Spread Knowledge

MGT301 - Principles of Marketing - Lecture Handout 21

User Rating:  / 0

Lesson overview and learning objectives:

In last Lesson we discussed the concept regarding some individual decisions about the product like product attributes, labeling and packaging. Today we will continue the same topic and will discuss the process of new product development again as well.


A. Product Line Strategies

Product Line Strategies

We have looked at product strategy decisions such as branding, packaging, labeling, and support services for individual products and services. But product strategy also calls for building a product line. A product line is a group of products that are closely related because they function in a similar manner, are sold to the same customer groups, are marketed through the same types of outlets, or fall within given price ranges. For example, Nike produces several lines of athletic shoes and Motorola produces several lines of telecommunications products. In developing product line
strategies, marketers face a number of tough decisions. The major product line decision involves product line length—the number of items in the product line. The line is too short if the manager can increase profits by adding items; the line is too long if the manager can increase profits by dropping items. Company objectives and resources influence product line length. Product lines tend to lengthen over time. The sales force and distributors may pressure the product manager for a more complete line to satisfy their customers. Or, the manager may want to add items to the product line to create growth in sales and profits. However, as the manager adds items, several costs rise: design and engineering costs, inventory costs, manufacturing changeover costs, transportation costs, and promotional costs to introduce new items. Eventually top management calls a halt to the mushrooming product line. Unnecessary or unprofitable items will be pruned from the line in a major effort to increase overall profitability. This pattern of uncontrolled product line growth followed by heavy pruning is typical and may repeat itself many times.

The company must manage its product lines carefully. It can systematically increase the length of its product line in two ways: by stretching its line and by filling its line. Product line stretching stretches its line downward, upward, or both ways.

Many companies initially locate at the upper end of the market and later stretch their lines downward. A company may stretch downward to plug a market hole that otherwise would attract a new competitor or to respond to a competitor's attack on the upper end. Or it may add low-end products because it finds faster growth taking place in the low-end segments.

B. New-product development

New-product development

Given the rapid changes in consumer tastes, technology, and competition, companies must develop a steady stream of new products and services. A firm can obtain new products in two ways. One is through acquisition—by buying a whole company, a patent, or a license to produce someone else's product. The other is through new-product development in the company's own research and development department. By new products we mean original products, product improvements, product modifications, and new brands that the firm develops through its own
research and development efforts. In this chapter, we concentrate on new-product development. New products continue to fail at a disturbing rate. One source estimates that new consumer packaged goods (consisting mostly of line extensions) fail at a rate of 80 percent. Moreover, failure rates for new industrial products may be as high as 30 percent.3Why do so many new products fail? There are several reasons. Although an idea may be good, the market size may have been overestimated. Perhaps the actual product was not designed as well as it should have been. Or maybe it was incorrectly positioned in the market, priced too high, or advertised poorly. A high-level executive might push a favorite idea despite poor marketing research findings. Sometimes the costs of product development are higher than expected, and sometimes competitors fight back harder than expected.

Because so many new products fail, companies are anxious to learn how to improve their odds of new-product success. One way is to identify successful new products and find out what they have in common. Another is to study new-product failures to see what lessons can be learned. Various studies suggest that new-product success depends on developing a unique superior product, one with higher quality, new features, and higher value in use. Another key success factor is a welldefined product concept prior to development, in which the company carefully defines and
assesses the target market, the product requirements, and the benefits before proceeding. Other success factors have also been suggested—senior management commitment, relentless innovation, and a smoothly functioning new-product development process. In all, to create successful new products, a company must understand its consumers, markets, and competitors and develop products that deliver superior value to customers.
So companies face a problem—they must develop new products, but the odds weigh heavily against success. The solution lies in strong new-product planning and in setting up a systematic new-product development process for finding and growing new products. Figure shows the eight major steps in this process.

