Marketing Strategies and the Difference Level of Sales and Profits Performance of the Batik SMEs in Malaysia

In this research, the business owners of batik SMEs were analyzed on their marketing strategies. The objective is to examine if the marketing strategies of high performance batik SMEs differently from low performance batik SMEs. The high and low performance batik SMEs were categorized based on their annual sales and profits for three consecutive years. For this purposes, the questionnaires were distributed among 186 batik business owners in Malaysia. The data collected were then analyzed using Mann-Whitney U Test and Chi-Square Test. The result found that the high sales and profits performers are not only focused in local market, but they expand their sales out-of their states. For high sales performers, they are frequently advertised their businesses and using various promotional tools. Hence, efforts should be made by the low performers to find and expand their sales outside of their states, advertise their businesses more frequently and not only depends on personal network as promotional tools. Using various promotional tools especially media, could help in increasing sales. The conclusion is that the way of how the business owners market their products impact the performance of the firms. By identifying the marketing strategies of high performance batik SMEs, this research will provide batik entrepreneurs some guidance and will help low performers identify their weaknesses by comparing the marketing strategies practiced by high performers.


Introduction
Small and Medium enterprises (SMEs) in Malaysia have played an important role in the nation's economy and are a major source of various economic contribution.(Saleh & Ndubisi, 2006).They provide large employment opportunities, stimulate competition, aid to large companies and as seed-bed from which large companies grow (Hashim, 1999).
Central Bank of Malaysia reported that in 2006, SMEs comprises 99.2% or 546,218 of Malaysian business establishment, of which about 80% are micro enterprises, while the remaining are the large companies totaling 4,486 (0.8%).The large numbers of SMEs contribute to the greater employment opportunities, about 5.6 million or 56% of the total employment market (BNM Report, 2007).
However, despite the large total in numbers, in 2003, the SMEs only contribute 31.9% to country's GDP with an export value totaling 18.1%.Two years later, the growth only at 0.1% on GDP (32%) with the export value totaling only 19%.Similarly, productivity levels in the SMEs were found to be significantly lower than large enterprises as they generated an average value added per employee of just RM14,740, far lower than the RM47,830 generated by large enterprises (BNM Report,2007).
Based on the above statistics, SMEs in Malaysia are still not able to reach their full potential.(Hashim and Osman, 2003) indicated that the business practices in small firms are still limited in focus.As a backbone of country's economy and their significant presence and role, the SMEs need to be more aggressive to improve their productivity and competitiveness (ACCCIM, 2007).SMEs in Malaysia should not totally rely on government agencies, they should to find their own path of progress by relying on strategies that allow them to access new markets, increase their revenue and expand their customer base (Saleh and Ndubisi, 2006).
In Malaysia, manufacturing sector (including batik industry) is the major sector besides service, mining and agricultural.SMEs account for 96.5% or 39,436 of all enterprises in manufacturing sector (BNM Report, 2007).In Malaysia, an enterprise will be classified as an SME if it meets either the specified number of employees or annual turnover definition.For manufacturing SMEs, The Small and Medium Industries Development Corporation (SMIDEC) defined it as an enterprise with full-time employees not exceeding 150 or with annual sales turnover not exceeding RM25 million (BNM Report, 2007).
Recently, batik industry has been recognized as a priceless heritage that has potential market in Malaysia as well as in international market.Recognised for its priceless heritage for the country, batik industry has been increasingly supported by the government and NGO through various programmes locally and internationally.The programmes such as Annual Piala Seri Endon and Piala Seri Iman Design Competitions, Batik Festival, Annual Malaysian Week in London, Annual Kuala Lumpur International and Batik Convention and Exhibition (KLIB).The government officers are also directed to wear batik on every Thursday in order to keep this industry continuously alive.
The local batik industry was valued at RM370 million in 2003 but with the government push, since 2004 it generated RM400 millions in revenue per year.Batik industry could also generate tourism income through the event that has been created.
Due to its potential, batik has been highlighted in Ninth Malaysian Plan as one of the traditional product that should be developed further (The Ninth Plan, 2006).Nowadays, after so many efforts carried out by the government and NGOs, batik is now generating big incomes to batik entrepreneurs.Some of them have their products marketed at international levels and their pieces have been recognized as the world standard.
However, despite the success of the firms, many others in the batik industry are still not able to reach their full potential.After so many efforts given by the government and NGO to elevate the batik industry, only a few firms have high performance while the others are just happy with day-to-day income.
After reviewing previous studies, it has been found that marketing is the important aspect of all business activities and critical for the survival and growth of small businesses (Huang and Brown, 1999).Therefore, the objective of this study is to examine the marketing strategies of Batik SMEs in Malaysia and analyzed whether it contribute to the different level of performance between them.Through identifying the factors of marketing, the result of this study is hoped to provide the low performers some guidance to be at par as their counterparts.
This study is organized into five sections.Following this introductory section is section 2 which provides a review of literatures.Then, the third section explains the details of the research methodology used in this study.Data analysis and research findings are laid out in section 4. The summary of the research, its implications and limitations are presented in the last section with suggestions put forward on the direction for future research.

