Our industry has been concerned for some time about declines in sympathy floral purchases in the US. Programs aimed at reducing "in lieu of flowers" announcements have been conducted to try to maintain or even grow the market for sympathy flowers. In this article, we model US consumer demographics that influence floral purchasing for funeral tributes in order to suggest better targets for these promotional programs. In addition, we combine this consumer trend information with US household Census projections and make some future demand projections for floral sympathy purchasing. Our projections reveal substantial opportunities for increased floral sympathy sales in the US over the next decade if the "in lieu of flowers" trend can be combated.

Since 1996, Prince & Prince has monitored the US consumer floral marketplace with its periodic nationwide survey, "Consumers Rate Their Floral Outlets". In these studies, we measured the consumer household incidence of purchasing floral products for numerous occasions, including those floral purchases for funeral tributes. By household purchasing incidence, we mean at least one floral purchase for a funeral tribute during the previous year.

In order to perform an in-depth analysis of demographic trends, we pooled purchasing incidence data across our three study periods (1996, 1998, and 2000). This provided us data from over 3,300 randomly-selected floral-buying households throughout the US. Due to the nature of our survey, our respondents tend to be floral buyers, as over 95% of our households had purchased at least one floral product over the prior year. The surveys also collected various consumer demographic measurements that are used in our analysis (Table 1).

Our methodology comprised "tree segmentation modeling", a data-mining research technique that is efficient in uncovering relationships in a database structure (see "About Our Analysis"). The output from this analysis results in a predictive tree diagram that succinctly shows how floral sympathy purchasing and household demographics are related.

The left-most box in the tree diagram (see Figure) reveals that across 3,300+ randomly-selected floral-buying households nationwide, 55% had purchased floral as a funeral tribute. This indicates that sympathy work remains a very important component of the retail floral industry. The household demographic characteristic that best distinguishes high and low sympathy purchase (from all of those listed in Table 1) is the age of the respondent. This means that the age demographic is the most important underlying consumer factor that distinguishes purchasing for funeral tributes. Households where the respondent is aged 45 -74 have the highest likelihood of sympathy purchase, whereas households where the respondent is under 25 have the lowest likelihood.

The top-third portion of our tree diagram further explores floral sympathy purchase likelihood for those with the highest likelihood, those aged 45-74. A strong regional profile emerges within this age group. Purchasing for funeral tributes is much lower in the West region than the rest of the country. This suggests a stronger cultural trend toward "in lieu of" in these Western states, as well as potential differences in funeral/memorial practices. In the two central regions, with respondents aged 45-74, sympathy purchasing is highest in rural areas. In metro areas and cities, the presence of children in the household further reduces floral sympathy purchasing. In the Northeast and South Atlantic regions, with respondents aged 45-74, education plays a role. Households were the most educated adult has a graduate or professional degree have a lower likelihood of floral sympathy purchasing than households with lesser education. This is the first of two instances in our tree diagram that suggests a high level of education leads one to support charities or causes "in lieu of flowers" as a funeral tribute. With these more educated households, we again see the rural-area effect of lifting funeral tribute purchasing above the level of those living in metro areas or cities.

For households where the respondent is aged 25-44 or those aged 75 or older (middle-third portion of tree diagram), education again further distinguishes funeral tribute purchasing. Households were the most educated adult has a graduate or professional degree have a lower likelihood of floral sympathy purchasing than households with lesser education. If our most educated households have more than one adult and income above $50,000, then funeral tribute purchasing is lifted. In our less educated households, we observed a decline in funeral tribute purchasing from other than single-male households in the year 2000 study.

Because the age demographic is key to floral sympathy purchasing, we examined projected Census trends in the age of US consumer householders, from the year 2000 through the year 2010. This provided us an understanding as to how age trends may affect future consumer demand for floral sympathy sales. Table 2 shows actual US household numbers and projections by age of householder for the discriminating age groups shown in our tree diagram (under 25, 25-44, 45-74, etc.).

For the 45-74 age group with the highest funeral-tribute purchase likelihood (61%), the predicted change in the number of these households through 2010 is a substantial 26% increase (Table 2). This compares with a 9% increase in US households overall for the same time period. This age group (45-74) is also significantly large in size, comprising about 45% of all US households in 2000 (over 50% in 2010). Thus, the relatively large age demographic shift through 2010 will likely have a favorable impact on floral sympathy purchasing by increasing the size of an already large-sized age group that tends to purchase more funeral tributes.

These US market findings on funeral-tribute purchasing and projections in household age demographics reveal substantial opportunities for florist sympathy flowers over the next decade. However, the higher purchase likelihood for this age group could be eroded by consumers' increasing support for "charity tributes" or more "in lieu of flowers" in the future. Thus, the continuing challenge for the floral industry over the next decade is to maintain or heighten the message that flowers convey in times of sorrow with the passing of a friend or family member.

