistically defensible claims. When your decision genuinely needs trained-panel rigor or consumer-scale data, it will tell you to bring in a research partner.

How is this different from a consumer taste test?

A sensory panel measures product attributes and differences using screened evaluators; a consumer taste test measures liking and preference among target shoppers. This skill designs the first and tells you when you need the second — typically once the panel shows you are close enough to parity or superiority that consumer preference becomes the deciding question.