The consumer preferences for new styrax based perfume products using a conjoint analysis approach

Thousands of tons of raw styrax resin are exported, while various derivative products with millions of dollars in value are imported for centuries. Tobarium Styrax Perfume is the first domestic product that explores the styrax essential oil formula for signature perfumes. The objective of study is to determine consumer preferences for a new styrax based perfume product. A conjoint analysis approach was utilized to determine consumer preferences for these innovation products. Fifty consumers were asked to provide an assessment of the combination of perfume products. Data processing was assisted by computer with SPSS 14.0 program. The influence of five product profiles on consumer preference including scent variant, package, volume, grade of essential oil, and prices were examined. The results show that higher utility values indicate greater preference. The highest utility value is indicated by factor grade (essential oil concentration), this illustrates the respondents’ greatest preference for the quality of signature perfume. The scent variant (importance value 53.958) has the most influence on overall preference. This means that there is a large difference in preference between product profiles containing the most and the least desired scents. This is because the preference for a particular scent is personal and unique which might have affected the perception of respondents. The highest consumer’s preference is a perfume with scent Rizla, package spray, volume 25 mL, grade signature perfume and price IDR 120,000. The grade and price play a significant role but not as significant as scent variant. Two statistics, Pearson’s R and Kendall’s tau, provides high correlation between the observed and estimated preferences. This information is expected to encourage this domestic innovation product development.


Introduction
Styrax resin has been traded throughout the civilization in Nusantara. Although it contributes significantly to community's livelihood, the added value of endemic and historic non-timber forest products has not yet been explored [1]. Thousands of tons of raw resin are exported, while various styrax resin derivative products in the form of medicines, perfumes, beauty and food ingredients with millions of dollars in value are imported annually [2]. One of the added value schemes is product innovation of Tobarium Styrax Perfume. This perfume is the first domestic product that utilizes styrax essential oil as a base-note for signature perfumes [3].
The main consideration of styrax perfume innovation is multi-dimensional benefits of styrax resin as a source of community's livelihood. Increasing added value is implemented by processing the raw materials into of high economic perfume products [3][4][5]. Knowledge about aromatic compounds of styrax resin is very limited in domestic communities. The resin associated with the scent of burned incense and it is underutilized for high-value products for centuries. The consequences, their aromatic formulas are unexplored, and potential for higher prices does not met in the local market.
Tobarium Styrax Perfume is a research invention which explores the incense fragrance formulation that known as god's resin and used by almost all religions and beliefs in the world. The market prospect to be targeted is to meet a demand for imported perfume products which reaches USD 401 million per year [6,7]. The consumer awareness for organic products is also an opportunity to develop these quality products.
Since the last decade, consumers have become more selective for a personal care product that they will use. Although many similar products are available, high-quality products and competitive prices will fulfill customer's satisfaction and increase the loyalty to the product. To fulfill the customer satisfaction, Tobarium Styrax Perfume provides six main scent variants including Rizla (floral fresh), Azwa (woody), Riedha (fruity fresh), Aphis (blue oceanic), Jeumpa (cananga) and Tiara (oud oriental). These products are packaged in various sizes that can be adjusted to consumer's needs and desires. The market segments being targeted are middle up society peoples and young executives who like exotic and energetic aromas; tourists visiting Lake Toba; aromatherapy oil collector; organic product buyers; and hospitality industry [4,6,7].
The objective of study is to determine consumer preferences for a new styrax based perfume product innovation. The information about some important attributes considered by consumers and how the combination of perfume product attributes that meet consumer needs and desires are expected to encourage the development of this domestic innovation product. Impacts of product development include increasing the value added of styrax resin that driving the local community welfares. The innovation product will increase the product efficiency and national competitiveness and it will build the national pride on domestic products.

Materials and Methods
A conjoint analysis approach was utilized to determine consumer preferences for Tobarium Styrax Perfume products. This approach was also applied in several previous studies [8,9]. Conjoint analysis is a research tool for developing effective product design. Using this analysis, the researcher can answer questions such as: What product attributes are important or unimportant to consumer? What levels of product attributes are the most or least desirable in the consumer's mind? The main result of this analysis is a design of a product or service, or a particular object desired by most respondents [9].
The scope for research is Tobarium Styrax perfume product with an analysis unit of 50 customers of this new innovation product. Respondents were selected purposively and interviewed during October 2019 to March 2020. The consumers were asked to provide an assessment of the combination of perfume products including scent, packaging, volume, essential oil concentrations and prices of perfume products.
Data processing includes generate orthogonal design and running the conjoint analysis were assisted by computer with SPSS 14.0 program. Two statistics analyses, Pearson's R and Kendall's tau were measured for correlation between the observed and estimated preferences.

Generating an orthogonal design
Generate orthogonal design generates a data file containing an orthogonal main-effects design that permits the statistical testing of several factors without testing every combination of factor levels. This design can be displayed with the Display Design procedure [9].
This study examined the influence of five factors on consumer preference-scent, package, volume, grade essential oil, and prices. There are six factor levels for scent, each one differing in scent variant (Rizla, Azwa, Riedha, Aphis, Jeumpa or Tiara); two package levels (roll on spray); two volume levels (15 mL and 25 mL); two grade levels (perfume or signature perfume) and three levels price (IDR 120,000; IDR 150,000; IDR 200,000) for last factor. Table 1  The first step is to create the combinations of factor levels that are presented as product profiles to the subjects. The generate-orthogonal design procedure creates an orthogonal array, also referred to as an orthogonal design. The combinations of factor levels are displayed in Table 2. Once an orthogonal design was created, it necessary to create the product profiles to be rated by the subjects or respondents. A listing of the profiles in a single table can be obtained or display each profile in a separate table or plan card. The information for each product profile is displayed Figure 1.

