Materials and methods

The 2014 National Green Industry Survey gathered information on business practices and operating results for calendar year 2013 or fiscal year 2013–14. The questionnaire and survey protocol were approved by the University of Florida Institutional Review Board for compliance with ethical standards for human subjects research.

This study represented the sixth national survey conducted by the Green Industry Research Consortium, following previous surveys in 1989, 1994, 1999, 2004, and 2009. The 2014 survey targeted plant dealer firms, for the first time, in addition to growers, with new questions added regarding retail marketing practices. For the 2014 survey, a list of over 110,000 grower and plant dealer firms in the United States was developed. The list contained information on company name, contact person, mailing address, and in some cases telephone numbers, e-mail addresses, and type of business (grower or dealer). The listings for each state were obtained from members of the National Plant Health Board, an organization representing the plant health regulatory agencies in each state, which in most cases is the state department of agriculture or its equivalent. All commercial growers and dealers of live plants are required to be registered and annually certified for compliance with phytosanitary regulations, so these lists of plant growers can be considered exhaustive to the extent of force of law. Some states made their lists of firms available on a website, whereas others provided it on request. Usable lists of certified nurseries and plant dealers were obtained from all states except Alaska, Montana, and New Mexico; for these states, lists of firms were obtained from the InfoSource USA database. After screening to eliminate duplicate entries and firms no longer in business, the effective population had over 104,000 firms. A total of 32,000 firms were targeted for the survey, including 15,000 grower or grower/dealer firms randomly selected to receive the questionnaire mailed via the U.S. Postal Service, and all 17,000 firms with e-mail addresses who received the survey via e-mail (Internet). Firms to be surveyed via e-mail were removed from the population considered for the mail survey to avoid duplication.

The surveys were distributed during July to Aug. 2014. Following best practices for survey research, an introductory letter was first sent to selected firms to explain the purpose and benefits of the project, and all printed survey materials contained the logos of the sponsoring organizations to enhance the credibility and legitimacy of the survey (Dillman et al., 2008). Two mailings of the survey questionnaire were sent to firms selected for the mail survey, along with postage-paid return envelopes. Reminder postcards were mailed to respondents about 1 week after each survey mailing. Mailed questionnaires were imprinted with a code number matched to the mailing list, to identify respondents for quality control purposes. Completed surveys were returned to the University of Florida for data entry, cleaning, and initial analysis.

The online version of the survey was implemented at the same time as the mail survey and followed the same general approach. SurveyMonkey^® web software (SurveyMonkey.com, Palo Alto, CA) was used to send batch e-mail invitations, record survey responses in security-encrypted form, and track respondents. Three invitations to participate in the survey were made in July and Aug. 2014, with the second and third e-mail invitations sent only to previously nonresponsive firms. Firms were invited to participate in the online survey by clicking a link in the e-mail message directing them to the survey website. Respondents were then explicitly asked for consent to participate in the survey, and were given the option to decline or “opt-out,” as required by laws governing electronic communications. Consenting respondents were asked a qualifying question: “Was your company actively involved in producing and marketing ornamental plants last year (2013)?” Respondents answering this question affirmatively were then directed to proceed with the survey, whereas those answering negatively were thanked and the survey was ended. It should be noted that the online version of the questionnaire and e-mailed letters of invitation closely matched the content of the printed/mailed surveys, except for the initial qualifying question, and some additional questions on retail marketing practices, so the results are comparable. There were questions concerning methods of collecting customer demographics, CLPs, POS usage, and QR code usage that were included in the Internet version of the survey only. A question on the “advertising media used as a percentage of advertising expenses” had the following possible responses: Internet, yellow pages, radio/television, billboards, gardening publications, catalogs (print or CD), trade journals, newsletters, trade shows, social media, or other. The “other” category included a field for writing in the type of “other” media used. Newspaper was not included in the list of potential responses, so newspaper responses reported in this study are a result of respondents’ write-in responses. This may have resulted in skewed results for newspaper responses.

