Materials and methods

A quantitative content analysis was used to determine how horticultural businesses were using ODS to sell live plants to customers. This research method allows for objective and systematic quantitative description of content (Berelson, 1952; Krippendorff, 2013) often in an online environment (Ary et al., 2010).

The census sample selected for this study consisted of 498 businesses with current membership in a national association for the horticulture industry, including breeders, greenhouse and nursery growers, retailers, distributors, interior and exterior landscapers, florists, students, educators, researchers, and manufacturers (AmericanHort, 2016). The mission of AmericanHort (Columbus, OH) is to “unite, promote and advance our industry through advocacy, collaboration, connectivity, education, market development, and research” (AmericanHort, 2014). This list was selected because of the association’s focus on leadership, advocacy, and growth of the entire horticultural industry in the United States. It is common in quantitative content analyses to begin with a predetermined list (Edgar et al., 2017), particularly in an area where previous work has not been conducted or a specific industry or organization is of interest (Abrams and Meyers, 2012; Baker and King, 2016). This allows for an initial description of the population.

A codebook is the major tool used in a quantitative content analysis and consists of variables and measurement for variables within the study (Krippendorff, 2013; Riffe et al., 2005). The codebook for this study was developed by studying published research papers that examine ODS in other industries (Dholakia and Zhao, 2010; Li et al., 2011; Li et al., 2013; Ng, 2013) and was based on recommendations by Riffe et al. (2005) and Krippendorff (2013), which are works that set forth the foundation of quantitative content analysis as a method of data collection. The codebook comprised 15 variables including 1) presence on Amazon. If a business was on Amazon, researchers coded for the next variables of 2) number of products available, and details about the first available product including 3) photo quality (categorized on a two-point scale of either good or bad); 4) shipping costs; 5) shipping method; 6) approximate shipping time; 7) richness of product description (measured by the number of sentences in the description); 8) average consumer review; 9) total number of reviews; 10) presence or absence of warranty information; 11) presence of company website; 12) presence or absence of sale/discount/loyalty or reward program; 13) presence or absence of return policy; and 14) presence or absence of suggested care instructions. The final variable was 15) selling on businesses’ own website. This variable included options for selling with a fully functional shopping cart, selling with a downloadable order form emailed to the business or otherwise received electronically, and not selling online directly to the consumer. Researchers did not count selling books, carts, or other hard-good supplies. The only products counted were live plants and cut flowers. After the study was conducted, researchers created a 16th variable of USDA Rural Classifications. Zipcodes of the businesses were used to create the 16th variable using the publicly available index provided on the USDA Economic Research Service (USDA, 2013).

The codebook was reviewed by a panel of experts [authors L.M. Baker (agricultural communications specialist), C.R. Boyer (horticulture specialist), and H.H. Peterson (agricultural economics specialist)] for face content and validity (Krippendorff, 2013) before moving forward with establishing interrater reliability. Three coders (undergraduate research students at Kansas State University) were trained to use the codebook. After the training, coders used 20% of the sample to establish interrater reliability using Cohen’s kappa (Krippendorff, 2013; Riffe et al., 2005). Interrater reliability is the standard measure of reliability for a quantitative content analysis (Riffe et al., 2005). Data were collected within 24 h of one another in Spring 2016 to reduce the chance of changes to products or price. Interrater reliability was achieved on each of the 15 variables with a Cohen's kappa of 0.70 or higher as recommended by Riffe et al. (2005) and Krippendorff (2013). After interrater reliability was achieved on the first 20% (100) businesses in the study population, the three researchers divided the remaining 398 businesses and coded independently. This is the highest standard of reliability as recommended by Riffe et al. (2005) and Krippendorff (2013). Data were analyzed for descriptive statistics using IBM SPSS Statistics 20 (IBM Corp., Armonk, NY).

This exploratory study includes limitations similar to all research in a new area. When establishing the codebook, there was no previous work with this specific population on which to base the variables. Researchers used variables from previous studies on diffusion and online selling of other products as a guide. However, some variables that would have been valuable to measure with this specific industry were not included in the scope of this research. For example, not included was whether the business was retail, wholesale, or a combination. Additionally, time was a factor, so some variables that would have been interesting to explore were outside the scope of this initial benchmarking study. This research describes this specific population at the time the data were collected.

