Secondary Sources Of Information

 “A few months of field work collecting data can save several hours in the library”

Secondary data (or secondary information) is information that has been collected by persons or agencies for purposes other than the solution of the marketing research problem at hand. These data may have been collected from sources within the researcher’s firm or from sources outside the firm. The key point is that the data were collected for some other project, or reason, than the current one.

In contrast, primary data is data collected for the researcher’s current research project. Primary data is often collected from a respondent, an individual who provides information either passively through the observation of his or her behavior, or actively through verbal response. Researchers using primary data must be concerned with information obtained by asking questions, by observing behavior or by examining the results of past behavior.

In addition to primary and secondary data, there exists commercial data sold in the form of syndicated services. These data are collected by commercial marketing research firms or industry associations and, as such, have characteristics of both primary and secondary data. Since these data relate to ongoing concerns of a marketer they can be viewed as primary data.

However, the commercial agency did not design its service solely to provide information for one company’s specific project. Thus, there are elements of secondary data. It should be clear that distinctions between primary and secondary commercial data may be minimal

In this chapter, we discuss the reasons for obtaining secondary information, types of secondary information, sources of external secondary data, and syndicated services that provide commercial data. Data, in all their forms, are the heart of research. Secondary research can help provide a clearer picture of a problem so that researchers and managers can make the necessary critical decisions.

Reasons For Obtaining Secondary Information

As a general rule, no research project should be conducted without a search of secondary information sources. This search should be conducted early in the problem investigation stage and prior to any organized collection of information from primary sources. There are several reasons for this.

Secondary Information May Solve the Problem

If adequate data are available from secondary sources, primary data collection will not be required. For example, Campbell Soup Co. based a long running advertising campaign on the theme “soup is good food.” This theme emerged from federal government data pertaining to eating habits, nutritional health, and related topics collected over a period of 15 years. 

1959 : Campbell's Soup

Secondary Information Search Costs Substantially Less

Comprehensive search of secondary sources can almost always be made in a fraction of the time and cost required for the collection of primary information. This is particularly true today with online access to research publications and databases. Searching for secondary research helps you avoid duplicating primary research and optimizes research expenditures by acquiring only information that cannot be found else where. Many marketing problems do not warrant expenditures for primary information collection, but are worth the time and cost of secondary information.

Secondary Information Has Important Supplementary Uses

When secondary information cannot solve the research problem, it can provide invaluable supplemental uses  :

1. Defining the problem and formulating hypotheses about its solution. The analysis of available secondary data will almost always provide a better understanding of the problem and its context, and will frequently suggest solutions not considered previously.
2. Planning the collection of primary data. An examination of the methods and techniques employed by other investigators in similar studies may be useful in planning the present one. It may also be of value in establishing classifications that are compatible with past studies so that trends may be more readily analyzed.
3. Defining the population and selecting the sample. Past information and samples may help establish classifications for current primary information collection.

The researcher must be careful when using only secondary data. To be useful, secondary data must be available, relevant to the information needs (which includes being timely), accurate, and sufficient to meet data requirements for the problem at hand. Often, little is known about the reliability of secondary research studies. It is important that the researcher know how the secondary data being considered for use were collected, if the data is reliable, and if the right techniques were used.

Example : One company wants to do a segmentation analysis on foreign markets with a particular emphasis on examining demographics (Albaum, Duerr, & Strandskov, 2005, Chap. 5). The company is considering using the official government census of the population. However, they become aware that the data are not available from all markets in equal quantity, aggregation, and detail, and the reliability of data is not the same.

What one gets from a census depends on what was on the census form in the first place typically a mixture of traditional questions and new items of interest to public policy makers and civil servants at the time. Some countries publish information about noncitizens, and others collect data on religion—both of these topics are ignored in U.S. Censuses. Income is one of the major dimensions of U.S. segmentation research, but many highly developed nations ignore the income question in their censuses.

In short, one cannot always expect to find the same range of data topics that you are interested in. More over, the data may not use the same categories when showing relevant distributions of demographic variables, such as age. In short, comparability and equivalence issues arise, and these can hinder your effectiveness in using secondary data.

Types Of Secondary Information

Secondary information falls into two categories, data that is available within the company (internal data) and that which must be obtained from outside sources (external data).

Internal Secondary Information

All companies collect information in the everyday course of conducting business. Orders are received and filled, costs are recorded, warranty cards are returned, sales people’s reports are submitted, engineering reports are made—all are collected for other purposes, but may be useful to the researcher (Andreasen, 1988, pp. 77–89). The key, of course is knowing where they are and how to access them. In order to do this efficiently, the firm must have an effective marketing information system.