Running the analysis
After an orthogonal design was generated, consumer preferences for each product profile were obtained from 50 respondents as listed in Figure 2.

Figure 2. Preference data for parfume consumer
The preference data collected from survey of 50 customers who bought and will buy a Tobarium styrax perfume product. The data consist of responses from 50 subjects, each identified by a unique value of the variable ID. Subjects were asked to rank the 31 product profiles from the most to the least preferred. The variables PREF1 through PREF22 contain the IDs of associated product profiles. Analysis of the data is a task that requires the use of command syntax [10]

Utility scores
Part-worth utilities (also known as attribute importance scores and level values or simply as conjoint analysis utilities) are numerical scores that measure how much each feature influences the customer's decision to make that choice [9]. Table 3 shows the utility (part-worth) scores and their standard errors for each factor level. Higher utility values indicate greater preference. The highest utility value is indicated by factor grade, this illustrates the respondents' greatest preference for the quality of signature perfume. Furthermore, the lowest utility value is obtained by the Tiara scent, this indicates the lowest preference for this variant. As expected, there is an inverse relationship between prices and utility, with higher prices corresponding to lower utility (larger negative values mean lower utility).
The presence of a grade or scent or volume corresponds to a higher utility, as anticipated. Since the utilities are all expressed in a common unit, they can be added together to give the total utility of any combination. For example, based on the statistics analysis, the total utility of a perfume with scent Rizla, package spray, volume 25 mL, grade signature perfume and price IDR 120,000 is: utility(scent Rizla) + utility(package spray) + utility(volume 25 mL) + utility(grade signature perfume) + utility(price IDR120,000) + constant or 2.763 + 1.727+ 1.587 + 12.873+ (-1.186) + 5.171 = 22.935 If the perfume had package scent Azwa, package rollon, volume 15 mL, grade perfume and price IDR200,000, based on the statistics analysis, the total utility would be: utility(scent Azwa) + utility(package rollon) + utility(volume 15 mL) + utility(grade perfume) + utility(price IDR200,000) + constant or 0.113 + 0.863 + 0.793+ 6.437+ (-3.557) + 5.171 = 9.82 Table 4 shows the linear regression coefficients for those factors specified as LINEAR. The utility for a particular factor level is determined by multiplying the level by the coefficient. For example, the predicted utility for a price of IDR 150,000 was listed as -2.371 in the utilities Table 3. This is simply the value of the price level, 2, multiplied by the price coefficient, -1.186.

Relative importance
Running the analysis, it can determine both the relative importance of each attribute as well as which levels of each attribute are most preferred. The range of the utility values (highest to lowest) for each factor provides a measure of how important the factor was to overall preference [9]. Factors with greater utility ranges play a more significant role than those with smaller ranges. Table 5 shows the importance value for each factor as below: The results show that scent (importance value 53.958) has the most influence on overall preference. This means that there is a large difference in preference between product profiles containing the most desired scent and the least desired scents. This is because the consumer's preference for a particular scent is personal and unique [11].
The results also show that a package and volume (importance value 3.490 and 3.876, respectively) play the least important role in determining overall preference. The grade and price play a significant role but not as significant as scent. Perhaps there are because the range of grade and prices are narrow. In addition, consumers also consider that the price offered is still relatively affordable for a signature quality if compared to imported similar quality perfumes.
This result showed that it would be critical to consider how a product performs in a testing panel before launching it. Regardless of how effective a product is, if consumer does not like its scent, it may have a completely different perception of product, making the product fail miserably. Considering the fragrances used in this study are smoother, more refreshing, more striking or sweeter, which might have affected the perception of the respondents. Therefore, the scents are the subject of a complex discussion because its interpretation varies with context, and external factors can influence since the recognition of a particular scent as its personal interpretation, linked also to the sensory memory [11,12].

Correlations
The Table 6 displays two statistics, Pearson's R and Kendall's tau, which provide measures of the correlation between the observed and estimated preferences. The table also displays Kendall's tau for just the holdout profiles. The conjoint procedure computed the correlations between the observed and predicted rank orders for these profiles as a check on the validity of the utilities. In this case, there are high significant correlations between the observed and estimated preferences.

Reversals
When specifying LINEAR models for package, volume, grade and price were chose, a linear direction (LESS or MORE) for the relationship between the value of variable and the preference for that value were expected. The conjoint procedure keeps track of the number of subjects whose preference showed the opposite of the expected relationship [9]. These cases are referred to as reversals.  Table 7 displays the number of reversals for each factor and for each subject. There are three subjects showed a reversal for volume. That is, they preferred product profiles with lower volume. Furthermore, only one subject showed a reversal for price. This subject preferred product profiles with higher price. These reversals are probably due to complexity of consumer's preference for an aromatic product and