Ch8 Measuring Market Opportunities Forecasting and Market Research

Ch8 Measuring Market Opportunities Forecasting and Market Research 8.1: Every Forecast Is Wrong! Some forecasts turn out too high, others too low. Forecasting is an inherently difficult task, because no one has a perfect crystal ball. The future is inherently uncertain, especially in today’s rapidly changing markets. Some forecasts are based on extensive research, others on small-scale inquiries, still others on uniformed hunches. Forecasting plays a central role in all kinds of planning and budgeting in all kinds of businesses and other organizations. 8.2: A Forecaster’s Toolkit: A Tool for Every Forecasting Setting What is to be estimated or forecasted: First, there’s the size of the potential market , the likely demand from all actual and potential buyers of a product or product values. An estimate of market potential often serves as a starting point for preparing sales forecast. The size of the currently penetrated market, those actually using the product at the time of the forecast. Investors will also want to know these figures- the size of the potential and penetrated markets for the market segments the company intended to serve, their target market. Established organizations employ two broad approaches for preparing a sales forecast: top-down and bottom-up. Under the top-down approach, a central person or persons takes the responsibility for forecasting and prepare an overall forecast. Under the bottom-up approach, each part of the firm prepares its own sales forecast, and the parts are aggregated to create the forecast for the firm as a whole. Using the bottom-up approach presents numerous advantages. First, this approach will force them to think clearly about the drivers of demand for each market segment or product line, and better understand the real potential of their business and its part. Second, they will be forced to explicit assumptions about the drivers of demand assumptions they can debate- and support evidence gathered from their research- with prospective investors and which they can later verify as the business unfolds. Third, such an approach facilitates “what if planning”. Various combinations of market segments and/ or product lines can be combined to build a business plan that looks viable. There are six major evidence-based methods for estimating market potential and forecasting sales: Statistical and Other Quantitative Methods Statistical methods use past history and various statistical techniques, such as multiple regression or time series analysis, to forecast the future based on an extrapolation of the past. Disadvantages: This method is typically not useful for ACG or other entrepreneurs or new product managers charged with forecasting sales for a new product or new business. There is no history in their venture on which to base a statistical forecast. Advantages: In established firms, for established products, statistical methods are extremely useful. When Michelin, the tire maker, wants to forecast demand for the replacement automobile tire market in Asia for the next year, it can build a statistical model using such factors as the number and age of vehicles currently on the road in Asia, predictions of GDP for the region, the last few years’ demand, and other relevant factors to forecast market potential as well as Michelin’s own replacement tire sales for the coming year. Such a procedure is likely to result in a more accurate forecast than other methods, especially if Michelin has years of experience with which to calibrate its statistical model. Limitation 1: statistical methods generally assume the future will look very much like the past. Limitation 2: if product or market characteristics change, statistical models used without adequate judgment may not keep pace. Observation Observe or gather existing data about what real consumers do in the product-market of interest. Observation-based forecasting: is attractive because it is based on what people do. If behavioral or usage data can be found form existing secondary sources- in company files, at the library, or on the Internet- data collection is both faster and cheaper than if a new study conducted must be designed and carried out. For new-to-the world products, observation is typically not possible and secondary data are not available, since the product does not exist, except in concept form. Surveys Survey of buyers’ intentions: buyers can also be asked about their current buying behavior: what they currently buy, how often, or how much they use. The salespeople can be asked how much they are likely to sell, completing a survey of salesforce opinion. Experts of various kinds – members of the distribution channel, suppliers, consultants, trade association executives, and so on – can also be surveyed. Surveys possess important limitations, however. For one, what people say is not always what people do. Consumer surveys of buyer intention are always heavily discounted to allow for this fact. Second, the persons who are surveyed may not be knowledgeable, but if asked for their opinion, they will probably provide it! Third, what people imagine about a product concept in a survey may not be what is actually delivered once the product is launched. statistical and observational methods, where adequate data or settings are available in which to apply them, are superior to survey methods of forecasting, because such methods are based, at least in part, on what people have actually done or bought While survey methods (Are you likely to buy replacement tires this year? How often are you likely to use a pay phone?) Are based on what people say a less reliable indicator of their future behavior. Analogy