Example : Spectra Physics Lasers Division (producing laser grocery store scanners) regularly performs customer satisfaction studies. Where as these studies are primary research to the Retail Systems organization, they are internal secondary information to other divisions that may want to look at them. Also, they can be secondary data to Retail Systems should they be used at a much later date for aiding in decision making or for purposes other than those originally intended when the studies were done.

External Secondary Information

External secondary information is available in staggering assortments and volumes. It also is applicable to all of the major types of marketing research projects and is mainly concerned with the noncontrollable aspects of the problem :

• Total market size
• Market characteristics
• Competitor products, prices, promotional efforts, and distribution methods

As an example, a consumer goods company is considering whether it should establish a direct selling operation. Direct selling is defined as personal contact between a sales person and a consumer away from a fixed business location such as a retail store. The Direct Selling Association (DSA) provides secondary information in the form of a regular survey of the industry. Some types of information on industry statistics and sales force demographics are available on a regular basis. Some of the types of industry statistics are the following :

• Estimated U.S. sales
• Estimated U.S. salespeople
• Percent of sales by major product groups
• Location of sales
• Percent of sales by census region
• Sales strategy
• Compensation structure by percent of firms
• Compensation structure by percent of sales dollars
• Compensation structure by percent of sales people

The following are some of the sales force demographics available :

• Gender
• Age Education
• Independent contractor/employee status
• Hours per week dedicated to direct selling
• Average time spent on direct selling tasks
• Main reasons for becoming a direct sales representative 
• Percent of sales people by distributor ship type

Age, education, average time spent on direct selling tasks, and main reason for becoming a direct sales representative data are from DSA’s National Sales force Survey. (Direct Selling Association, 2003;)

Sources Of External Secondary Data

The major original sources of external secondary information are :

1. Government (supranational, federal, state, and local)

2. Trade associations and trade press

3. Periodicals and professional journals

4. Institutions (e.g., universities)

5. Commercial services

Government Data Sources

The federal government is by far the largest single source of this type of data. Both governmental and trade sources are so important that the experienced researcher will be thoroughly familiar with them in his or her field of specialization. Periodicals and research publications of universities and research institutes frequently provide valuable information. Commercial services of many types are available that are highly useful for specific research problems.

Market performance studies on consumer products, for example, will normally provide such demographic information as the number of consumers (or consuming units) by age group, income class, gender, and geographic area. Such data are usually available on a reasonably recent basis from censuses conducted by federal, state, local, and, when needed, supranational governments.

Often, a good first source is the Statistical Abstract of the United States, available online from the Bureau of the Census http://www.census.gov/compendia/statab/. This reference abstracts data from original reports and gives some useful material on social, political, and economic matters. The source is a good reference to the more detailed data in the original sources.

The State and Metropolitan Area Data Book is a publication of the Bureau of the Census that is available online in PDF format. It provides detailed comparative data on states, metropolitan areas and their component counties, and central cities. It covers information about numerous topics relevant for both B2C and B2B marketing, including population, income, labor force, commercial office space, banking, health care, housing, and so forth. 

The Census of Population http://www.census.gov/population/www/ and the Census of Housing http://www.census.gov/hhes/www/housing.html taken by the U.S. Department of Commerce every 10 years are the most comprehensive of such censuses. Updates of various census measurements based on smaller yearly surveys are available in Current Population Reports  and Current Construction and Housing Reports Many other up-to-date estimates are made periodically by governmental and non–governmental agencies.

Data from the U.S. Census Bureau is available online for custom data analysis, on CDROM, and in report form as downloadable PDF files. There are, however, private companies that make such data available for a fee in more processed form, which, in effect, adds value to the Census Bureau data. The company previously mentioned, GeoLytics (http://geolytics.com/), markets a line of census data products and a variety of custom data retrieval services.

􀁸 Demographic reports (and maps)

• Custom data sets and reports Area segmentation
• Area to area correspondence files
• Banking and realtor tract level data and maps

􀁸 Services

• Geocoding (GPS) addresses
• Custom-built databases
• Normalized data for across census comparisons

Other companies include census data in mapping software that is used for geographic market analysis. This type of software is potentially useful for such applications as retail site analysis, real estate site reports, direct marketing, database creation, and so forth. One supplier is Scan/US, Inc. (http://www.scanus.com ), whose software product Scan/US Streets and Data U.S.A. includes maps for the entire United States that include all types of demographics.