Theoretical Background
This research is using the resources-based view (RBV) approach as a theoretical basis.RBV approach has been first introduced by Wernerfelt (1984) and it further expanded by Barney (1986).The RBV is the way in which internal resources contribute toward a firm sustainable competitive advantage (Way, 2002).
Firm resources include all assets, capabilities, organizational process, firm attributes, information, knowledge, etc controlled by a firm that enable the firm to conceive of and implement strategies that improve its efficiency and effectiveness (Barney, 1991).Barney (1991) classified resources into 3 categories; Physical Capital Resources, Human Capital Resources and Organizational Resources.However, a subsequent distinction made by Amit & Shoemaker (1993) which split up the resources into resources and capabilities was frequently used by researchers.Capron and Hulland, (1999) defined resources as stocks of knowledge, physical assets, human capital and other tangible and intangible factors that a business owns or controls, which enable a firm to produce, efficiently and / or effectively, market offerings, that have value for some market segments.The characteristics of the resources are tradable and non specific to the firm (Amit & Schoemaker,1993) e.g: plan, equipment, capital, material, human resource etc.
Capabilities have been defined by O'Regan and Ghobadian (2004a) as firm's capacity to deploy its assets, tangible or intangible, to perform a task or activity to improve performance.The characteristic of the capabilities are firm-specific, non transferable and use to utilize the resources within the firm (Amit & Schoemaker, 1993).e.g: advertising, product range, design, brand name, etc.
The central question of this research is why firms show varying levels of performance?The RBT could provide the answer for this question as indicated by previous literatures, "varying performance between firms is a result of heterogeneity of assets and RBV is focused on the factors that cause these differences to prevail" (Amit and Schoemaker, 1993), "firms differ from each other since each has its own bundle of resources and capabilities" (O'Regan and Ghobadian, 2004a).Therefore, the application of resource-based theory could help in explaining why some firms consistently outperform other firms (Barney, Wright, & Ketchen, 2001).Many studies of firms' performance used RBV as a ground theory (e.g Mooreman and Slotegraaf, 1999;De Carolis, 2003;O'Regan and Ghobadian, 2004a).From the above explanation, resources are comprises of various assets which owned by the firms.This study choose to compare the marketing strategies of the firms as it is one of the important resources in organization (Huang and Brown, 1999;Panayides, 2004).

Performance
Refer to the accounting term; performance is the relationship of the income and expenses of an enterprise, as reported in the income statement (Kothari and Barone, 2006).Some studies relate the performance as a success (Walker & Brown, 2004).In the literature, performance is measured by either subjective or objective criteria.Profit and Total sales are among objective performance measures which widely used in the studies of SMEs (e.g Garg, Mittal, & Goyal, 2005;Arbuthnot, Sisler, & Slama, 1993;Craig, Martin, & Horridge, 1997).
Profit usually used because achieving profit is frequently identify as an overriding business goal (Kent, 1994) and total sales are considered a valid measure for presenting overall performance, especially in relatively homogeneous sample or in the same industry (Haber & Reichel, 2005).
Although some researchers described an objective data as sensitive, confidential information and difficult to obtain from the respondents (Berthon, Ewing, & Napoli, 2008), it would seem to be the simplest way to assess performance (Haber & Reichel, 2005).Subjective measures would be the preference of some researchers due to the difficulty in obtaining the objective data (eg.O'Regan & Ghobadian, 2004b;Durham & Littrell, 2000;Walker & Brown, 2004) or to overcome a problem when a sample contains a variety of industries (Allen and Helms, 2006).However, since subjective measures of performance are based on the owner's perception, they increase the possibility of measurement error and the potential for bias (Kotey, 2005).Therefore, if researchers limit themselves to a single industry (as the current study), the objective performance measure may be more meaningful (Allen and Helms, 2006).This research considered to use objective performance as it represents the narrowest conception of business performance.As for this study, the objective is to examine the factors that caused the sales and profits performance between batik SMEs differ, this study therefore operates firm's performance in terms of sales and profits.