Results

A total of 32,000 firms were contacted for the survey by both mail and e-mail (Internet) methods. Valid responses were received from 2657 firms, including 1712 (64%) from the mail survey and 945 (36%) from the e-mail survey. A total of 299 (2.0%) of mailed surveys were returned as undeliverable, and 958 e-mail addresses were considered undeliverable. In addition, 377 firms refused to participate (“opted-out”) of the e-mail survey. After deducting the undeliverable and noncompliant firms, the overall response rate for the survey was ≈8%. A summary of general findings from the study can be found in Hodges et al. (2015).

For the purposes of this report, only the 699 firms reporting 100% retail sales (had no wholesale sales) of at least $1000 per year were evaluated, following the U.S. Department of Agriculture definition of a farm (USDA, 2016). Of those, 247 were returned via the Internet and 452 were returned by mail. Year established, employee numbers, advertising as a percentage of sales, and advertising media as a percentage of advertising expenses were asked in both the Internet and mail surveys (n = 699). Questions concerning methods of collecting customer demographics, CLPs, POS usage, and QR code usage, included in the Internet version of the survey, had 247 respondents (n = 247). For firm size comparisons, we were able to classify 273 of the 699 into four groups based on sales dollar volume: large retailers had sales ≥$250,000 (n = 32); medium retailers had $100,000 to $249,000 in sales (n = 19); small retailers had $20,000 to $99,999 in sales (n = 60), and “hobbyist” retailers had <$20,000 in sales (n = 162). All analyses were conducted using SPSS (version 22; IBM Corp., Armonk, NY). We compared those four sizes of firms using analysis of variance, comparing the different-sized firms using Dunnett’s T3 significance test.

The oldest retail firm was established in 1843, whereas the youngest was established in 2014; the average age of the retail firms was 24 years. The number of permanent employees ranged from 0 (21.0% of firms) to 600 (0.2%), with a mean of 4.46, median of 1.0, and mode of 1. The total number of seasonal employees ranged from 0 (36.4% of firms) to 150 (0.2%) with a mean of 4.17, median of 1, and mode of 0.

Estimated annual sales ranged from $1000 (cutoff point below which response was not considered sufficient for this analysis) to $80 million with a mean of $540,902, median of $12,000, and mode of $5000. Only 20 of 699 firms had online sales, with an average online sales of $1360, and a range from $140 to $160,000. Thus, a very few retail outlets were generating substantial sales revenue online.

Our first objective was to describe the percentage of sales retailers allocate to promotion and advertising, including a breakdown of media used. Some firms reported exceedingly high percentages of sales spent on advertising [n = 14 (2.5%)] possibly due to a misunderstanding of the question, skewing results, so firms reporting greater than 30% of sales expenditure on advertising were removed from only this analysis. The median expenditure as a percentage of sales on advertising was 3.6% (median 2.0%) for all retail firms responding, with 33.7% of firms responding $0.

In examining the distribution based on media type, the Internet was the most frequently listed by firms [226 firms (32.3%)] with a mean expenditure of 13.7% of total advertising dollars (Table 1). Social media was listed second most frequently [150 firms (21.5%)] with a mean expenditure of 6.4%. Newspapers were listed as the third most frequently used type of media [126 firms (18.0%)]. When examining the types of social media used by respondents, we report a percentage of social media type from all write-in responses and a percentage of social media type from all respondents (Table 2). Facebook [60.0% (from the write-ins), 17.6% (from the responses to the category)] far exceeded any other type of social media. Twitter [11.2% (from the write-ins), 3.3% (from the responses to the category)] and websites [7.3% (from the write-ins), 2.1% (from the responses to the category)] were distant second and third choices of types of social media.

Table 1.

Distribution of type of media used and average percent of advertising budget for retail firms responding (n = 699) to the 2014 Trade Flows and Marketing Survey (Hodges et al., 2015).

View Table

Table 2.

