Enabling Innovation Getting Started With Sensory Evaluation - - PowerPoint PPT Presentation

Enabling Innovation Getting Started With Sensory Evaluation

Introduction

I want to do some sensory….

What do you want to find out ? What will you use data for? What actions will you take in response to data?

Introduction

Taste Sound Trigeminal sensations Aromas Sight Mouthfeel &

SENSOR SENSORY Y RESP RESPON ONSES SES SUBJECTIVE Opinions Preferences Expectations OBJECTIVE Differences /similarities Descriptions Intensity

Goods in Market awareness / competitor analysis CONSUMERS Existing Product Development (EPD) Manufacturing process Quality Monitoring (QA / QC) New Product Development (NPD) Storage Sales Purchasing Shelf life determination FACTORY CONSUMERS Preference Tests Descriptive Test Benchmarking Test Discrimination Tests Descriptive Tests Discrimination /Descriptive Tests

Raw material spec: Sensory parameters Product Sensory spec

Focus group Preference Tests Marketing Descriptive Test Process Control FACTORY CONSUMERS

Product Sensory spec

Acceptance Tests Discrimination Tests

Customer Sensory spec Label descriptions /advertising

Difference from control tests

Can people notice the change made?

CONSUMERS Goods in Market awareness / competitor analysis CONSUMERS Existing Product Development (EPD) Manufacturing process Quality Monitoring (QA / QC) New Product Development (NPD) Storage Sales Purchasing Shelf life determination FACTORY CONSUMERS Preference Tests Descriptive Test Benchmarking Test Discrimination Tests

Raw material spec: Sensory parameters Product Sensory spec

Focus group Preference Tests Marketing Descriptive Test Process Control FACTORY

Product Sensory spec

Acceptance Tests Discrimination Tests

Customer Sensory spec Label descriptions /advertising What are consumer expectations? How does my product taste compared to the competition? Which does the consumer prefer? Does the consumer like this ? And by how much ? Which does the consumer prefer? Can people notice the change made?

Discrimination /Descriptive Tests

How long before it tastes different? Is it consistent to last batch?

Descriptive Tests

How is it different? Informs root cause analysis.

Difference from control tests

Is this adequate quality? Hold / release decisions. How is my product effected by the supply chain? ‘Buy backs’ How to use sensory as a selling point ?

Research & Controlled Studies

Sensory evaluation for SMEs – market awareness Benchmarking posh chocolates - Christmas Market:

Sensory evaluation for SMEs – sorting based on sensory profile Cider “Napping” – Dad’s Dining Room Table

Sensory evaluation for SMEs – looking for a difference Discrimination Test, Apple Juice - Bramley Apple Festival Southwell Minister

Sensory evaluation for SMEs – checking there’s no difference Triangle Test, Botanical Extracts – At a grower / supplier’s farm office, Hungary

Sensory evaluation for SMEs – Descriptive Profile & Ranking Test

Part 1: Beer Profiling – Pub in Derby (local beer drinkers) Part 2: Beer Ranking – My Dining Room Table!

Practical – Basic Taste Recognition You will taste solutions of the basic tastes using store cupboard ingredients. Do not contain:

• any of the official allergens as listed in 1169/2011
• alcohol

Vegan and vegetarian-friendly. IF THERE IS ANYTHING ELSE YOU DON’T WANT TO CONSUME TELL ME NOW …

Basic Tastes - Recognition Test

Answers

658 705 319 548 536 310

Discrimination Tests – THE TRIANGLE TEST Scenario: Artisan Elderflower drink Production Manager: “I can make an efficiency saving by changing the process”. Owner: “Great, but won’t that affect the flavour of the drink…. ?” Technical Manager: “The difference from trial runs is certainly not that huge…”

Golden Rules for Sensory

PEOPLE:

• 1. Use a group of people to make sensory assessments.
• 2. Be aware of individual differences in detection / perception.
• 3. Avoid coffee, mouth wash and smoking 2 hours before assessment.
• 4. Use palate cleansers:

(and leave time to palate cleanse).

• 6. Don’t do too many

samples / have breaks. 7. Due diligence – food safety.

