Customer Survey Results

An innovative study conducted by the Pacific Coast Farmers’ Market Association (PCFMA) has revealed important details on customer buying habits and responses to advertising by farmers’ markets.

In September 2006, PCFMA was awarded a grant from USDA’s Farmers’ Market Promotion Program for a project entitled "Growing Customers at the Farmers’ Market: Testing Marketing and Advertising Strategies for Increasing Customer Visits and Producer Sales."

The project was designed to help farmers’ market operators develop and implement more effective and cost-efficient promotional activities by testing different marketing and advertising strategies in farmers’ markets operated by PCFMA in the San Francisco Bay Area, then measuring the effectiveness of each strategy on its ability to bring customers to farmers’ markets, encourage repeat visits and increase customer purchases.

More than 4,600 customer intercept surveys were completed at eight different farmers’ markets in 2007 and 2008. The surveyed farmers’ markets were:

  • Alameda on Tuesday morning
  • Alameda on Thursday evening
  • Brentwood on Saturday morning
  • Cupertino Square on Friday morning
  • Danville on Saturday morning
  • Fairfield on Thursday evening
  • Fillmore (San Francisco) on Saturday morning
  • Jack London Square (Oakland) on Sunday morning

In addition to exploring the impact of advertising and marketing strategies, the surveys revealed attitudes and buying habits of farmers’ market shoppers, allowed for a demographic comparison of shoppers to community residents, and facilitated an estimate of the economic impact of farmers’ markets on surrounding businesses.

The four-phase project included: collection and analysis of baseline data; development and implementation of marketing and advertising strategies; collection and analysis of follow-up data; and distribution of project lessons to peer organizations.

Collection of Baseline Data

Baseline data was collected from 2,070 customers through customer intercept surveys conducted over 19 days in late July and early August of 2007. Staff from PCFMA and a temporary agency approached adult shoppers and handed them a clipboard with a survey form and pen. In rare instances staff recorded verbal responses from the shopper. Each survey included a $2 incentive coupon toward any purchase in the farmers’ market. Market managers provided estimated adult crowd counts for all survey days, suggesting a 12.3% response rate across the eight farmers’ markets in 2007.

The report summarizing the results of the baseline surveys can be downloaded here.

Implementation of Strategies

Using the results of the baseline surveys, various marketing and advertising strategies were selected for testing. In Fairfield and Brentwood, electronic media advertising was tested. For Danville and Jack London Square, print media advertising was tested. With Cupertino Square, community sponsorships were tested. In the Fillmore, mass transit advertising was tested. The Tuesday morning and Thursday evening farmers’ markets in Alameda were designated as control sites, where no new advertising or marketing strategies were tested. In each farmers’ market, advertising strategies that had been used in 2007 were repeated in 2008. For example, in the Brentwood Farmers’ Market, print advertising was purchased in the Brentwood Press in 2008 as it had been in 2007, while electronic media advertising—the test strategy for that market—was added.

Collection of Follow-up Data

Follow-up data was collected through 2,541 customer intercept surveys in the same eight farmers’ markets over 19 days exactly one year later in 2008. Using adult crowd count estimates provided by the market managers for the eight survey days, the response rate in 2008 was estimated at 13.7%. The surveys were only available in English and Spanish, but Spanish response was low. Those without English proficiency likely had a higher refusal rate which may have contributed to the observed under-representation of some minority groups.

Those who were shopping together were allowed to complete separate surveys, though in many instances one member of the group would suggest that the other complete the survey. This too could have impacted the survey results as each response was analyzed as a separate shopping party.

The survey process and instrument used both years was virtually identical to allow pooling and comparison of data. It was composed of 18 to 21 questions concerning customer behavior, marketing and demographics. (Further information on the demographic comparison is on page 7.)

Distribution of Project Results

A publication summarizing the results of the baseline surveys was developed and distributed in January 2008. It is available online at This project final report was completed in January 2009, providing a summary of the data from the follow-up surveys with comparative data from the baseline surveys.

The final report, summarizing the results of the follow-up surveys can be downloaded here.