An approach often used for new product forecasting where neither statistical methods nor observations are possible is to forecast the sales or market potential for a new product or product class by analogy. Under this method, the product is compared with similar products for which historical data are available. Forecasters consider related product introductions with which the new product may be compared. Early forecasts for high-definition television (HDTV) were done this way, comparing HDTV with historical penetration patterns for color TV, videocassette recorders (VCRs), camcorders, and other consumer electronic products. Limitation 1: the new product is never exactly like that to which analogy is drawn. Limitation 2: market and competitive conditions may differ considerably from when the analogous product was launched. Judgment Sometimes forecasts made solely on the basis of experience judgment, or intuition. Some decision makers are intuitive in their decision processes and cannot always articulate the basis for their judgments. Those with sufficient forecasting experience in a market they know well may be quite accurate in their intuitive forecasts. Unfortunately, it is often difficult for them to defend their forecasts against those prepared by evidence-based methods when they differ. The importance of experience judgment in forecasting, whether it is used solely and intuitively or in a concert with evidence-based methods, cannot be discounted. Market Tests Usually largely for new products, market tests such as experimental market tests may be done under controlled experimental conditions in research laboratories, or in live test markets with real advertising and promotion and distribution in stores. Use of test markets has decline over the past two decades for two reason: First, they are expensive to conduct because significant quantities of the new product must be produced and marketing activities of various kinds must be paid for. Competitors can engage in marketing tactics to mislead the company conducting the test, by increasing sampling programs, offering deep discounts or buy-one-get-one-free promotions, or otherwise distorting normal purchasing patterns in the category. The coming of the internet has made possible a new kind of market test: an offer directly to consumers on the web. Offers to chat rooms, interest groups, or email lists of current customers are approaches that have been tried. Use of such techniques has increase, due to companies’ ability to carry out such tests quickly and at low cost. SWAG : a not evidenced-based method (silly wild Guess) Mathematical Entailed in Forecasting The ultimate purpose of the forecasting exercise is to end up with numbers that reflect what the forecaster believes is the most likely outcome, or sometimes a range of outcomes under different assumptions, in terms of future market potential or for the sales of a product or product line. The combination of judgment and other methods often leads to the use of either of two mathematical approaches to determine the ultimate numbers: the chain ratio calculation or the use of indices. Both mathematical approaches begin with an estimate of market potential (the number of households in the target market The market potential is then multiplied by various fractional factors that, taken together, predict the portion of the overall market potential that one firm or product can expect to obtain. 8.3: Cautions and Caveats in Forecasting Keys to Good Forecasting There are two important keys to improve the credibility and accuracy of forecasts of sales and market potential. The first of these is to make explicit the assumptions on which the forecast is based. This way, if there is debate or doubt about the forecast, the assumptions can be debated, and data to support the assumptions can be obtained. The resulting conversation is far more useful than stating mere opinions about whether the forecast is too high or too low. The second key to effective forecasting is to use multiple methods. When forecasts obtained by different methods converge near a common figure, greater confidence can be placed in that figure. Where forecasts obtained by multiple methods diverge, the assumptions inherent in each can be examined to determine which set of assumptions can best be trusted. Ultimately, however, any forecast is almost certainly wrong. Contingency plans should be developed to cope with the reality that ultimately unfolds Biases in Forecasting Several sources of potential bias in forecasts should be recognized. First, forecasters are subject to anchoring bias, where forecasts are perhaps inappropriately ‘anchored’ in recent historical figures, even though market conditions have markedly changed, for better or worse. Second, capacity constraints are sometimes misinterpreted as forecasts. Someone planning to open a car wash that can process one car every seven minutes would probably be amiss in assuming sufficient demand to actually run at that rate all the time. Another source of bias in forecasting is incentive pay. Bonus plans can cause managers to artificially inflate or deflate forecasts, whether intentionally or otherwise. ‘Sandbagging’ – setting the forecast or target at an easily achievable figure in order to earn bonuses when that figure is beaten – is common. Finally, unstated but implicit assumptions can overstate a well-intentioned forecast. Assumptions of awareness and distribution coverage at levels less than 100 per cent, depending on the nature of the planned marketing programme for the product, should be applied to such a forecast, using the chain ratio method. 