Post on 20-Dec-2015
21112005
Using Consumer Research for Product Development in
the Wood Processing Industries
Anders Q. Nyrud (NTI)
Anders Roos (Swedish University of Agricultural Sciences)
Oloav Høibø (Norwegian University of Life Sciences)
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
MARKET DRIVEN INNOVATION: A challenge for the wood processing industries What are the customers’ needs?
What product attributes can meet the needs?
Products with the desired properties?
The key properties must be easy to define and measure and possible to implement in industrial production
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
An increasing number of consumer studies on product and service attributes of wood products
Most have been focusing on industrial customers
Few studies employ well-established elicitation methods for product attributes (Brandt and Shook 2005)
Much remains to be done, especially to support a more market oriented and innovative forest sector
CONSUMER REASEARCH IN THE FOREST SECTOR
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
PURPOSE
Identify key properties/attributes of wood products
Investigate customers’ preferences of different types of wood products
Investigate costomer satisfaction and loyalty
Discuss different methods for consumer research and product development in the industry
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
THEORY
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
A FRAMEWORK
Understanding customer needs
Segmentation
Creativity enhancement techniques
Screening Planning tools
Product testing
Market comm.
SCM
(1) Opportunity identification (2) Development (3) Optimization (4) Launch
Human capitalresources
Physical capitalresources
Organizational capitalresources
Sources: van Kleef et al. (2005); Barney (1991); and Eisenhardt and Martin (2000)
Product innovation
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
The good, per se, does not give utility to the consumer; it possesses characteristics (attributes)
A good will usually possess more than one characteristic, and many characteristics will be shared by more than one good
ATTRIBUTES, NOT GOODS, GIVE UTILITY
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
THE PRODUCTS
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
DECK MATERIALS
Familiar product
Tree species
Treatment
Aesthetics
Durability
Environmental aspects
Context
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
BIOENERGY HEATING
Residential heating = End-consumers
Wood heating vs. electricity
Old inefficient stoves polluteNew improved woodstoves since 1998
Experiences of improved woodstoves
Factors influencing satisfaction
Factors influencing intention to continue to use
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
METHODS FOR ATTRIBUTE ELICITATION
SOME APPICATIONS
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Wood as a building material FOCUS GROUPS
Two groups of house owners (8, 6 respondents)
One group consisting of architects (7 respondents)
Procedure
– Presentation of participants
– Opinions of wood as material
– Reactions to various pictures of wooden housing
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Wood as a building material FOCUS GROUPS: Results
Wood is a popular building material Architects emphasize the artistic aspects of
material choice, home owners focus on practical issues
Architects are more detailed in description of wood, both aesthetic and practical
Little knowledge about different species of wood (for practical use) among house owners
Wood most appreciated on floor, in living room and in kitchen
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Deck materialsSENSORY ANALYSIS: Data collection
Evaluation of wooden deck materials
Consumer liking/preference
96 respondents
Physical product attributes
Sensory panel
Compare stated consumer liking and product attributes
Multivariate analysis
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Deck materials RESULTS: Hedonic sensory analysis
1
2
3
4
5
6
7
8
9
UntreatedIpé
Metal free treatment, pine
Furfurylatedpine
UntreatedRussian larch
Coppertreated pine
aab
b
c c
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
-1.0
-0.5
0
0.5
1.0
-1.0 -0.5 0 0.5 1.0
Color hue
Color intensityGrowt rings
Surplus Color
Gloss
Knot sizeKnot shape
Fracture
Rich surface
Growth ring hom
Hardness
Metal free teratement, pine
Furfurylated pine
Russian larch
Copper treated pine
1
23
4
56
7
8
9
10
11
12
13
14
15
16
17
1819
20
21
22
23
24 2526
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
4344
45
46
47
48
49
50
515284
89
90
91
92
93
94
9596
97
98
99
100101
102
103
104
105
106
107
108
109
110
111
112113
114 115
116
117
118
119
120
121
122
123
PC1
PC2
Whiteness
Dry knots
Knot patternKnot desnenessUntreated Ipé
Growth ring densitySmotthness
Evenness
(X-expl:64%,21%, Y-expl: 38%,25%)
Deck materials RESULTS: Preference map
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
-1.0
-0.5
0
0.5
1.0
-1.0 -0.5 0 0.5 1.0
Color hue
Color intensityGrowt rings
Surplus Color
Gloss
Knot sizeKnot shape
Fracture
Rich surface
Growth ring hom
Hardness
Metal free teratement, pine
Furfurylated pine