a) Idea generation

New-product development starts with idea generation—the systematic search for new-product ideas. A company typically has to generate many ideas in order to find a few good ones. Major sources of new-product ideas include internal sources, customers, competitors, distributors and suppliers, and others. Using internal sources, the company can find new ideas through formal research and development. It can pick the brains of its executives, scientists, engineers, manufacturing, and salespeople. Some companies have developed successful "entrepreneurial" programs that encourage employees to think up and develop new-product ideas. Good newproduct ideas also come from watching and listening to customers. The company can analyze customer questions and complaints to find new products that better solve consumer problems. The company can conduct surveys or focus groups to learn about consumer needs and wants. Or company engineers or salespeople can meet with and work alongside customers to get suggestions and ideas. Finally, consumers often create new products and uses on their own, and companies can benefit by finding these products and putting them on the market. Customers can also be a good source of ideas for new product uses that can expand the market for and extend the life of current products. Competitors are another good source of new-product ideas. Companies watch competitors' ads and other communications to get clues about their new products. They buy competing new products, take them apart to see how they work, analyze their sales, and decide whether they should bring out a new product of their own. Finally, distributors and suppliers contribute many good new-product ideas. Resellers are close to the market and can pass along information about consumer problems and new-product possibilities. Suppliers can tell the company about new concepts, techniques, and materials that can be used to develop new products. Other idea sources include trade magazines, shows, and seminars; government agencies; newproduct consultants; advertising agencies; marketing research firms; university and commercial
laboratories; and inventors.
The search for new-product ideas should be systematic rather than haphazard. Otherwise, few new ideas will surface and many good ideas will sputter in and die. Top management can avoid these problems by installing an idea management system that directs the flow of new ideas to a central point where they can be collected, reviewed, and evaluated. In setting up such a system, the company can do any or all of the following:

  • Appoint a respected senior person to be the company's idea manager.
  • Create a multidisciplinary idea management committee consisting of people from R&D, engineering, purchasing, operations, finance, and sales and marketing to meet regularly and evaluate proposed new-product and service ideas.
  • Set up a toll-free number for anyone who wants to send a new idea to the idea manager.
  • Encourage all company stakeholders—employees, suppliers, distributors, dealers— to send their ideas to the idea manager.
  • Set up formal recognition programs to reward those who contribute the best new ideas.

The idea manager approach yields two favorable outcomes. First, it helps create an innovationoriented company culture. It shows that top management supports, encourages, and rewards innovation. Second, it will yield a larger number of ideas among which will be found some especially good ones. As the system matures, ideas will flow more freely. No longer will good ideas wither for the lack of a sounding board or a senior product advocate

b) Idea Screening

The purpose of idea generation is to create a large number of ideas. The purpose of the succeeding stages is to reduce that number. The first idea-reducing stage is idea screening, which helps spot good ideas and drop poor ones as soon as possible. Product development costs rise greatly in later stages, so the company wants to go ahead only with the product ideas that will turn into profitable products. As one marketing executive suggests, "Three executives sitting in a room can get 40 good ideas ricocheting off the wall in minutes. The challenge is getting a steady stream of good ideas out of the labs and creativity campfires, through marketing and manufacturing, and all the way to consumers."
Many companies require their executives to write up new-product ideas on a standard form that can be reviewed by a new-product committee. The write-up describes the product, the target market, and the competition. It makes some rough estimates of market size, product price, development time and costs, manufacturing costs, and rate of return. The committee then evaluates the idea against a set of general criteria such as these: Is the product truly useful to
consumers and society? Is it good for our particular company? Does it mesh well with the company's objectives and strategies? Do we have the people, skills, and resources to make it succeed? Does it deliver more value to customers than do competing products? Is it easy to advertise and distribute? Many companies have well-designed systems for rating and screening new-product ideas.

c) Concept Development and Testing

An attractive idea must be developed into a product concept. It is important to distinguish between a product idea, a product concept, and a product image. A product idea is an idea for a possible product that the company can see itself offering to the market. A product concept is a detailed version of the idea stated in meaningful consumer terms. A product image is the way consumers perceive an actual or potential product.
Concept testing calls for testing new-product concepts with groups of target consumers. The concepts may be presented to consumers symbolically or physically For some concept tests, a word or picture description might be sufficient. However, a more concrete and physical presentation of the concept will increase the reliability of the concept test. Today, some marketers are finding innovative ways to make product concepts more real to consumer subjects. For example, some are using virtual reality to test product concepts. Virtual reality programs use computers and sensory devices (such as gloves or goggles) to simulate reality.