8.4: Why Data? Why Marketing Research? For gaining a better understanding of market and competitive conditions and of what buyers in a given market want and need – what we call market knowledge. Obtaining market knowledge also requires data, and so far we’ve provided little discussion of exactly how one might best find the necessary data. Market knowledge is generally incomplete and often ill-informed, based perhaps on hunches or intuition that may or may not be correct. Without adequate market knowledge, marketing decisions are likely to be misguided. Products for which there is little demand may be introduced, only to subsequently fail. New markets may be entered, despite market or industry conditions that make success unlikely. Attractive product-markets may be overlooked. Products may be marketed to the wrong target market, when consumers in another market segment would like the product better. Pricing may be too high, reducing sales, or too low, leaving money on the table. Advertising and promotion monies may be poorly spent. Second-best distribution channels may be chosen. These outcomes are all too common. Thoughtfully designed, competently executed marketing research can mitigate the chances of such unpleasant outcomes. 8.5: Market Knowledge Systems: Charting a Path toward Competitive Advantage There are four commonly used market knowledge systems on which companies rely to keep pace with daily developments: internal records regarding marketing performance in terms of sales and the effectiveness and efficiency of marketing programmes, marketing databases, competitive intelligence systems, Systems to organize client contact. Taken together, these systems lie at the heart of the systematic practice of customer relationship management (CRM). Effective use of CRM is likely to result in happier, higher volume, more loyal customers. Internal Records Systems Additional reports aggregate sales information by style and color; by merchandise category (e.g., dress or casual); store, area, or region; and for various time periods. The information provided by these reports constitutes the backbone of Nine West’s decision making about which shoes to offer in which of its stores. Every marketer, not just retailers, needs information about ‘what’s hot, what’s not.’ Unfortunately, accounting systems generally do not collect such data. Typically, such systems just track revenue, with no information about which goods or services were sold. Thus, marketers need internal records systems to track what is selling, how fast, in which locations, to which customers, and so on. Providing input on the design of such systems so that the right data are provided to the right people at the right time is a critical marketing responsibility in any company. The salesforce, too, needs information about status of current orders, customer purchasing history, and so on. For those charged with developing or updating internal record systems in their companies, we provide a series of questions to help marketing decision makers specify what internally generated sales data are needed, when, for whom, in what sequence, at what level of aggregation. Information varies from company to company and industry to industry. Exhibit 8.5 Designing an internal records system for marketing decision makers Questions to ask Implications for a chain footwear retailer Implications for an infomercial marketer of kitchen gadgets What information is key to providing our customers with what they want? Need to know which shoes sell, in which stores and markets, at what rate Need to know which gadgets sell, in what markets, at what rate What regular marketing decisions are critical to our profitability? Decide which shoes and shoe categories to buy more of, which to buy less of or get rid of, in which stores and markets to sell them Decide on which specific TV stations, programmes, and times of day to place infomercials for which gadgets What data are critical to managing profitability? Inventory turnover and gross margin Contribution margin (gross margin less media cost) per gadget sold Who needs to know? Buyers and managers of merchandise categories Media buyers, product managers When do they need to know, for competitive advantage? For hottest sellers, need to know before competitors, to beat them to the reorder market. For dogs, need to know weekly, to mark them down. Need to know daily, for prior night’s ads, to reallocate media dollars In what sequence and at what level of aggregation should data be reported? Sequence of report: hot sellers first, in order of inventory turnover Sequence of report: hot stations/programmes first, in order of contribution margin per gadget sold Aggregation: by style and color for buyers, by category for merchandise managers Aggregation: By stations/ programmes for media buyers, by gadget for product managers Marketing Databases Online marketers like Amazon.com use “cookies”, electronic signatures place at a customer’s personal computer, so they not only keep track of what each customer has bought, but also recognize the customer when he or she logs on to their site. Airlines track members of their frequent flyer programmes and target some with special promotions. Supermarket chain Tesco in the UK uses its loyalty cards to track and analyze customer buying patterns, and to offer customers coupons and incentives tailored to their buying behavior. Tesco uses their analysis in deciding product placement on shelves, managing coupon campaigns, and to tailor product portfolios to individual stores. Designing marketing databases that take effective advantage of customer data that companies are in a position to collect requires that several major issues be considered: the cost of collecting the data, the economic benefits of using the data, the ability