Types of social media listed as “write-in” responses (n = 205) by retail respondents to the 2014 Trade Flows and Marketing Survey (Hodges et al., 2015).

View Table

Since newspapers were not listed as an option in the survey and respondents had to write in newspapers as an “other” type of advertising media, results were potentially skewed and newspapers may have received a lower overall mean expenditure in this survey. The percent mean expenditure for the entire “other” category was 64.3%; over half of which (54.8%) was from newspaper advertising. Newsletters, radio/television, and yellow pages were used by similar numbers of firms (118, 105, and 110, respectively) with 32.9%, 33.0%, and 23.0% mean expenditure, respectively. If the digital responses were combined [Internet, social media, and digital responses from the “other category” such as websites, blogs, Craigslist (San Francisco, CA)], the total number responses from firms listing a digital type response was 397 (56.8%) of all firms. This is less than the total number of responses listing some type of print media, such as newspaper, yellow pages, newsletter, gardening publications, direct mail [609 (87.1%)]. Newsletters, yellow pages, and gardening publications were not categorized as print vs. digital, so a large percentage of these responses are potentially digital, but there is no way to distinguish digital from print in the data collected.

Our second objective was to describe the POS materials used. Almost all of the total respondents, 96.8% (239), reported the use of POS materials. Of those, 56.9% (136) made their own materials, 18.8% (45) received materials free from a supplier, 20.9% (50) purchased materials from a supplier, while 13.8% (33) purchased materials from another source.

Our third objective was to describe the use of QR or smart codes. Of the total firms analyzed here, 2.8% (7) reported using QR codes. Of those, four reported creating them and three reported that the codes were supplied. These low results are not surprising given anecdotal comments from horticultural printing companies at various trade shows.

Objective four was to gain a better understanding of how firms used CLPs. Only 48 (19.8%) reported having a CLP. Of those, 45.8% (22) used customer purchase cards, whereas 35.4% (17) used POS software.

Our fifth objective was to better understand the methods retailers use to collect customer demographics. Collection of demographic information can be helpful in developing a marketing plan. Nearly 33% of the firms collected demographic information about their customers. Of those, the method with the highest percentage use (multiple responses were permitted) was social media (50.7%) followed by CLP (48.9%), web visits (34.5%), questionnaires (15.7%), social coupons (13.5%), census data (3.9%), and marketing firms (3.1%). The Internet plays a role in many of the retail firms’ methods for collecting customer demographics.

Firm size comparisons.

Finally, we made comparisons among 262 retail-only firms (who reported the percentage of sales spent on advertising <30%) segregated by annual sales into four categories. We classified hobby retailers as having sales of $1000 to $19,999 (n = 157); small retailers had sales of $20,000 to $99,999 (n = 56); medium firms had sales between $100,000 and $249,999 (n = 19); and large firms had annual sales of ≥$250,000 (n = 30).

Hobby and small firms were ≈14 years younger than medium or large firms (Table 3). Although there was tremendous variance in the number of total, permanent, and seasonal employees, we found a difference only for seasonal employees. On average, large firms employed 12.3 seasonal employees, whereas the hobby, small-, and mid-sized firms employed half that or less. Average sales per employees was substantially higher for large firms, but hobby, small, and mid-sized firms were statistically similar.

Table 3.

Analysis of variance and comparison of 262 retail-only firms segregated by annual sales: hobby retailers had sales of $1000 to $19,999 (n = 157); small retailers had sales of $20,000 to $99,999 (n = 56); medium firms had sales between $100,000 and $249,999 (n = 19); and large firms had annual sales of ≥$250,000 (n = 30).

View Table

Examining advertising expenditures as a percentage of total sales, we found no difference among the four retail firm sizes with firms spending, on average, 3.02% to 5.06% of sales on advertising with no difference in the percentage spent on different outlets. Among the various methods of advertising expenditures, we found no differences by firm size. Companies of different sizes spent about the same percentage of advertising expenditures on yellow pages, radio/ television, billboards, gardening publications, trade journals, newsletters, trade shows, or other media.