Results and discussion

Objective 1: Determine if and how ODS is used by horticultural businesses on Amazon.Of the 498 horticulture businesses, only four businesses (0.8%) were selling live plants or flowers on Amazon. Of the four businesses selling plants through Amazon, the number of products ranged from 2 to 24 with 1 to 2 photos of each product (Table 3). Shipping time ranged from 4 to 22 d with shipping costs ranging from $4.89 to $40.00. Two of the businesses selling on Amazon did not have any customer reviews, one had a 1-star review, and one had a 5-star review. None of the products were available with Prime shipping, which is Amazon’s free, guaranteed 2-d shipping method. Two businesses provided warranty information and two did not. None of these businesses had loyalty programs, sales, or discounts. Two businesses offered suggested care instructions for products, and two did not. None of the businesses mentioned a return policy.

Table 3.

Few horticultural businesses (n = 4) in the sample population [Spring 2016 membership list of a national horticulture industry association (n = 498)] were selling live plants on Amazon (Seattle, WA) at the time of data collection. Characteristics listed below describe the nature and quality of live-plant products listed for sale by association member businesses selling on Amazon. Online direct sales characteristics important to consumers are compared.

View Table

Objective 2: Determine if and how ODS is being used by horticultural businesses through businesses’ own websites. A total of 44 (8.8%) were selling on a business website through a fully functional shopping cart system, and 48 (9.6%) had an online order form that could be downloaded and emailed or submitted online in some other way (Table 4). The remaining 406 (81.5%) businesses were not selling live plants through the businesses’ website in any form.

Table 4.

Comparison of the quantity and type of online selling exhibited by horticultural businesses in the sample population [Spring 2016 membership list of a national horticulture industry association (n = 498)].

View Table

The USDA rural-urban commuting area (RUCA) system was applied to the business zip codes, which revealed 435 (87.3%) of the businesses were located in metropolitan areas, 47 (9.4%) were in micropolitan areas, 14 (2.8%) were in small towns, one (0.2%) was in a rural area, and one business was in Puerto Rico where the USDA RUCA system did not apply (Table 5). The one rural business in the current sample was not selling online; and of the 14 businesses in small towns, only one was selling online with a fully functioning shopping cart (Table 6). Two of the 14 businesses in small towns were selling online with an online order form, and 11 were not selling plants online.

Table 5.

Location of horticultural businesses in sample [Spring 2016 membership list of a national horticulture industry association (n = 498)] indicating population demographics using the U.S. Department of Agriculture (USDA) rural-urban continuum codes (RUCA) system based on zip codes (USDA, 2013). The RUCA system forms a classification scheme that distinguishes metropolitan counties by the population size of their metro area, and nonmetropolitan counties by degree of urbanization and adjacency to a metro area.

View Table

Table 6.

Comparison of the quantity and type of online selling exhibited by horticultural businesses in the sample population [Spring 2016 membership list of a national horticulture industry association (n = 498)] located in small towns (n = 14).

View Table

It is clear that a majority of horticultural businesses have not adopted ODS and that the industry is in the early stages of using this technology. While the sample of businesses selling on Amazon was too low to draw many useful conclusions, it appears that horticultural businesses are struggling to meet the current expectations of online customers. As online shopping becomes more ubiquitous, customers have high expectations when engaging in ODS, regardless of the type of product purchased.

Within the sample, many websites were not user-friendly and lacked cohesive branding and marketing strategy. The 48 businesses selling online with an order form (9.6%) made it difficult for consumers to purchase products online. Some order forms contained language such as “after submitting the form if someone has not contacted you within seven days, please contact us again,” or had to be downloaded, printed, completed, and scanned to e-mail back to the business. Moreover, some businesses required consumers to contact the company to obtain a password before viewing available plants or before making pricing available. In a world where consumers expect free 2-d shipping of products and immediate responses, this type of language can turn consumers away from purchasing products from these businesses.

As previously mentioned, businesses in this sample were all members of a prominent horticulture industry association, which presumably indicates that the owners have an interest in marketing and potentially the funds to invest in marketing of products, as association dues are not inexpensive… and yet, these businesses were not well represented on Amazon or in ODS through the businesses’ websites at the time of the study. If the large businesses with resources for marketing, like the vast majority of the businesses in this study, are struggling to sell online, it stands to reason that small, rural businesses are at a bigger disadvantage for entering the ODS market.

There are many likely reasons why horticultural businesses have not entered into the ODS market. Chief among them is a lack of knowledge and skills in online sales infrastructure, coupled with the challenge of educating consumers and keeping plants alive and in good condition upon delivery to the consumer. Shipping a plant is significantly more complex than an inorganic good such as a book. Standard packaging must fit a variety of plant sizes (or the business can specialize in a few sizes) and must be designed to prevent damage regardless of how the box is handled by the shipping carrier. Additionally, phytosanitary regulations and plant quarantines [such as for invasive fire ants (Solenopsis richteri and S. invicta)] apply to individual shipments of plants just as they do to truckloads of nursery stock across interstate shipping lanes. This may require more processing time to appropriately prepare plants for shipping and delivery across state boundaries. Anecdotal evidence suggests that consumers want instant gratification with minimal mess; they desire to go straight from the box into the garden with minimal handling. Customers likely expect a warranty and care instructions, and in this study no warranties were available for the businesses’ products and only half offered care instructions.