of the company to keep the data current in today’s mobile society, and the rapid advances in technology that permit the data to be used to maximum advantage. Collecting information, then storing and maintaining it, always costs money. If a company wants to know more about the demographics and lifestyles of its best customers, in addition to their purchasing histories, it must obtain demographic and lifestyle data about them. Doing so is more difficult than it sounds; most people are unwilling to spend much time filling out forms that ask nosy questions about education, income, whether they play tennis, and what kind of car they drive. The cost of collecting such information must be weighed against its value. What will be done with the information once it is in hand? Various commercial marketing databases are available, with varying depth and quality of information. Virtually every credit card issuer, magazine publisher, affinity group and others who sell to or deal directly with consumers sell their customer databases. Marketers who consider buying lists or other services from any of these commercial database providers need to inquire exactly how and where the data are collected. They should also compare the costs of databases containing names about which is more known to the extra value, compare to simpler compiled databases, such as those taken from telephone directories or automobile registrations. Marketers planning to build their own databases need also to consider several increasingly important ethical issues. For firms with deep pockets, advances in computing power and database technology including data-mining technology, are permitting firms to combine databases from different sources to permit a more complete understanding of any member of the database. Keeping current with what is possible in database technology is important, as technological advances often make possible that which was only a dream a short time ago. Competitive intelligence Systems Competitive intelligence (CI) is a systematic and ethical approach for gathering and analyzing information about competitors’ activities and related business trends. It is based on the idea that more than 80 per cent of all information is public knowledge. The most important sources of CI information include companies’ annual and other financial reports, speeches by company executives, government documents, online databases and trade organizations, as well as the popular and business press. he challenge is to find the relevant knowledge, analyze it, and share it with the decision makers in the organization, so they can use it. The critical questions that managers setting up a CI system should ask are: How rapidly does the competitive climate in our industry change? How important is it that we keep abreast of such changes? What are the objectives for CI in our company? Who are the best internal clients for CI? To whom should the CI effort report? What budget should be allocated to CI? Will it be staffed full- or part-time? In companies that operate in industries with dynamic competitive contexts, the use of full-time CI staff is growing. Client Contact Management Systems Salesforce automation software: helps companies disseminate real-time product information to salespeople to enable them to be more productive and more able to satisfy customer needs. Such software also allows companies to effectively capture customer intelligence from salespeople, keep track of it for use on later sales calls, and even transfer it to other salespeople in the event of a salesperson leaving the company. These programmes keep track of client’s’ names, addresses, phone and fax numbers, and they also provide an organized way to make notes about each contact with the customer. They also can remind the user when it is time to follow up with the customer on topic left pending. Most whose livelihood depends on face-to-face selling now use such systems to keep themselves organized. Other kinds of Market knowledge Systems New applications are being developed every day. Ultimately, the potential that many of these systems share is to enable marketers to serve target markets of one; that is, to know enough about any given customer and the competitive context that an offering can be tailored to fit each customer so well that the customer’s needs are met perfectly. Doing so is many a marketer’s dream! 8.6: Marketing Research Resolves Specific Marketing Challenges Marketing research task: the design, collection, analysis, and reporting of research intended to gather data pertinent to a particular marketing challenge or situation. Marketing research is intended to address carefully defined marketing problems or opportunities. Research carried out without carefully thought-out objectives usually mean time and money down the tubes! Some marketing problems commonly addressed through marketing research include tracking customer satisfaction from unit to unit or year to year (tracking studies); testing consumer responses to elements of marketing programmes, such as prices or proposed advertising campaigns; and assessing the likelihood that consumer will buy proposed new products. The steps in the marketing research process are shown in exhibit below. As this exhibit shows, the marketing research process is fraught with numerous opportunities for error. That’s why it’s so important that all who play influential roles in setting strategy for their firms or who use marketing research results for decision making be well-informed and critical users of the information that results from market research studies. To this end, we now address each of the steps in the marketing research process, from a decision-making point of view. Exhibit 8.7 Steps in the marketing research process: what can go wrong? Steps What frequently goes wrong? 