Marketing Strategies and Performance
Panayides (2004) defines marketing as the process of planning and exacting the conception, pricing, promotion and distribution of ideas, goods and services to create exchanges and satisfy organizational objectives.Marketing has been widely acknowledged as being the most important aspect of all business activities and critical for the survival and growth of small business (Huang and Brown, 1999).The implementation of a series of actions towards the organization and the market would achieve marketing objectives and improve organizational performance (Panayides, 2004).
However, sales and marketing issues were found to be the most dominant problems encountered by the small business entrepreneurs.Study by Huang and Brown (1999) indicated that problem in the promotion and marketing research were most frequently faced by the small business operators.The problems included the selection of promotional media, content design and format of promotional media, market size, location and addresses of the potential customers.Due to the above problems, many owner-managers rely largely on their intuition, experience or judgment to make decision in marketing their products or businesses (Berthon et al, 2008).Therefore, marketing strategies are frequently adopted by researchers to be studied as by doing so, it might give a clue for the entrepreneur on how to solve the marketing problems.
Marketing strategy is also thought to be one of the related issues to the ability of SMEs in order to gain competitive advantage (Pavic, Koh, Simpson & Padmore, 2007) and having a pivotal role in enhancing business performance (Sharma, 2004).Competitive advantage and improved performance are the objectives of business organization (Panayides, 2004).
In order to gain a competitive advantage and to deal with the current challenges such as growing competition in the domestic and international markets, more demanding and assertive customers and rapid advancement in technology, the businesses must have more distinctive and purposeful marketing strategies and they should be effectively implemented (Sharma, 2004).
Among the marketing strategy studies, target market, advertising and promotion are frequently cited as factors that contribute to the high performance of SMEs.According to Lincoln and Naumann (1982), advertising plays a critical role in the success or failure of many small businesses.Study by Bhaskaran (2006) on SMEs in Australia indicated that firms which developed new markets for their products recorded higher sales and profits.
Other than target market and advertising, promotional activities are also critical factors for business performance (Craig et al., 1997).Promotion can enhance a firm's profitability by securing increased sales volume at a decrease total expense percentage and by accelerating turnover (Shim and Drake, 1991).In Shim and Drake (1991) study, profitable firms gain competitive advantages by having well-planned promotional strategies and attempting to expand their customer base to a large geographic area while less profitable businesses do not extend promotional advertising beyond their local area (Littrell, Stout, & Reilly, 1991).
The above studies provide evidence that marketing strategies have an important role to play in explaining firm performance.Fail to adapt and practice in suitable condition, place and need, will contribute to the low performance.On the basis of the above discussion, this study then hypothesized that: H1a: There are differences in the volume of direct sales between high and low sales performers of batik SMEs.
H1b: There are differences in the volume of direct sales between high and low profit performers of batik SMEs.
H2a: There are differences in the volume of out-of-state sales between high and low sales performers of batik SMEs.
H2b: There are differences in the volume of out-of-state sales between high and low profit performers of batik SMEs.
H3a: There are differences in the volume of international sales between high and low sales performers of batik SMEs.
H3b: There are differences in the volume of international sales between high and low profit performers of batik SMEs.
H4a: There are differences in frequent advertising between high and low sales performers of batik SMEs.
H4b: There are differences in frequent advertising between high and low profit performers of batik SMEs.
H5a: There are differences in the type of promotional tools between high and low sales performers of batik SMEs in term of (i) personal network, (ii) media and (iii) competition and exhibition.
H5b: There are differences in the type of promotional tools between high and low profit performers of batik SMEs in term of (i) personal network, (ii) media and (iii) competition and exhibition.

Research Instrument
In order to achieve the objective of examining the marketing strategies that contribute to the different level of performance in the batik industry, a questionnaire was designed to obtain the relevant data.

Design of the Questionnaire
There are seven variables involved in this research.Two are dependent variables, and the other five are independent variables.The instruments for measuring these variables are discussed in this section.It consists of two sections, one for the independent variables (marketing strategies) and one for the dependent variables (sales and profits).Some questions required the respondents to tick on the correct answer from the given choices and some other questions asked the respondent to fill in the blank.The detailed measurement of the variables in each section will be explained in section 3.1.2