For those households where the respondent is aged 25-44 and 75 or older, we predict softened demand for floral funeral-tribute purchasing, as purchase likelihood is 5% lower than the overall US average (50% vs. 55%). In addition, the number of households aged 25-44 (representing 40% of all households in 2000) is expected to decline by 10% through 2010.

For those households where the respondent is under 25, we predict weak floral funeral-tribute demand. Households in this age group have the lowest purchase likelihood for floral funeral-tribute purchasing (36%), and the number of households in this age group is predicted grow by only 3% through 2010. This age group (under 25) is also comparatively small, comprising only 5% of all US households in 2000. However, we think the projected weak demand for this age group (and for the 25-44 age group) will be more than offset by the larger projected gains in funeral-tribute purchasing of the 45-74 age segment.

This reporting has presented a model that predicts US household purchase incidence for floral funeral-tribute purchasing. A relatively large projected rise in the number of households aged 45-74 in the US over the next decade, combined with a moderately higher sympathy purchase likelihood for this segment, provides the basis for a projected increase in demand for funeral tribute purchasing.

Our research findings showed that age, education level, and region are key demographic factors that currently drive floral sympathy purchase. Educational campaigns aimed at reminding consumers of the sentiment that only flowers can express at times of grief could be targeted at these key segments and perhaps be made more efficient. Young consumers, and older consumers living in the Western US or possessing graduate or professional degrees would be prime targets for such campaigns. Since we tend to regard floral products as sophisticated, it is somewhat surprising to see a negative education effect in our tree segmentation. Apparently, a high level of education makes one more susceptible to "in lieu of" messages.

In our demand projections, a key assumption of our analysis is that purchasing patterns differ among age groups, and that the generational purchasing patterns of one age group do not carry-over unaltered into older age groups as the population matures. Research with some agricultural products show that the generational effect is small, compared with the age-group effect. Thus, as we age, we change and become more like our parents and grandparents, in tastes, habits and purchasing. However, this assumption has yet to be verified in the US floral industry with long-term studies over decades.

This marketing research also has implications for various floral suppliers that provide numerous products for the floral sympathy market. Suppliers and florists are challenged to meet this projected increase in market demand with suitable and appropriate product offerings. Retail florists may use the segmentation profiles of funeral-tribute purchasing to compare with their own market demographics and identify market opportunities for more floral sympathy sales. However, the "in lieu of" message trend may become a significant barrier to realizing the full future market potential of floral sympathy sales in the US. This phenomenon needs to be countered by our industry if we are to stem further losses in the floral sympathy market.

About Our Analysis. To identify household demographic characteristics that most influence the incidence of funeral tribute purchasing, we performed a "tree segmentation" analysis of our database. This research technique searches through our database to develop a predictive tree diagram that depicts groups of households that have high and low purchasing incidence for floral funeral tributes. The first branch in the tree diagram represents the best predictor demographic characteristic (of all those measured in Table 1) that most distinguishes high and low purchase incidence for floral funeral tributes. The overall group of households is then split on this characteristic, and the analysis repeats itself on the resulting household demographic groups. Subgroups of the demographic (e.g. age, with 7 original age category groups) are merged if they do not result in statistically different purchasing incidence. Subsequent branches in the tree diagrams represent additional predictors of purchasing incidence. Thus, all demographic splits and branches in the tree diagrams are statistically significant. In essence, our tree segmentation model identifies the highly significant household demographics that largely predict funeral tribute purchase, and reveals the priority sequence of those demographic characteristics.

The five Census regions shown in Table 1 are identified by the following states: Northeast (CT, MA, ME, NH, NJ, NY, PA, RI, VT), North Central or Mid-West (IA, IL, IN, KS, MI, MO, MN, NE, ND, OH, SD, WI), South Atlantic (DC, DE, FL, GA, MD, NC, SC, VA, WV), South Central (AL, AR, KY, LA, MS, OK, TN, TX), and West (AK, AZ, CA, CO, HI, ID, MT, NM, NV, OR, UT, WA, WY).

About the Authors. Drs. Tom and Tim Prince, formerly of The Ohio State University, are brothers and co-founders of Prince & Prince, Inc., a leading marketing research specialist in the floral and green plant industries. Prince & Prince has completed more than 50 major marketing research reports for the floral and floral-related industries in the US, and has also conducted floral marketing research in Canada, the United Kingdom, Holland, and Germany. They conceptualize, design, and implement market studies and product tests for floral and green-plant suppliers, floral importers, wholesale florists, retail florists, and floral-industry associations. For more information about their marketing research and reports for sale, visit their web site at www.FloralMarketResearch.com.