1. Identify managerial problem and establish research objectives Management identifies no clear objective, no decision to be made based on the proposed research. 2. Determine data sources (primary or secondary) and types of data and research approaches (qualitative or quantitative) required Primary data are collected when cheaper and faster secondary data will do. Quantitative data are collected without first collecting qualitative data. 3. Design research: type of study, data collection approach, sample, etc. These are technical issues best managed by skilled practitioners. Doing these steps poorly can generate misleading or incorrect results. 4. Collect data Collector bias: hearing what you want to hear. 5. Analyze data Tabulation errors or incorrect use or interpretation of statistical procedures may mislead the user. 6. Report results to the decision maker Some users do not really want objective information – they want to prove what they already believe to be true. Step 1: Identify the Managerial Problem and Establish Research Objectives A good place to start it start is to ask what managerial problem or question is that a proposed programme of research might address The result is a set of research objectives (e.g., determine market size and growth rate; assess supplier power in this industry, and so on) that will drive the research. Step 2: Determine the Data Sources and Types of Data Required This step is critical in determining the cost-effectiveness and timeliness of the research effort. The researcher must answer two key questions at this stage: Should I gather data from primary or secondary sources? Whichever type of data sources are called for, do I need qualitative or quantitative research to satisfy my research objectives, or both? Primary or Secondary Sources? Primary data are data collected from individual research subjects using observation, a survey, interviews, or whatever. The data are then gathered and interpreted for the particular research objective at hand. Secondary data already exist – on the Internet, in government documents, in the business press, in company files, or wherever. Someone has already done the primary data collection and placed the data where others can access it, whether easily or with difficulty, whether free or at some cost. Which is better – primary or secondary data? If (and it’s an important if) a research objective can be met using secondary data, that’s usually the best course to follow. Why? First, it’s usually quicker to find the data somewhere than to collect information from scratch. Second, it’s usually less costly to simply find existing secondary data than to collect the information as primary data all over again. Third, secondary data are typically based on what people actually do, or how they actually behave. Surveys, a common form of primary data, are based on what people say. The two are not the same. Qualitative or Quantitative Data and Research Approaches? Most secondary research studies require both qualitative and quantitative data. If primary data are necessary, a decision must be made about whether to collect that data using qualitative or quantitative research approaches. Qualitative research usually involves small samples of subjects and produces information that is not easily quantifiable. Qualitative data may yield deeper insights into consumer behavior than are available from quantitative research. For this reason, qualitative research is often conducted first and used to guide subsequent quantitative research. An important drawback of qualitative research, is that it’s generally small samples may not fairly represent the larger population. Most experienced marketing researchers would say, ‘Never generalize from qualitative research. Always follow up with a quantitative study to test the hunches developed in the qualitative study.’ Such statements presume, however, that adequate research resources are available to conduct additional studies. Often, and particularly in entrepreneurial settings, such is not the case, and decision makers are forced to rely, albeit tenuously, on small-scale qualitative studies. Quantitative research collects data that are amenable to statistical analysis, usually from large enough samples so that inferences may be drawn with some confidence to the population from which the subjects in the sample are drawn. The principal benefit of quantitative research lies in its measurement of a population’s attitudes toward or likely response to products or marketing programmes. Because of their larger sample sizes and quantitative metrics, greater confidence can be placed in quantitative studies, when conducted properly, using appropriate sampling procedures and statistical techniques. Qualitative Research Techniques There are seemingly as many qualitative research techniques as there are stars in the sky. The most common ones, however, are focus groups and interviews of various kinds. A focus group typically consists of 8 to 12 consumers from the marketer’s target market brought together at a research facility to discuss a particular marketing problem, such as attitudes toward a proposed new product and various possible features. A skilled moderator conducts the focus group, records the conversation on audio and/or videotape, and writes a report of the findings. Typically two or more groups are conducted for a single research project. Focus groups have significant limitations: They are subject to data distortion caused by a dominant person in the group, their results are difficult to interpret, and they are neither representative of nor generalizable to