Private Data Sources

Private organizations are another source of demographic information useful to marketers. To illustrate, SRDS  publishes The Lifestyle Market Analyst. This annual provides demographic and lifestyle information for 210 Designated Market Areas (DMAs) in the United States. As shown in Table 3.1 and Figure 3.1, this market data can be accessed in graphic and tabular formats for demographic and lifestyle variables:

• Demographic categories for each DMA: Start with a specific demographic segment, such as dualincome households, and identify lifestyles and geographic locations.
• Most popular lifestyles for each DMA: Specify a lifestyle and then identify what other interests
• frequently appeal to those consumers and what demographic information corresponds to that profile.

Nielsen’s Claritas division, a provider of solutions for geographic, lifestyle and behavioral target marketing, has developed a demographic widget that is available as a free download for personal electronics.

Market size studies (e.g., size in sales dollars or units) often are conducted by trade associations, media, firms in the industry, and private research organizations. These studies are published and made available to interested parties. Industry type studies may be concerned with such types of information as total market size, market characteristics, market segments and their size and characteristics, and similar types of information.

Example: Mediamark Research, Inc. conducts a single source continuing survey, primarily aimed at the advertising industry that provides demographics, lifestyles, product usage, and exposure to all advertising media data. One part of this study is a series of studies on specific products/services that is published as syndicated reports. 

Information on new products and processes is available from such sources as patent disclosures, trade journals, competitors’ catalogs, testing agencies, and the reports of governmental agencies, such as the Food and Drug Administration, the Department of Agriculture, and the National Bureau of Standards.

Starch Readership Reports

The best way to create print ads for the future, and for the long term, is to get feedback on a constant basis in order to find out what works and what doesn’t.

Each year, Starch measures over 25,000 ads in over 400 magazine issues. On the most basic level you get raw readership scores the percent of readers who saw the ad and read the copy. Then the data are put

into a context: The ad is ranked not only against other ads in the issue but also against other ads in its product category over the past two years. These norms are a fast and easy way to judge the performance of your ad over time and against the competition.

The Benefits of Starch Ad Readership 

In Depth Analysis

Campaign analyses inform clients not only about the scores of the ads but also why they performed as they did and what can be done to improve the ads. Moreover Starch also is unique in its ability to tell clients about the best positions in various publications (e.g., whether farforward positioning is superior to ads in the back of the book).

Extra Questions

• To give you information on advertising like ability, persuasiveness and intent to purchase
• Many times, if you ask a publisher to Starch an issue your ads will appear in, they will assume the cost and pass on the data to you for free

The Starch Ad Readership Program

Through the Book, Recognition Method

• One-to-one in-person interview
• Generally, 100–200 sample, but can be more if client desires
• Sample approximates readership of publication, but is not representative
• Reports present data on 

            o Noted: percent who saw any part of the ad

            o Associated: percent who saw advertiser’s name

            o Read Some: percent who read any of the copy

            o Read Most: percent who read more than half the copy

• Most reports also offer indexed scores, based on ads of the same size, color, product category

Internet Databases

The Internet has become the staple of research and provides access to most commercial electronic databases. Thousands of such databases are available from numerous subscription systems, such as DIALOG (http://www.dialog.com ), LexisNexis (http://www.lexisnexis.com/), or Dow Jones News/Retrieval http://www.dowjones.com/Products_Services/ElectronicPublishing/EnterpriseMedia.htm.

In general, there are five categories of commercial databases :

1. Bibliographic databases that index publications
2. Financial databases with detailed information about companies
3. Statistical databases of demographic, econometric, and other numeric data for forecasting and doing projections
4. Directories and encyclopedias offering factual information about people, companies, and organizations
5. Full text databases from which an entire document can be printed out.

The advantages of such current databases are obvious. All that is needed is personal computer with internet access or a CD-ROM/DVD.

Computerized databases have led to an expanded focus on database marketing. Database marketing has been defined as an extension and refinement of traditional direct marketing, but uses databases to target direct response advertising efforts and tracks response and/or transactions. In database marketing, the marketer identifies behavioral, demographic, psychographic, sociological, attitudinal, and other information on individual consumers/households that are already served or that are potential customers. Data may come from secondary and/or primary sources. Qualtrics clients are increasingly APIs (Application Programming Interface) to link and integrate customer databases with survey data and respondent panels. APIs can be used to link and integrate data from multiple sources in real time. Thus, information in database profiles is augmented by new contact and survey data, and can be viewed in dashboards that report current information and can be used to better target and predict market response. Databases can be used to estimate market size, find segments and niches for specialized offerings, and even view current customer use and spending (Morgan, 2001). In short, it helps the marketer develop more specific, effective, and efficient marketing programs.