Managing these expectations has become an important role of the seller. Explaining to customers that plants appear differently when they are shipped in dormant form or are not available to ship yet because of growth, weather, or geographic location can be difficult when customers likely do not have a basic understanding of plant physiology. It can be challenging to recommend plants specific to the consumer’s USDA plant hardiness zone. E-commerce websites may require an algorithm to identify consumers’ location, or to add or remove plants as they become available for shipping. Phytosanitary restrictions and pest issues may also complicate ODS. Any one of these concerns could lead to “poor customer service” in a world where consumers expect free shipping with free returns.

It should be noted that since the data in this study were collected, a large national retailer, Spring Meadow Nursery (Sycamore, IL), began selling their Proven Winners plant brand on Amazon through Amazon’s “Add-on Item” program. While this indicates shipping and delivery can be done in a reasonable timeframe, challenges still exist for the majority of the industry, particularly small, rural growers.

Competing with Amazon in the online live plant marketplace requires many tactics currently used by local independent garden centers: knowledgeable customer service, activities, engagement, and a sense of community. Horticulture industry journalists argue that these tactics outweigh the many features of online shopping (Wells, 2018) and recommend strategies for addressing online plant sales (Sparks, 2018a, 2018b). A Fast Company article “The Future of Retail in the Age of Amazon” highlighted the challenges of both brick-and-mortar and online sales (Carr, 2017). The author made the case that as the size and presence of Amazon increases online, the demand for “warm and individualized service” will increase and customers will “crave the insight and personal connection of fellow humans.” Carr (2017) suggests that retailers “figure out what makes them special and use those weapons to compete.” He further argues that successful retailers in the future will deliver a satisfying in-person experience, will “challenge the fundamental assumptions of commerce,” and “resurrect the art of selling.” In the future, ODS and in-person retail will both be viable business opportunities, though each require different resources.

Learning to navigate the challenges of ODS offers the opportunity to capitalize on new-media marketing tools such as websites, social media, and e-newsletters to lessen the distance that rural businesses and customers must traverse (Barkley et al., 2007; Barnes et al., 2014; Barton and Behe, 2017). For many retailers, the risk associated with ODS (perishable goods, investing in new infrastructure, lack of the “experience” of shopping in a brick and mortar store) could be worth the reward as online businesses have the ability to charge premium prices to customers expecting specialized products matching a high-quality photo on the business website (Dholakia and Zhao, 2010; Kim and Kim, 2004). Higher profit margins, lower start-up costs, expanded customer reach, few geographical limitations, and the ability to sell 24 h daily, 7 d weekly are opportunities ideal for rural horticultural retailers. Providing a service to consumers who prefer to shop on mobile technology such as tablets and smartphones can move the horticulture industry into the modern era.

Recommendations

Since this study was the first of its kind, it provides a benchmark for horticultural businesses selling online and via Amazon at the time of data collection. There are many future avenues for research in this area that should be considered. Specifically, future research should empirically investigate specific barriers for horticultural businesses to enter the ODS market and identify specific customer needs and expectations related to ODS in the horticultural industry. As a part of a study on barriers, it would be interesting to investigate businesses that started selling on Amazon and switched to another online selling strategy, as well as how businesses are addressing phytosanitary and shipping restrictions in rural areas.

In addition, future research should investigate the requirement by certain businesses in this sample of passwords or logins to view available products. It is unclear why businesses choose to require this; perhaps it is a form of security to prevent competitors from learning details about inventory and pricing. It may also be the difficulty of managing inventory in multiple systems, or perhaps these businesses do not or are not interested in selling retail directly to consumers. Future research should investigate this phenomena and details surrounding it.

In future content analyses in this area, it would be of interest to investigate which types of businesses within horticultural industries are selling online, as this initial work categorized all businesses collectively as horticultural businesses. Moreover, future research should explore what the most common type of product sold online by horticultural businesses is, to provide a better description of current online selling efforts by the industry. Additionally, future research should consider the infrastructure of horticultural businesses for selling online and investigate challenges related to shipping and packaging. Studying ODS practices of businesses currently involved in shipping may also reveal useful information. While using the AmericanHort list for this initial sample provided a benchmark for which businesses were selling online, future research should consider sampling through keyword search terms on Amazon and Google (Mountain View, CA) or other search engines to broaden the scope of this work.