a larger population, due to their small sample size and convenience samples. They are a good way, however, to begin a research inquiry or to gather at least some information when research budgets are tight. Quantitative Research Techniques In most quantitative research, questionnaires are used that enable the researcher to measure the subjects’ responses on quantitative scales. These scales enable the researcher to compare product attributes, the responses of demographically different consumers, and other differences in order to better understand what consumers prefer, how satisfied they are with one product compared to others, and so on. Novice researchers, or those whose budgets are limited, can sometimes obtain useful market knowledge from small-scale research that begins with some qualitative research, perhaps several interviews, and concludes with a quantitative study using measures such as those shown in exhibit below. Gaining experience with such research, even in a class project setting, provides future managers with some appreciation for the conduct of marketing research and the limitations to its interpretation. Step 3: Design the Research Designing secondary research is a simple matter of finding sources of information sufficient to satisfy the research objectives and ensuring that the sources are credible. For primary quantitative research, research design is the most technical and most difficult step in conducting the research. The key decisions to be made in primary research design are to determine the data collection method and prepare the research instrument, determine how to conduct the participants in the research, and design the sampling plan. Determine the Data Collection Method and Prepare the Research Instrument The most common methods of collecting primary data are observation, survey, and experiment. Observation is just that: observing subjects using pay phones. Typically, a form is prepared on which the observer records what is being observed, perhaps minutes of use and gender of the user, among other things. Surveys involve writing a questionnaire, which will include questions and either scaled answers or spaces for open-ended answers. Demographic information about the respondent is also usually requested to aid in market segmentation and market targeting decisions. Constructing survey questions and formats for the answers is more difficult than one might expect and is beyond the scope of this book, but several sources cited in this module can help bring the reader up to speed on these tasks. Experiments are studies in which the researcher manipulates one or more variables, such as price or product features, either within the context of a survey or in a laboratory or field setting, in order to measure the effect of the manipulated variable on the consumer’s response. One common use of experiments is to examine the consumer’s likelihood to buy a new product at different price points. Different respondents are given different prices for the product, and the researcher tests differences in consumers’ likelihood to buy as the price changes. This procedure entails less bias than asking consumers what they would be willing to pay for a product, the typical answer to which is ‘as little as possible!’ Exhibit 8.9 Some commonly used types of scales for quantitative market research Type of scale Description Example Semantic Differential Scale A scale connecting two bipolar words or phrases How satisfied are you with your provider of cable TV? Not at all satisfied? 1?2?3?4?5?6?7? Extremely satisfied Likert Scale A statement with which the respondent shows the amount of agreement/ disagreement I am extremely satisfied with my provider of cable TV. Strongly agree? 1?2?3?4?5?6?7?Strongly disagree Quality Rating Scale Rates some attribute on a scale from ‘excellent’ to ‘poor’ My cable TV service, overall, is: Poor ? Fair? Good? Very Good ?Excellent Importance Scale, using semantic Rates the importance of some attribute How important are the following criteria to your satisfaction with your cable TV provider? differential Not at all Extremely format important important Answers the phone quickly 1?2?3?4?5?6?7 Prompt repair service 1?2?3?4?5?6?7 Cleans up after installation 1?2?3?4?5?6?7 Service never goes dark 1?2?3?4?5?6?7 Intention-to-Buy Scale Measures how likely the respondent is to buy at How likely are you to sign up for the new InterGalactic Channel for an extra $4.95 per month? some price Definitely ____ Probably ____ Might or might not ____ Probably not ____ Definitely not ____ Determine the Contact Method Once a data collection method is chosen, the researcher must decide how to contact those who will participate in the research. Common choices include face-to-face (perhaps in a shopping mall or a public place), mail, telephone, fax, email, and the Internet. A significant problem with survey research is that those who choose not to participate when asked (‘we’re eating dinner now, and please don’t call back!’) may differ from those who do participate. This nonresponse bias may distort the results of the research. Response rate can also be a problem, since many who are asked to participate will not do so. Response rates for mail surveys generally run about 15 to 20 per cent. The other types are better or worse. Thus, for a mail survey, five to six times the number of surveys the researcher hopes to receive must be mailed. Highly representative group, not simply a bunch of survey junkies. Exhibit 8.10 Pros and cons of different contact methods for survey research Method Response rate Cost Timeliness Nonresponse bias Face-to-face High High Slow Low Mail Low Low Slow High Telephone Moderate Moderate Fast Moderate Fax Moderate