Firms used a variety of means to capture consumer demographics. The different-sized firms used questionnaires, the U.S. census, and marketing firms similarly. However, a higher percentage of larger firms used web visits, social coupons, and social media compared with mid-sized, small-sized, and hobby retailers (Table 4). A similar percentage of different-sized firms had a CLP (Table 4). There was no difference between firm sizes in the use of POS materials, except that medium- and large-sized firms were more likely to purchase POS materials from other sources. We also found no difference by firm size for the use of QR codes.

Table 4.

Chi square analysis of 262 retail-only firms on their method of collecting customer demographics, and use of loyalty programs, point-of-sale materials, and quick response (QR) codes segregated by annual sales: hobby retailers had sales of $1000 to $19,999 (n = 157); small retailers had sales of $20,000 to $99,999 (n = 56); medium firms had sales between $100,000 and $249,999 (n = 19); and large firms had annual sales of > $250,000 (n = 30).

View Table

Discussion

Online connectivity with current/future/potential customers has revolutionized how green industry (and other) firms communicate with people. Social media outlets are potential sources of market intelligence (Bolton et al., 2013). Green industry firms reporting 100% retail sales with total sales greater than $1000/year (i.e., the group studied in this paper) are becoming aware of the power of social media and harnessing it. Of the total responses, 16.6% are using social media to collect customer demographics, which is more than any other mechanism for collecting consumer information. Web visits, another Internet source, was listed by 11.3% of respondents, bringing the total of Internet-based methods to 28.9%. These findings would logically follow from the high BCR for the Internet reported by Palma et al. (2012). But, with no difference in the allotment of advertising funds spent on different advertising media by firm size, it would appear as though firms of all sizes realize and are beginning to embrace the power of the Internet.

The Internet is also the most popular advertising medium among green industry retailers (226 responses with mean expenditure of 42.5%). Social media (150 responses with mean expenditure of 29.6%) is also strong and among social media platform; Facebook (60%) far exceeds any other platform. Purely digital forms of advertising media were used by 56.8% of firms. More traditional media (radio/television, newsletters, catalogs, yellow pages, and gardening publications) are still popular among green industry retailers and in total were used by 87.1% of firms responding. Social media is becoming a powerful marketing force. Trusov et al. (2009) showed that referrals on social network sites have substantially longer carryover effects than traditional advertising and produce substantially higher response elasticities. Overall, user-generated content (UGC), the type developed through social media, exhibits a stronger impact than marketer-generated content (MGC), more traditional advertising, on consumer purchase behavior (Goh et al., 2013). Small firms, not surprisingly, were just as likely to use social media or the Internet for advertising as large firms. However, large firms were more likely to use digital media for collecting customer demographic information. Green industry retailers have started to and should continue following marketing trends and harnessing the marketing power of social networking.

A POS display is a specialized form of sales promotion located in the retail space. It is intended to draw the customers’ attention to products or special events. POS displays can include shelf edging, display stands, mobiles, posters, and banners. Due to their appealing and informative nature, POS materials are capable of stimulating purchases directly at the point of sale. Almost all (96.8%) of green industry retail respondents use POS materials. Most of them are making their own materials (56.9%), although they also receive POS materials free from a supplier (18.8%), purchase POS materials from a supplier (20.9%), or purchase POS materials from another source (13.8%) with medium- and large-sized firms being more likely to purchase materials from other sources.