Measurement
The research on performance in the batik industry has not attracted many researchers.Therefore, since there is lack of references on performance in the batik industry, the variables in this research were inspired from research findings on small business studies which have been identified as associated with SMEs performance.Some of the variables however were modified in order to suit with the actual conditions of the batik industry.The previous literatures which have several strengths relevant to this research were Littrell et al. (1991), Shim and Drake (1991) and Chaganti and Chaganti (1983).
Marketing strategies measurement comprised of the number of sales made directly to the customers (Chaganti & Chaganti, 1983), the number of out-of-state sales, the number of sales marketed at international level, the frequency of entrepreneur advertising his/her product (Littrell et al.,1991) and promotional tools (Shim and Drake, 1991).Under this section, the respondents were asked to select the most relevant answer by ticking the corresponding boxes The questions for promotional tools were classified under three categories named as personal network, media and competition and exhibition promotional tools.Personal network includes the promotion done by the family members or friends, through promotion sales and distributors.Through media, the promotional tools were magazine/newspapers, television and internet.For competition and exhibition, the promotions were done through fashion show, design/art competition, expo and exhibition and sponsored cloth.The items for design/art competition, and expo and exhibition were added by the researcher from the observation of recent ways the batik industry being promoted in Malaysia while cloth sponsor's item was identified by the researcher during pilot study.
The performance measures used in this research were the annual sales and annual profits for three consecutive years from 2005 to 2007.In this section, the respondents were required to fill in the answers with the appropriate figures.The average sales were then derived by adding the annual figures of annual sales for over three years period and divided by three.The same procedures were repeated on the annual profits.Finally, the high and low performance of batik SMEs were obtained by splitting the data at the median of average sales and profits.

Validity of the Instrument
As stated in section 3.1.2,the variables used in this study were inspired from various small business studies especially those related to the craft and textile industries.For example, study by Littrell et al. (1991) was in the craft industry and Shim and Drake (1991) was in the textile industry.The craft and textile industries were chosen since they are very close to the batik industry.In Malaysia, batik is also classified as craft and usually used or drawn on textile.
However, since the measurement of the variables in previous studies has originally developed on textile and craft industries, this study only inspired the measurement from those studies and not totally taken item by item.For example, in Littrell et al. (1991) study, they measured the variable out-of-state sales in term of the region, either Midwestern, Eastern, Western or Southern, however this study measured out-of-state sales in term of the amount of sales made out of Kelantan and Terengganu either none, less than half, half, more than half or all.The material was then pre-tested in the survey development stage to enhance the validity of the instrument.A pilot study was conducted.The detail information and results of the pilot test as presented in section 3.1.4.

Pilot Study
For pilot test, a random sample (n=10) was selected from batik SMEs population (N=186).Ten questionnaires were then distributed to 10 SMEs in Malaysia in order to identify and overcome the problems that will be faced related to the proposed survey instrument.The pilot study was also conducted to verify that the questions were clearly understood and the responses would be consistent to the purpose of the study.The outcome of the pilot testing required several corrections be made to the developed questionnaires as follows: (1) There were several questions in the questionnaire which had been misunderstood and difficult to understand.Therefore a number of changes were made especially to the wording of the sentences in order to improve the clarity of the questions.
For example, during pilot test, the respondents were required to state their amount of sales directly to the customer, amount of out-of-state sales, and amount of international sales in the form of percentage.However, many respondents found it was difficult to state the answer compared to if they were asked in the form of multiple choice question.Some of the respondents straightly gave the answer in the form of either less than half, half or more than half.Therefore those questions were changed to the multiple choice question and the list of answers was in the term of none, less than half, half, more than half and all.
(2) Sponsored cloth for celebrities was identified as another promotional tool used by the batik SMEs in promoting their products.Therefore, sponsored cloth had been included into the list of promotional tools under competition and exhibition construct.

Population
Empirical data was collected from 186 batik insiders in Malaysia which met the definition of SMEs by Small and Medium Development Council (SMIDEC).As the batik industry is included in the manufacturing sector, the SMEs in this study were defined using the general definition given by SMIDEC on manufacturing sector.In manufacturing sector, the business will be classified as SMEs if the annual turnover of the firm is below RM25 millions or the total employees is below 150.
The states of Kelantan and Terengganu in Malaysia were chosen in this study for two reasons.Firstly, it was because of their backgrounds as the two states which batik first grew in Malaysia (World Batik Council, 2005).Secondly because most of the batik insiders are also in these two states.The number of SMEs which involved in the batik industry in Kelantan and Terengganu were collected from the directories of SMIDEC, Batik Guild and MARA.After sorting the redundant lists, the total populations obtained from the directories in the two states were 186 batik SMEs.The number of population obtained as represented in Table 3.1.Due to the small number of population, a complete survey was conducted in order to get the maximum data.