Low Fast High Email Low Low Fast High Internet Low Low Fast High Design the Sampling Plan Selecting a sample of participants for observational, survey, or experimental research requires that three questions be answered: Who is the population (or universe) from which the sample of respondents will be drawn? What sample size is required to provide an acceptable level of confidence? By what method, probability sampling (also called random sampling) or nonprobability sampling (such as convenience sampling), will the sample be selected? First, the population from which the sample is to be drawn must be clearly specified. Typically, it consists of the target market, defined in demographic or behavioral terms. Second, the sample must be large enough to provide confidence that statistical data, such as mean responses to survey questions, are truly within some narrow-enough range, sometimes called the margin of error. In general, the larger the sample size, the smaller the margin of error. Third, the idea behind probability or random sampling is that every person in the population has an equal chance of being selected. If nonprobability samples, such as convenience samples, are used, the sample may be biased. Convenience samples are used quite often for marketing research because true random samples are more difficult and costly to reach. The nonresponse problem makes almost all samples potentially biased in the same way. If the method is not random, the user should inquire about how the sample was selected to look for any obvious source of bias that might distort the research results. Step 4: Collect the Data The data collection contributes more to overall error than any other step in the process. In some cases, especially where entrepreneurs or marketers conduct marketing research themselves instead of contracting with a third party for data collection, collector bias can be a problem. The person collecting the data might, in his or her enthusiasm for the product, bias the respondents so they tell the researcher what they think he or she wants to hear. Errors in face-to-face or telephone surveys include those that derive from nonresponse by some respondents; selection errors by the interviewer; the way the interviewer asks the questions; the interviewer’s interpretation and recording of answers; and even interviewer cheating. In surveys conducted by fax, email, or over the Internet, an additional problem is that the researcher does not know who actually replied to the survey. The data collection effort can be substantial. To complete 100 surveys in the United Kingdom with randomly selected homes using random digit dialing, several hundred phone numbers will likely be required and 1000 dialing’s! Step 5: Analyze the Data When the data have been collected, the completed data forms must be processed to yield the information the project was designed to collect. Typically, the data are then entered into computer files, percentages and averages are computed, and comparisons are made between different classes, categories, and groups of respondents. Often, sophisticated statistical analyses are required. Step 6: Report the Results to the Decision Maker If the research study began with clearly defined objectives, reporting the results simply returns to those objectives and reports what was found. Where research is carried out without clear objectives, reporting can be difficult, as no clear conclusions may be available. Lots of marketing research money is wasted in some companies because of poorly specified research objectives. 8.7: What Users of Marketing Research Should Ask The research process described in the preceding section makes clear where many of the potential stumbling blocks are in designing and conducting marketing research. The informed and critical user of marketing research should ask the following questions, ideally before implementing the research or if necessary subsequent to its completion, to ensure that the research is unbiased and the results are reliable. What are the objectives of the research? Will the data to be collected meet those objectives? Are the data sources appropriate? Are cheaper, faster secondary data used where possible? Is qualitative research planned to ensure that quantitative research, if any, is on target? Are the planned qualitative and/or quantitative research approaches suited to the objectives of the research? Qualitative research is better for deep insights into consumer behavior, while quantitative research is better for measurement of a population’s attitudes and likely responses to products or marketing programmes. Is the research designed well? Will questionnaire scales permit the measurement necessary to meet the research objectives? Are the questions on a survey or in an interview or focus group unbiased? (‘Isn’t this a great new product? Do you like it?’) Do the contact method and sampling plan entail any known bias? Is the sample size large enough to meet the research objectives? Are the planned analyses appropriate? They should be specified before the research is conducted. 8.8: Rudimentary Competence: Are We There Yet? Given the importance of marketing research in strategic decision making today, we encourage every business student from every business discipline to try his or her hand at it. A wide variety of software applications have been developed to aid marketers in conducting marketing research and applying it and other data to specific marketing problems. In subsequent modules, we’ll point out specific applications for which such systems are commonly used. Various trade magazines publish annual directories that list providers of these tools and other services that facilitate marketing research.