QR code content should be carefully crafted to provide meaningful, usable information for involved consumers, which is relevant, interesting, and easily passed on to other consumers (Atkinson, 2013). A very small percentage of green industry retail respondents (2.8%) reported using QR codes for retail sales with no difference by firm size. QR codes have the potential to alter consumers’ purchases at the point of purchase. When retailers offer a special promotion, they must alert the consumer to the promotion. Advertising through traditional means forces the consumer to remember the promotion and act on it later. The most important time to alert a consumer to a promotion is when they are in the store, interested in specific merchandise, but have not yet decided to purchase that merchandise. POS materials can serve this promotional function, but require printing and display—additional costs and steps. Electronic customer loyalty cards can track customer purchases and offer promotions on desirable items but discounts or promotions are applied either after the purchase decision has been made—at the cash register—or for future purchases. QR codes can target information on promotions to consumers while they are shopping, offering on-the-spot incentives or the ability to price compare in the store.

Typically, a distinction is made between push and pull mobile advertising (Bamba and Barnes, 2007; Barnes, 2002). Push advertising describes messages that are initiated by the advertiser, whereas pull advertising refers to communication of promotional material initiated by the consumer (Unni and Harmon, 2007). Push advertising is sent by the marketer to individual consumers, usually after they have completed a one-time opt in to receive such messages. They might take the form of a text message, for example, sent by an advertiser announcing a sale on a particular brand. Although the consumer has elected to receive these kinds of messages (mobile ads), they may or may not be deemed useful. Depending on various contextual factors, the mobile ad might be considered more distracting and bothersome than useful. Pull advertising, on the other hand, refers to communication that is sent by advertisers at the direct request of the consumer. The shopper might be in the bedding plant section of the garden center, for example, and notice an on-shelf promotional announcement inviting the consumer to visit a web site or download a coupon. Given the shopper’s spatial–temporal context of being in the store at the very moment they are considering bedding plants, this kind of marketing communication is usually perceived as more relevant and meaningful (Atkinson, 2013). In today’s rapidly changing technological environment, QR codes are being replaced by texting options. Retailers should therefore consider a variety of mobile-centric technologies.

A CLP is a rewards program offered by a company to customers who frequently make purchases. A loyalty program may give the customer free merchandise, rewards, coupons, or even advance released products (Kolowich, 2015). CLPs are designed to develop a relationship between a business and a consumer. The goal should be to attract the right customers and create profitable long-term relationships. According to the 2015 Colloquy Customer Loyalty Census, American households hold memberships in an average of 29 loyalty programs, but are active (meaning earn or redeem at least one per year) in only about 12 of them (Colloquy, 2015). Most CLPs are based on accumulating points or dollars spent with a reward at a preset volume of business. In fact, 73% of loyalty programs are point based [selfstartr.com; Selfstartr (2016), Jacksonville, FL]. A higher percentage of larger firms in this study used a CLP; it may take time to realize the benefits of these programs and generate the resources to offer them. Tiered programs allow the business to provide advanced rewards for larger purchase quantities. Customer purchase cards, sometimes called punch cards, have traditionally been used by bricks and mortar businesses, but are not very adaptable to e-commerce. This was the method used by about half (45.8%) of green industry retailer respondents with a CLP, with no difference by firm size.

POS software was being used to track customer purchases by 35.4% of respondents with a CLP. CLPs also offer the benefit of collecting marketing data from customers. CLPs (16.0%) were the second most frequently cited method for collecting customer demographic information of all respondents in the study. Close to 20% of respondents had a CLP, which means 80% are missing out on the opportunity developed in this type of relationship with customers.

Green industry retailers are currently using social media for marketing green industry goods. But, they have much more opportunity to use electronic media for CLPs and to begin using mobile-centric technologies to deliver in-store promotional information to consumers. This study helps document the current retail plant centers’ advertising practices and expenditures in the 21st century. It also provides a glimpse into their consumer research practices, especially the use of CLPs and methods for profiling consumer demographics.

Larger firms were more likely to use websites, social coupons, and social media. This represents an opportunity for small- to medium-sized firms who could benefit from the use of these media and more than ever have equal access to the technology as large firms.

Future research should continue to monitor the use and impact of online advertising, especially the use of social media and the ever-changing platforms for communication. Additionally, future work should monitor the changes in more traditional media (e.g., newspaper, television, radio) to determine if their use continues to decline.