Data Collection
The data was obtained through distributing the questionnaires to the batik SMEs which were answered by the entrepreneur or someone with complete information about the firm.This is to ensure the information obtained is accurate and reliable.
The data had been collected between September and October 2008.The researcher recruited 10 undergraduate students from the department of Accounting and Finance, University Malaysia Terengganu to collect the data.For this research, the enumerators were asked to use the method of premise-to-premise to ensure the questionnaires were interpreted in the correct way by the respondents.A personalized cover letter that explained the purpose of the study accompanied each questionnaire.After two months, the target of 186 answered questionnaires was successfully achieved.

Results and Discussions
The data collected was analyzed using Statistical Package for Social Science (SPSS) Software Version 12.
Statistical procedure employed in this study included descriptive analysis, the Chi-square Test and Mann-Whitney U Test.

Descriptive Statistics of Performances
Respondents were asked to provide the figure of annual sales and annual profit for three consecutive years from 2005 to 2007.These figures were then calculated to derive the average sales and average profits by adding the annual figures of annual sales / profits for over three years period and divided by three.For skewness and kurtosis, a value of 0 corresponds to a normal distribution (Maltby and Day, 2002).However, this table showed the value of 3.537 and 4.603 for skewness and the figure of 14.882 and 25.542 for kurtosis.This means that the distributions of performance data in this study were positively skewed and more pointed than a normal distribution.According to Field (2005), highly skewed distribution brings mean too far from the majority of the values to accurately reflect the distribution.In this case, the median is more appropriate than mean since it represents better description of the data.Therefore, for this study, the high and low performance of batik SMEs were obtained by splitting the data at the median of annual sales and annual profits figures.
This research was carried out with the aims to examine whether marketing strategies of one group differ from one another.Therefore, for that purpose, the population has to be categorized into high and low performance firms before the hypotheses could be tested.
Based on the descriptive statistics on both performance measures in Table 4.1, the conclusion was made as follows: those SMEs below RM 189,333.33 in average annual sales were categorized as the low sales performers.Those SMEs with annual sales exceed RM 189,333.33,were categorized as high sales performers.For low profit performers, their annual profits were below RM 53,833.33 and to be classified as the high profit performers, the SMEs should have the annual profits more than RM 53,833.33.The other figures in this table represent the medium, minimum and the maximum values for each category.

Descriptive Statistics of Marketing Strategies
Table 4.2 presents the descriptive statistics on marketing strategies items.Based on the table, most of the entrepreneurs (60.2%) practised sales directly to the customers.Only 1.1% of the respondents answered that they do not practice this type of marketing.4.8% answered less than half of their product sold directly to the customers, 12.9% answered half of their products and quite a large number (21.0%) answered that more than half of their products sold directly to the end customers.
The cumulative frequencies showed that 66.7% of the respondents in this study, marketed their products outside their states.In this figure, 27.4% of the respondents marketed less than half, 23.1% marketed half and 16.1% marketed more than half of their products out of their states.The remaining 33.3% focused on local market.
Regarding international sales, the data showed that in cumulative number, only 19.3% of the respondents exported their products while the majority 80.5% only concentrated in the domestic market.
The result showed that only 30.8% in cumulative or less than half respondents advertised their products.From this number, only 5.4% of the respondents often do the advertisements while the others 20.5% and 4.9%, rarely and sometimes.These results showed that the advertising was not given much attention by the entrepreneurs since most of the respondents (69.2%) responded that they never advertise their products.
The result for promotional tools appeared to be consistent with the attitude towards advertising.An examination of the data showed that over 93% of the respondents were just rely on personal network as their promotional tools compared to media 15.6% and competition and exhibition promotional tools which only used by 37.6% of the respondents.