Russian larch
Copper treated pine
1
23
4
56
7
8
9
10
11
12
13
14
15
16
17
1819
20
21
22
23
24 2526
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
4344
45
46
47
48
49
50
515284
89
90
91
92
93
94
9596
97
98
99
100101
102
103
104
105
106
107
108
109
110
111
112113
114 115
116
117
118
119
120
121
122
123
PC1
PC2
Whiteness
Dry knots
Knot patternKnot desnenessUntreated Ipé
Growth ring densitySmotthness
Evenness
(X-expl:64%,21%, Y-expl: 38%,25%)
Deck materials RESULTS: Preference mapOPPORTUNITY FOR NEW
PRODUCT DEVELOPMENT
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Deck materials FREE ELICITATION
Sample Reasons to Prefer Reasons not to prefer
Untreated Ipé
Suits the clor of my house, Nice color, Solid impression, Looks exclusive, Trendy colorNice dark color, Tough color, resistantDoesn’t need treatment, Would be nice in my garden, Few knots, Seems resistantWill last over years
Too dark, Deforestation?, Rainforest species?I don’t like dark, Uneven colors, Ser kjedelig ut
Metal free treatment, pine
Looks real,color, Light and nice, Inviting, Lighnt and pleasant
Pink color?, Reddish coulour – seems unnaturalToo light red – unpractical, doesn’t fit anywhere
Furfurylated pine
Nice color, Conforms to the color of my house, OK color, Dark and even colorGives a solid impression, Like a roof of a cabin
Too dark!, This type of color is not trendy, Looks dirty
Untreated Russian larch
Light, pleasant colorAesthetically appealing wood, Looks naturalNeautral and nice color, Looks like good, quality wood
Gets easily stained, Needs regular treatment, Too much patterns, Too light
Copper treated pine
Nice color, Green is a known color for preserved wood
Don’t like the color!, Associated with pressure treatment, Too greenLooks poisonous, Low quality wood!, Uneven color
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Deck materialsCONJOINT ANALYSIS: Data collection
House and garden fair in the Oslo area
294 respondents
Rating of 17 samples
Fractional factorial design
12 stimuli and 5 holdouts
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Deck materialsCONJOINT ANALYSIS: Preliminary results
Factor Levels No of levels
Photo and product
Copper treated pine
Metal free treatment
Furfurylated pine
3
Price
90 NOK
140 NOK
190 NOK
3
Environmentally certified
Yes / No 2
Service Yes / No 2
Ready to assemple box
Yes / No 2
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Deck materialsCONJOINT ANALYSIS: Preliminary results
No. of responses 210
Part worth Importance Picture an product Copper -0.45 18 information Metal free 0.17
Furfurylated 0.28 Price 90 NOK/m2 0.57 25
140 NOK/m2 0.10 190 NOK/m2 -0.46
Certified Yes 1.03 51No -1.03
Service Yes -0.03 2
No 0.03 Module system Yes 0.09 4
No -0.09
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Deck materialsCONJOINT ANALYSIS: Segmentation
Envir. Aesthetic Price No. responses 90 40 75 Picture an product Copper -0.10 14 -1.76 52 -0.20 10 information Metal free 0.36 0.31 -0.09
Acetylated -0.26 1.45 0.30 Price 90 NOK/m2 0.14 8 -0.22 8 1.49 54
140 NOK/m2 0.06 -0.03 -0.33 190 NOK/m2 -0.21 0.25 -1.16
Certified Yes 1.57 71 0.84 27 0.88 22 No -1.57 -0.84 -0.88
Service Yes -0.07 3 0.37 12 0.01 0 No 0.07 -0.37 -0.01
Module system Yes 0.08 4 0.03 1 -0.35 14 No -0.08 -0.03 0.35
Share women 68 Share experience 77 Share plans 83
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Mail survey in Oslo
1200 questionnaires, 860 respondents
Experiences with bioenergy heating
Opinions about bioenergy heating
Intention to use bioenergy heating in the future
Lickert-scale
Bioenergy heatingUSER SURVEY: Data collection
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
Economic benfits
Subsidyeffect
Warmth
Time spent(Convenience)
Subjective norm
Environmental assessment
Know-how
Overallsatisfaction
H1
H2
H3
H4
H5
H6
H7
H8
Bioenergy heating INITIAL MODEL AND HYPOTHESES
Intention to continue using
bioenergy heating
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no Paths and standardized parameter estimates
Bioenergy heatingSTRUCTURAL EQUATIONS MODEL
KOSTBREN0.61
KOSTDRIF0.56
ENOK10.28
ENOK30.27
VARMEFF0.33
VARMKOMF0.22
VARMKONT0.39
VARMATM0.55
BRUKTID0.43
BRUKANST0.41
BETJKJEN0.33
VEDLKUNS0.80
EKSTRESP0.49
EKSTSTAT0.63
MILJO0.21
MILJGLOB0.16
MILJLOK0.27
Costs
Subsidy
Warmth
Time
Knowledg
Subjnorm
Environm
Satisf
Inttouse
OPPFTILF 0.18
OPPFHAND 0.64
INTFREMT 0.51
INTANBEF 0.50
Chi-Square=434.00, df=159, P-value=0.00000, RMSEA=0.046
0.91
0.60
0.70
0.71
0.62
0.66
0.85
0.86
0.82
0.88
0.78
0.67
0.76
0.77
0.82
0.45
0.72
0.61
0.89
0.92
0.85
0.61
0.09
-0.10
0.34
0.23
0.32
0.01
0.31
0.20
NO
RW
EG
IAN
UN
IVER
SIT
Y O
F LIF
E S
CIE
NC
ES
www.umb.no
CONCLUSIONS Sensory analysis and preference mapping can be
used for a more consumer focused product development
Focus grups and free elicitation can provide information on how preferences are formed
Conjoint analysis is a powerful tool in product development (e.g. screening)
Structural equations modelling can be used to evaluate consumers’ satisfaction with a given product
Socio-economic and demographic data can be used to segment markets, but psychograpic data can probably improve the results