Today, data mining is in high demand as a research focus. Data mining involves sifting through large volumes of data to discover patterns and relationships involving a company’s products and customers. Viewed as a complement to more traditional statistical techniques of analysis, two of the more powerful data mining techniques are neural networks and decision trees (Garver, 2002). Further discussion of data mining techniques is beyond the scope of this text, but good discussions are found in Berry & Linoff (1997, 2002) and Dehmater & Hancock (2001).

Elkind and Kassel (1995) provided essential guidelines for attaining market knowledge from online sources :

• Develop an online research plan. The plan will outline all the key areas of inquiry and will provide a systematic pathway to search, retrieve, and arrive at the desired data, information, and knowledge.
• Clearly define your information needs, knowledge gaps, and issue to be resolved. One of the best  ways is to do a knowledge inventory or review to determine what you already know or have in your possession, both primary and secondary research.
• Focus the search. Start by applying the learning from your knowledge inventory and specify the new areas that are critical to your project. The focus can be further enhanced by specifying key hypotheses regarding possible findings, information categories relevant to the issue, and other criteria such as product categories, consumer targets, market areas, time frames, and so on.
• Search across multiple sources. Don’t expect to find what you need in single pieces of data or sources of information. You only rarely will find what you need in one place
• Integrate information from the multiple sources. Use techniques of trend analysis to identify patterns that emerge when various information elements are combined; for example, content analysis, stakeholder analysis, paradigm shift, trendlines, critical path analysis, sector analysis (technological, social/cultural, occupational, political, economic), or other analytic techniques that facilitate integration of diverse data and information and identification of underlying patterns.
• Search for databases that contain analyses rather than limiting the search to just data or information. Many of the professional online database producers and vendors offer thousands of full text articles and resources that contain analyses. You may be able to find material that already provides some interpretation that may be helpful.
• Enhance the robustness of your data or information through multiple source validation. You can increase confidence in the validity of the findings of your secondary searches by looking for redundant patterns that cut across different sources and studies

Syndicated Service

Some of the aforementioned commercial services are examples of what are called syndicated services. Research organizations providing such services collect and tabulate specialized types of marketing information on a continuing basis for purposes of sale to a large number of firms. In general, syndicated data are made available to all who wish to subscribe. Reports are made available on a regular basis (for example, weekly, monthly, quarterly). Since these data are not collected for a particular firm for use in a specific research problem situation, they can properly be viewed as secondary data. Syndicated services are widely used in such areas as movement of consumer products through retail outlets, direct measures of consumer purchases, social trends and lifestyles, and media readership, viewing, and listening.

The syndicated Survey of American Consumers (based on surveying more than 25,000 adults) by Mediamark Research, Inc. provides an illustration of syndicated services. This survey provides data useful for detecting a marketer’s best prospects by providing answers to such questions as the following:

• How many customers are there for the products or services we market? Is the size of the market growing? Stabilizing? Or shrinking?
• Who are the customers? How old are they? What do they earn? Where do they live?
• How do customers differ in terms of how often and how much they buy? Who are the heaviest purchasers of the product?
• What brands are customers buying? How have shares of the market changed? What differences are there among brand buyers?
• What’s the best way of reaching prospects? Which media vehicles and formats are most efficient in delivering the message to the customer?

Mediamark is able to profile American consumers on the basis of more than 60 demographic characteristics and covers usage of some 500 product categories and services and 6,000 brands.

Types of Syndicated Services

Syndicated data may be obtained by personal interviews, direct observation, self reporting and observation, or use of certain types of mechanical reporting or measuring devices. One of the most widely used approaches is the continuous panel, which refers to a sample of individuals, households, or firms from whom information is obtained at successive time periods. Continuous panels are commonly used for the following purposes:

1. As consumer purchase panels, which record purchases in a consumer diary and submit them periodically.
2. As advertising audience panels, which record programs viewed, programs listened to, and publications read.
3. As dealer panels, which are used to provide information on levels of inventory, sales, and prices.

Such panels have been established by many different organizations, including the federal government, various universities, newspapers, manufacturers, and marketing research firms. These types of panels furnish information on at regular intervals on continuing purchases of the products covered.

For example, typical consumer panels might report the type of product purchased by brand, weight or quantity of unit, number of units, the kind of package or container, price per unit, whether a special promotion was in effect, store name, and date and day of week of purchase. Data are recorded in diaries, either online or are mailed in each month.

One of the largest consumer panels is maintained by NPD Research This panel comprises 13,000 families and is national in coverage. NPD also maintains self-contained panels in 29 local markets. 