Response on Research Objectives and Hypotheses Testing
RO: To examine whether there are differences in marketing strategies between high and low performance of batik SMEs in term of direct sales, out-of-state sales, international sales, frequent advertising and promotional tools.
In order to achieve research objective and to test the hypotheses, Mann-Whitney U Test was conducted on direct sales, out-of-state sales, international sales and frequent advertising while Chi-square test was used on promotional tools.Commonly, the independent-samples t-test is often appropriate to compare the levels, or averages, of two independent samples of data in the form of measurements.However, for the case of where there are some highly deviant scores, outliers or non-normal data distribution, which can inflate the values of the denominators of t-statistics, the non-parametric Mann-Whitney U Test is the alternative (Sprinthall, 2007).The results of these test as presented in table 4.3 and 4.4.The first result from table 4.3 is sales made directly to the customers.The data under sales performance measure showed that the mean rank for low performers was greater than mean rank of high performers.The mean rank of low performers was 99.41 compared to the mean rank of high performers which was 87.59.However, this difference was not significant at 0.05 levels as presented by the p-value of 0.088.Under profit performance measure, the mean rank for low profit performers was 96.80 and the mean rank for high sales performers was 90.20.This result was also not significant at 0.05 as indicated by the p-value equals 0.341.Thus, hypothesis 1a and 1b are rejected.This means that there is no difference in the number of sales made directly to the customers between high and low performance batik SMEs in Malaysia.
For out-of-state sales, by looking at table 4.3 reveals that the mean rank for the low performers was lower than the mean rank of the high performers for both performance measures.The mean rank for low sales performers was 81.46 while the mean rank of high sales performers was 105.54.The mean rank for low profit performers was 83.57while the mean rank of high profit performer was 103.43.With the p-value of 0.002 for sales performance measure and 0.009 for profit performance measure, this variable is significant at 0.01 levels for both measures.Thus h2a and h2b are accepted.These results confirmed that there is significant difference in the number of out-of-state sales between high and low performance of batik SMEs in Malaysia.
Regarding international sales, the mean rank of low sales performers was 88.85 and the mean rank for high sales performers was 97.20.Although the mean rank for low sales performers was lower than the mean rank of high sales performers, this difference was not significant at 0.05 levels as indicated by p-value of 0.122.This result was similar in the profit performance measure.However the result showed the perfect figure of non-significance with p-value equals 1.000.This result occurred since both high and low profit performers had the mean rank of 93.00.These non significant results rejected hypothesis 3a and 3b.Therefore, these results suggest that there is no difference in the number of international sales between high and low performance of batik SMEs in Malaysia.
In term of frequency of advertising, the result of Mann-Whitney U Test on sales performance measure in table 4.3 showed the p-value of 0.002, indicated this variable was significant at level 0.01.The mean rank for low sales performers was 83.23 while the mean rank of high sales performers was 102.67.This result however differs under profit performance measure.Although the mean rank of low performers was 88.35 while the mean rank for high performers was 97.60, this difference was not significant at 0.05 levels as indicated by the p-value of 0.148.
The results from Table 4.3 mean that the difference in frequency of advertising appeared between high and low sales performers but not between high and low profit performers.Therefore, hypothesis 4a is accepted but hypothesis 4b is rejected.This means that in Malaysia, there is a difference in the frequency of advertising between high and low sales performances of batik SMEs.However, there is no difference in this variable between high and low profit performance of batik SMEs.
Regarding promotional tools, data analysis in table 4.4 showed that 93.5% of the respondents chose personal network as promotional tools while 6.5% did not.Of the 'yes' total, 48.4% were low performers and 45.2% were high performers.Although the total of low performers was higher than the high performers, the different was too small as indicated by p-value of 0.073 which means this variable was not significantly different even at 0.05 levels.This result repeated in profit performance measure.The p-value of 0.551 indicates the difference was not significant at 0.05 levels.Both non-significant results were then rejected hypotheses 5a (i) and 5b s(i).This suggested that there is a difference between high and low performance of batik SMEs in Malaysia in term of using personal network as promotional tools.
For media promotional tool, the difference exists between high and low sales performers.Table 4.4 showed that there was only 4.8% of low performers used this promotional tool compared to 10.8% of high performers.This difference as indicated by the p-value of 0.026 which was significant at 0.05 levels.Under profit performance measures, however, the result showed the p-value of 0.840 indicated that the number was not significant at 0.05 levels.This could be seen from the distribution in which media was used by 8.1% of low profit performers and 7.5% of high profit performers.This gives the contrast result from sales performance measure.Thus, hypothesis 5a (ii) is accepted but hypothesis 5b(ii) is rejected.These mean that in Malaysia there are differences in using media as promotional tool between high and low sales performance of batik SMEs but not between high and low profit performance of batik SMEs.
The result from table 4.4 also showed that there was no difference in the data distribution between high and low performance of batik SMEs in regard to competition and exhibition as promotional tools for both performance measures.Of the similar total of 37.6% which answering 'yes', both high sales and profit performers contribute 21.5% compared to low sales and profit performers which contribute 16.1% of the total.These differences however were not significant even at 0.05 levels as indicated by p-value equals 0.130.Therefore, h5a(iii) and h5b(iii) are rejected.This indicated that there is no difference in using competition and exhibition promotional tools between high and low performance of batik SMEs in Malaysia.

Discussions
In this section, the findings which obtained from section 4.2 were discussed.Each variable and its finding were separately discussed in each sub-section.