Samples type tested Palate cleanser

Most products Products with strong flavours or tastes that might persist e.g.: beer bitterness / spicy food / artificial sweeteners Still, unchlorinated water (not mineral water). + Plain crackers/ water biscuits. Unflavoured, no salt. For spicy food – milk. Fatty /creamy / oily products Slice bland apple (e.g: Gala) Products high in astringency e.g.: tea (other products high in polyphenols – cider, rhubarb, cranberry juice) 0.1% pectin solution Piece of melon

Golden Rules for Sensory

SAMPLES

• 4. CONSISTENT – size / shape / format / sample container / temperature
• 5. REPRESENTATIVE of batch / product

ENVIRONMENT

• 6. Avoid distraction
• dours

conferring

• 7. Get out of factory wherever possible.

Golden Rules for Sensory - Sample Presentation

Create your own Random 3 digit numbers in Excel: =RANDBETWEEN(1,999) Triangle Test:

Golden Rules for Sensory - Sample Presentation Labelling is inconsistent, draws attention to one sample.

Golden Rules for Sensory - Sample Presentation Again different manner of labelling - font, colour, number of digits.

Different sample sizes – bias / not enough to evaluate for some samples, too much to drink in others. Golden Rules for Sensory - Sample Presentation

Golden Rules for Sensory - Sample Presentation

Golden Rules for Sensory - Sample Presentation Same cups Same labels Same label position 3 digit random codes All contain same amount: 30ml.

Practical – Triangle Test

Elderflower drink contains: Water, sugar, lemon juice, fresh elderflowers, citric acid. Does not contain:

• any of the official allergens as listed in 1169/2011
• alcohol

Vegan and vegetarian-friendly.

Discrimination Tests – THE TRIANGLE TEST

• Often called a DISCRIMINATION / OVERALL DIFFERENCE TEST.
• “In general (using all senses) can you tell a difference between 2 samples”.
• Common situations: INGREDIENT CHANGE. PROCESS CHANGE.
• Allows you to declare if the samples are SIGNIFICANTLY DIFFERENT or not.
• Does NOT allow you to say the samples are ‘SIMILAR’ where a significant

difference is found.

• Typically you have already determined that the samples pretty close in profile

and you want some data to decide one way or the other.

• Good number of people to use: 24 – 30 (most balanced : nos. divisible by 6)

Discrimination Tests – THE TRIANGLE TEST: Set up

• BALANCED DESIGN reduces bias.
• 6 possible orders of presentation:

where A = test sample B = control sample

• BEST PRATICE:

Give first 6 people one each of the above; then repeat each presentation

• rder an equal number of times.
• Get organised – make a plan….

A A B A B A B A A B B A B A B A B B

Discrimination Tests – THE TRIANGLE TEST: Results Analysis

Number of respondents from your test Number of people needed to correctly identify the different sample in order to declare a significant difference (p = 0.05; 95% confidence level) 6 5 7 5 8 6 9 6 10 7 11 7 12 8 13 8 14 9 15 9 16 9 17 10 18 10 19 11 20 11 21 12 22 12 23 12 24 13 25 13 26 14 27 14 28 15 29 15 30 15 31 16 32 16 33 17 34 17 35 17 36 18 … …

Critical Value Table for Triangle Test Data Analysis Data source: BS ISO 4120:2004 Triangle Test 95% confidence level

• Typical for food industry sensory
• Only a 5% risk that you wrongly conclude the

samples are different.

• r

You are 95% confident of the conclusion from the test.

Discrimination Tests – A SME Difference from Control Test

• QA during production
• e.g: raw materials /

intermediates …preventative

• QC finished product

…hold / release decision IS THE PRODUCT / RAW MATERIAL ‘IN SPECIFICATION’ ? IS THE PRODUCT / RAW MATERIAL CONSISTENT WITH PREVIOUS BATCHES? IS THERE A DETECTABLE FAULT? }

• Shelf – life testing NPD /

EPD …determine BBE

• Buy back testing

….QC once in supply chain …pick up misuse / abuse IS THE PRODUCT STILL RECOGNISABLE / REPRESENTATIVE OF BRAND?

Discrimination Tests – A SME Difference from Control Test Control / Reference Point Test samples

….