Advertising audience panels are undoubtedly more widely publicized than other panels. It is from these panels that television and radio program ratings are derived. These panels are operated by independent research agencies rather than the media both for reasons of economy and to avoid any question of partisanship.

For example, ACNielsen uses a metering device that provides information on what TV shows are being watched, how many households are watching, and which family members are watching. The type of activity is recorded automatically; household members merely have to indicate their presence by pressing a button. The sample is about 5,000 households. In local markets, the sample may be 300 to 400 households.

Single source data tracks TV viewing and product purchase information from the same household. Mediamark’s national survey and IRI’s Behavior Scan are examples of such single source data. The single source concept was developed for manufacturers who wanted comprehensive information about brand sales and share, retail prices, consumer and trade promotion activity, TV viewing, and household purchases.

The information obtained from the types of syndicated services described previously has many applications. The changes in level of sales to consumers may be analyzed directly without the problem of determining changes in inventory levels in the distribution channel. Trends and shifts in market composition may be analyzed both by type of consumer and by geographic areas. A continuing analysis of brand position may be made for all brands of the product class. Analyses of trends of sales by package or container types may be made. The relative importance of types of retail outlets may be determined. Trends in competitor pricing and special promotions and their effects can be analyzed along with the effects of the manufacturer’s own price and promotional changes. Heavy purchasers may be identified and their associated characteristics determined. Similarly, innovative buyers may be identified for new products and an analysis of their characteristics made to aid in the prediction of the growth of sales. Brand switching and brand loyalty studies may be made on a continuing basis. One reported use of this syndicated service has been to design products for specific segments.

The products from syndicated services are continually changing with client needs and new technological opportunities.

Summary

This chapter has been concerned with secondary information and sources of such information. We started with some reasons why secondary information is essential to most marketing research projects. Then, various sources and types of secondary information internal and external were discussed in some depth. Also given more than cursory treatment was syndicated data, a major type of service provided by commercial agencies.

Rerferences

ACNielsen: http://en-us.nielsen.com

ACNielsen – Claritas Demographic Widget, http://www.claritas.com/target-marketing/nielsenclaritas-

demographic-widget.jsp

Albaum, G., Duerr, E., & Strandskov, J. (2005). International marketing and export management

(5th ed.). Harlow, UK: Pearson Education.

Andreasen, A. (1988). Cheap but good marketing research. Homewood, IL: Dow-Jones Irwin.

Berry, M., & Linoff, G. (1997). Data mining techniques: For marketing, sales, and customer

support. New York: Wiley.

Berry, M., & Linoff, G. (2002). Mining the Web: Transforming customer data. New York:

Wiley.

Campbells 1959 attribute commercial http://www.youtube.com/watch?v=qMJtLP8jMWQ

Campbells 1985 Soup is good food commercial http://www.youtube.com/watch?v=4PtZYlcyXfI

Dehmater, R., & Hancock, M. (2001). Data mining explained: A manager’s guide to customer

centric business intelligence. Burlington, MA: Digital Press

DIALOG http://www.dialog.com

Direct Selling Association, 2003 http://www.dsa.org/pubs/

Dow Jones News/Retrieval

http://www.dowjones.com/Products_Services/ElectronicPublishing/EnterpriseMedia.htm

Elkind, F., & Kassel, A. (1995, July 31). A marketer’s guide for navigating the information

superhighway. Marketing News, 29, 2ff.

Garver, M. S. (2002, September 16). Try new data mining techniques. Marketing News, 36, 31–

33.  

IRI: infores.com 

LexisNexis http://www.lexisnexis.com

Morgan, M. S. (2001, October 8). Research boosts database’s power. Marketing News, 35, 16.

NFO: mysurveys.com

NPD Research http://www.npd.com

NPD Research, Inc. http://www.npdor.com

RoperASW. (2002). Available at RoperASW.com. Accessed January 9, 2002.

SRDS Life Style Market Analyst, http://www.srds.com/frontMatter/ips/lifestyle/index.html

SRDS Graphical Reports,

http://www.srds.com/frontMatter/ips/lifestyle/images/LMAA_overview-with-map.jpg

Synovate: synovate.com

U.S. Census Bureau, Statistical Abstract, http://www.census.gov/compendia/statab/

U.S. Census of Population, http://www.census.gov/population/www/

U.S. Census of Housing, http://www.census.gov/hhes/www/housing.html

U.S. Current Population Reports, http://www.census.gov/main/www/cprs.html

U.S. Current Construction and Housing Reports, http://www.census.gov/prod/www/abs/conshou.

html