Sales Made Directly to the Customer and Different Level of Performance
The findings from this study showed that there were no differences in the number of sales made directly to the customers between high and low performance of batik SMEs in Malaysia under both performance measures.This means that the number of sales made directly to the customers was not differ between high sales performers and low sales performers.This also means that there were no differences in the number of sales made directly to the customers between high profits performers and low profits performers The findings of this study is contrast to the study by Siu (1999) which found that high performing firm was less involve in personal selling.The reason of the contrast finding may be due to lack or unclear target market among batik SMEs.They might identify the general public as their markets.The target market was not well defined because products were made for any interested customer.This might be the reason as noted by Shim and Drake (1991), lack of information about target market is a major problem among small firms.

Out-of-state Sales and Different Level of Performance
In term of out-of-state sales, the results indicated that there were differences in this variable between high and low performance of batik SMEs in Malaysia.This difference was appeared under both sales and profits performance measures.This means that the number of out-of-state sales was differed between high and low sales performers.It means that the number of out-of-state sales made by high profits performers was also differed from the low profits performers.
High performers were expected to have broader target market than the low performers.This findings support the expectation when the results showed that the mean rank of high performers which marketed products out-of-state sales was greater than those of low performers.This finding is in agreement with the study by Siu (1999) which found that higher performing firms tend to use market expansion as their marketing strategies.However, the finding in this study was contrast to the finding in Chaganti and Chaganti (1983) study which found that the profitable firms concentrated on local markets while the less profitable firms tended to serve mostly regional and national markets -markets far from plant locations.

International Sales and Different Level of Performance
The data analysis showed that there was no difference in the number of international sales between high and low performance of batik SMEs in Malaysia under both performance measures.This means that the number of sales made at international level was not differ between high sales performers and low sales performers.The number of international sales was also not differed between high profits performers and low profits performers.
From the descriptive statistics results, only 19.3% of batik SMEs in Malaysia were involved in international sales.Conclusion can be made that batik SMEs in Malaysia are still not ready since many SMEs respondents are much more rely on domestic sales.This finding is in agreement with the study by Durham and Littrell (2000).Study by them on handcraft producers in 11 developing countries indicated that it was hard to reinforce international market because of time constraints, labor constraints and transportation constraints.Accepting these constraints, the entrepreneurs emphasized market diversification by concentrating on local and national rather than international sales.Therefore, Durham and Littrell (2000) concluded that marketing internationally was not always desirable and possible for the handcraft industry.

Advertising and Different Level of Performance
The finding concluded that there were differences between high and low performance of batik SMEs in Malaysia in term of frequency of advertising.However, the difference occurs only under sales performance measure but not under profit performance measure.This means that the attention is given on advertising of high sales performers were differed from the low sales performers.This result is consistent with the suggestion by Lincoln and Naumann (1986) in that advertising plays a critical role in the success or failure of many small businesses.
As stated previously, the difference however only exists between high and low sales performers but not between high and low profit performers.Therefore, regarding profit performance, this study is in agreement with De Carolis ( 2003) study in US which found that advertising is not significantly related to profitability.In addition to that, Fillis (2004) study also agreed that advertising was not an option for many craft firms.His study on smaller crafts firm in the UK found that instead of using advertising tools, the entrepreneurs prefer to use a combination of lower cost approaches such as word of mouth and trade show attendance in order to construct a reputation over time.
From the overall results, suggestion can be made as advertising is a necessary expense that does not immediately contribute to accounting returns but may in fact contribute to the future value of the firm.Therefore, for their long survival, the efforts still have to be made by low performers to advertise their products or businesses.