Brand std. Sample OR Documented Standard

• Spec.
• QAS
• Photos

People – know product well / trained

Discrimination Tests – A SME Difference from Control Test – Set up IN SPECIFICATION ACCEPTABLE OUT OF SPECIFICATION A scoring system : assesses & decision making RELEASE REJECT Sensory profile matches control Profile close but some variation Taint Off flavour Different profile RELEASE Root Cause Analysis

PRACTICAL – Basic Different from Control Test Scenario: Daily QA Taste Panel in Brewery

Sample ingredient list: Water, BARLEY malt, rice, hops.

• Remember to include Hidden Controls sometimes.
• Validates Method
• Validates Panel Ability
• Include a space for descriptive comment.
• Can develop a number of scales for key attributes.

Discrimination Tests – summary

• Are OBJECTIVE TESTs – no opinions / emotions involved!
• Who ? People from the company.
• Triangle Test is a commonly used overall difference test when you have 2

samples to test for general differences.

• Difference from Control Test is commonly used in QA/QC. Compare multiple

samples to a reference (specification/control sample) for overall difference.

• Paired Comparison Test is used for a specific attribute test when you have 2

samples to test for a difference in specific attribute (eg: which sample is sweeter?).

• Ranking Test is used to test for difference of specific attribute test when there

are 3 or more samples.

Consumer Testing - PRACTICAL Scenario: Acceptance test – Cloudy premium apple juice

I’ve just inherited an orchard, there are loads of apples, I’ve had some juice

• made. I think it tastes wonderful! I’m going to start a business…

1) Do consumers like the taste of my new juice? Why ? and by how much? 2) Is it sweet enough for them ? (I could store the fruit, press later; make it sweeter) 3) How does their opinion of my juice compare to their opinion for competitor juices? 4) Will people buy my juice?

Consumer tests – general rules PEOPLE: Who ? : How many ? : Everything we said on slide 15 about people applies. ENVIRONMENT: Where?: SAMPLES & PRESENTATION: Safe, representative, consistent, 3 digit random codes, balanced presentation

• rders…

Branding ?/ Not? TARGET MARKET CONSUMERS THE MORE THE BETTER – AT LEAST 100 SPECIFIC VENUE WITH INVITES or SHOWS/FESTIVALS/MARKETS Context will affect the result you get.

Consumer Testing - PRACTICAL

Shall I do a preference test against Copella ?

• What would that show us?

Whether the consumers prefer mine more than Copella’s.

• What wouldn’t it show us?

Whether they like my juice at all. Preferable = not necessarily likeable. How much they do or don’t like my juice in its own right. How big the difference is between their liking for mine and Copella’s. If they’ll buy it. Whether the sweetness level is perfect; only if preferable/not.

Consumer Testing – PRACTICAL – ACCEPTANCE TEST ACCEPTANCE TEST will allow me to: See if & how much they like: a) my juice. b) Copella juice. …..in their own right /against expectations. Ask some more questions to understand key attributes important to product. Each sample 1 by 1 + 9 point category scale:

“Please taste the sample & indicate your opinion by ticking a box below”

Consumer Testing – PRACTICAL – ACCEPTANCE TEST

Sample ingredient list: Apples, ascorbic acid (vitamin C). Does not contain:

• any of the official allergens as listed in 1169/2011
• alcohol

Vegan and vegetarian-friendly.

Consumer Testing – PRACTICAL – ACCEPTANCE TEST Diagnosis questions: 2) Comment: “Please explain why you scored it as you did?” (I have not yet influenced them). 3) Just About Right scale: Sweetness (specific attribute) “Please indicate by circling below, your opinion on the sweetness of the juice?”

(I have now influenced them, but I captured their unbiased thoughts first! So OK to ask this at this point).

Consumer Testing - PRACTICAL– ACCEPTANCE TEST

1) Do consumers like the taste of my new juice? Why ? and by how much? 2) Is it sweet enough for them ? (I could store the fruit, press later; make it sweeter) 3) How does their opinion of my juice compare to their opinion for competitor juices? 4) Will people buy my juice?

How to improve design?

Consumer Testing -– ACCEPTANCE TEST - Results Basic analysis results: 1) FREQUENCY of responses to each LIKING or JAR CATEGORY added up. 2) CALCULATE MODAL CATEGORY most frequently selected category. 3) PLOT GRAPHS 4) LOOK AT TRENDS - generally liking / disliking 5) TALLY COMMENTS > SEE TRENDS > USE TO EXPLAIN TRENDS More advanced analysis:

• Statistical testing for significant difference can be applied.