Promotional Tools and Different Level of Performance
The promotional tools measured in this study are in term of personal network promotional tool, media promotional tool and competition and exhibition promotional tool.
Personal network promotional tool was used by the majority of the respondents.In this study, personal network promotional tool comprised of promotion through family / friends (asking friends or relatives to advertise), sales promotion (special sale price, purchase with purchase, giveaway with purchase and free sample) and distributors (which moving around office, schools, clinics, houses and any premises showing the products).
The data analysis showed that there was no difference in using personal network as promotional tool between high and low performance of batik SMEs in Malaysia under both performance measures.This means that the personal network has been used by both high and low sales performers.This also means that the personal network was the choice of both high and low profits performers.This is somehow contradicts to the findings of Littrell et al. (1991) study which indicated that promotion through personal networks was linked with less successful producers.One reason for this contrast results might be that previous research dealt with the tools individually while the current research refers the tool as a type of group.
Media promotional tool comprised of promotion through television, internet, radio, newspaper and magazines.
The data analysis suggested that there was a difference between high and low performance of batik SMEs in Malaysia in term of using media as promotional tool.However, the difference occurs only under sales performance measure but not under profit performance measure.This means that the number of high performers which use media as promotional tool were differed from the number of low sales performers.However, the number of those using media as promotional tool was not differed between high profit performers and low profit performers.On the other hand, the SMEs may find media advertising is too expensive.As noted by Littrell et al., (1991) media advertising appears expensive for this size of businesses since the advertising and promotion budgets for small business may have to be very small.
Competition and exhibition promotional tools comprised of promotion through participation in batik design and art competition, fashion show, tradeshow and cloth sponsored.The results showed that there was no difference in using this tool between high and low performance of batik SMEs in Malaysia under both performance measures.This means that use competition and exhibition as promotional tool do not cause any differences in the performance levels since both high sales and high profits performers was not differed from the low sales and low profits performers in using these tools.
The explanation might be as suggested by Shim and Drake (1991).They recommended that SMEs should understand the efficiency and effectiveness of each promotional and advertising tool.According to them, there were two types of promotional strategies which frequently used by profitable firms, institutional strategy and special promotional strategy.Institutional strategy is designed to build the reputation of a firm.Its primary objectives are to keep customers' products are sold over a long period of time and to seek the steady patronage of customers.However, special promotional strategy is designed to influence the sale of products and services with its primary objective being the immediate response to specific merchandise.From the explanation, we could say that competition and exhibition is an excellent tool to achieve institutional objectives (designed for reputation of the firms) but not to achieve special promotional objective (designed for increasing sales immediately).As a conclusion, competition and exhibition are not the correct tools to achieve higher performance in sales and profits in a short run, however, in a long run it might give an effective impact.
From the findings in promotional tools variable, clearly that the high performers used various type of promotional tools.They do not rely only on promotion through personal networks.This finding is in agreement with the study by Wood (2006) on small event firms in the United Kingdom which found that small firms are likely to perform better if they use a variety of promotional methods.

Implications and Conclusions
The initial aim of this study was to examine why some firms achieved higher in sales or profits than the other firms in the same industry.In answering the question, the marketing strategies as one of the important firm's resources of high and low performers (in sales and profits) of Batik SMEs were compared on a number of factors.The factors were direct sales, out-of-state sales, international sales, frequent advertising and promotional tools.In order to examine in which factors that the high performance of batik SMEs were significantly different from low performance of batik SMEs, the Mann-Whitney U Test and Chi-square Test were performed on those factors.The majority of performance research is founded in a non-comparative case study analysis.In contrast, the findings of this study offered the opportunity to compare and contrast the firm's resource of high and low performance of the firms.
As a summary from the findings, the high sales and profits performers are not only focused in local market, they expand their sales out-of their states.For high sales performers, they are frequently advertised their businesses and using various promotional tools.Hence, efforts should be made by the low performers to find and expand their sales outside their states, advertise their businesses more frequently and not only depends on personal network as promotional tools.Using various promotional tools especially media, could help in increasing sales.
The study has implications for both research and practices.For research, this study is one among a few which empirically test the performance levels of the batik industry.Most of the previous studies in the batik industry are likely to focus on the process of batik making such as arts, tools, colors of batik design and designer's problem (e.g.Hodge, 1999;Lillethun, 2002;Beament, 1972;Joseph, 1982).Thus this study could overcome the lack of research on performance in the batik industry.For practitioners, the findings of this study are interesting for the batik industry entrepreneurs.The high performers would know their strength and low performers could identify their weaknesses by comparing the marketing strategies practicing by high performers.Thus, if the SMEs want to perform at a significantly higher level than their competitors, the SMEs must excel at the critical and most statistically significant practices identified in this study.The current findings also has the implication on other related agencies include government and related private agencies as well as bankers in constructing new policy in order to elevate this priceless heritage industry.However, there are two limitations which might be the suggestions for future research.Firstly, as this study is limited in two states, there is a need to get other states to make a generalization.Secondly, the findings are limited to the batik industries, future research may investigate how the findings reported here translate to other industries.

Table 1 .
Population data from the directories

Table 2 .
Descriptive statistics on performance measures

Table 4 .
1 shows the descriptive statistics on performance measures.For annual sales, the mean value was RM 392,218.44 with a standard deviation of 595,073.55.For annual profits, the mean value was RM141,758.30with a standard deviation of 271,531.05.The SMEs that included in this study had a median sales of RM189,333.33 with a range from RM6,600.00 to RM4.2 millions.Looking at annual profits, the average ranged from RM 2,533.33 to RM2.2 millions with a median of RM53,833.33.

Table 3 .
Descriptive results on marketing strategies

Table 4 .
Results of Mann-Whitney U test on marketing strategies