Consumer Testing -– ACCEPTANCE TEST - Results

Consumer Testing - ACCEPTANCE TEST – Sum up ACCEPTANCE TEST useful for: 1) Testing out NEW / NOVEL product on own or multiple recipes. 2) BENCHMARKING against competition. 3) Can add WHY? questions – discover KEY DRIVERS LIKING ….. drivers you weren’t thinking about

With packaging Do they like the packaging and branding ? How much? Will they try it ? With packaging & price & context Is everything meeting expectations? Will they buy it? Blind samples No packaging Do they like its sensory properties? How much? Will they repeat purchase ?

Wording & order questions CRITICAL!

Descriptive Tests

• Putting words to sensory sensations, so you can

communicate about the product to others: 1) CHARACTERISE the product in detail 2) DEFINABLE 3) RELEVANT to those using the data

• Describes the facts ; no opinion.
• Sensory description important for: 1) specifications

2) quality attribute sheets Often useful for marketing

Descriptive analysis – A qualitative profiling method for SMEs PEOPLE: Who ? : How many ? : Everything we said on slide 15 about people, sample presentation, consistency applies.

• Familiar / experienced with product.
• Interested / willing.
• People already good at verbalising sensations

6 - 8

• Writing marketing material – consumers > consumer language.
• Writing a specification – employees > language familiar to those in QA.

Pre-screening questionnaire

Descriptive analysis – A qualitative profiling method for SMEs Scenario: Qualitative descriptive profiling – Crisps A crisp manufacturer wants to characterise their new crisp product in order to create a sensory description for a specification and quality attribute sheet.

Ingredient list/ special dietary info: 1st sample – Potatoes, sunflower oil, sea salt. Gluten free. These chips are suitable for vegans. 2nd sample - Potato, high oleic sunflower oil (33%), salt.

Descriptive analysis – A qualitative profiling method for SMEs 1) GENERATE WORDS – ALONE – note down in grid provided. 2) DISCUSS as a group. 3) DEFINE the words.

• Name reference benchmarks > training

4) REFINE TERMS – merge common descriptors. 5) KEEP A TALLY – no. of times a word is used by multiple people

• Common usage > key attribute

A Basic Sensory Specification

Ap Appearance: Ar Aroma: Tas aste & fla flavour: Mou

• uthfeel/texture/trigeminal:

Aft Aftertaste:

Basic Sensory Profile : X brand Ready Salted Crisps

This qualitative profile represents brand standard quality.

More Advanced Sensory Specification

Traffic Light Sensory Specification: X brand Ready Salted Crisps

BRAND STANDARD QUALITY ACCEPTABLE – for release UNACCEPTABLE – reject Appearance Aroma Taste & flavour Trigeminal sensations Mouthfeel/ Texture Aftertaste

Insert photo Insert photo Insert photo

Descriptive analysis – A qualitative profiling method for SMEs

Descriptive Profile

Key sensory properties

Acceptance Test Data

How much liked & why

Your product. Competing products

POWERFUL STUFF!!

• Market awareness.
• Informed product development.
• properties to create
• properties to avoid

In Summary

• You can get a lot of info. from some very simple tests, on a low budget.
• Have a clear objective and a good plan.
• Do everything possible to reduce bias in:
• Samples
• Way test is conducted
• Use the most appropriate people for each test & use as many as possible -

remember about individual differences in sensory acuity & perception.

• Give the people best possible environment to make good measurements for you.
• Practice required to aid descriptive ability – you & employees.
• Sensory evaluation has benefits operationally & commercially.

Other Sensory Workshops For more detail: 19th September 2018 Sensory Evaluation Course

• Accredited by Institute of

Food Science and Technology.

• Exam to get an IFST certificate.

Please take a flyer ….

Further Work

For more information on sensory evaluation

• r

bespoke sensory work free to SMEs in Nottinghamshire & Derbyshire contact: Alice Jones – Sensory Scientist University of Nottingham 0115 951 6739 alice.jones@nottingham.ac.uk www.foodinnovationcentre.co.uk