d) Marketing strategy Development

The next step is marketing strategy development, designing an initial marketing strategy for introducing this car to the market.
The marketing strategy statement consists of three parts. The first part describes the target market; the planned product positioning; and the sales, market share, and profit goals for the first few years. The second part of the marketing strategy statement outlines the product's planned price, distribution, and marketing budget for the first year. The third part of the marketing strategy statement describes the planned long-run sales, profit goals, and marketing mix strategy:

e) Business Analysis

Once management has decided on its product concept and marketing strategy, it can evaluate the business attractiveness of the proposal. Business analysis involves a review of the sales, costs, and profit projections for a new product to find out whether they satisfy the company's objectives. If they do, the product can move to the product development stage.
To estimate sales, the company might look at the sales history of similar products and conduct surveys of market opinion. It can then estimate minimum and maximum sales to assess the range of risk. After preparing the sales forecast, management can estimate the expected costs and profits for the product, including marketing, R&D, operations, accounting, and finance costs. The company then uses the sales and costs figures to analyze the new product's financial attractiveness

f) Product Development

So far, for many new-product concepts, the product may have existed only as a word description, a drawing, or perhaps a crude mock-up. If the product concept passes the business test, it moves into product development. Here, R&D or engineering develops the product concept into a physical product. The product development step, however, now calls for a large jump in investment. It will show whether the product idea can be turned into a workable product.
The R&D department will develop and test one or more physical versions of the product concept. R&D hopes to design a prototype that will satisfy and excite consumers and that can be produced quickly and at budgeted costs. Developing a successful prototype can take days, weeks, months, or even years. Often, products undergo rigorous functional tests to make sure that they perform safely and effectively. The prototype must have the required functional features and also convey the intended psychological characteristics.

g) Test Marketing

If the product passes functional and consumer tests, the next step is test marketing, the stages at which the product and marketing program are introduced into more realistic market settings. Test marketing gives the marketer experience with marketing the product before going to the great expense of full introduction. It lets the company test the product and its entire marketing program—positioning strategy, advertising, distribution, pricing, branding and packaging, and budget levels.
The amount of test marketing needed varies with each new product. Test marketing costs can be enormous, and it takes time that may allow competitors to gain advantages. When the costs of developing and introducing the product are low, or when management is already confident about the new product, the company may do little or no test marketing. Companies often do not testmarket simple line extensions or copies of successful competitor products.

h) Commercialization

Test marketing gives management the information needed to make a final decision about whether to launch the new product. If the company goes ahead with commercialization—introducing the new product into the market—it will face high costs. The company will have to build or rent a manufacturing facility. The company launching a new product must first decide on introduction timing Next, the company must decide where to launch the new product—in a single location, a region, the national market, or the international market. Few companies have the confidence, capital, and capacity to launch new products into full national or international distribution. They will develop a planned market rollout over time. In particular, small companies may enter attractive cities or regions one at a time. Larger companies, however, may quickly introduce new models into several regions or into the full national market.

Speeding Up New-Product Development

Many companies organize their new-product development process into the orderly sequence of steps starting with idea generation and ending with commercialization. Under this sequential product development approach, one company department works individually to complete its stage of the process before passing the new product along to the next department and stage. This orderly, step-by-step process can help bring control to complex and risky projects. But it also can be dangerously slow. In fast-changing, highly competitive markets, such slow-but-sure product development can result in product failures, lost sales and profits, and crumbling market positions.
"Speed to market" and reducing new-product development cycle time have become pressing concerns to companies in all industries.
In order to get their new products to market more quickly, many companies are adopting a faster, team-oriented approach called simultaneous (or team-based) product development. Under this approach, company departments work closely together, overlapping the steps in the product development process to save time and increase effectiveness. Instead of passing the new product from department to department, the company assembles a team of people from various departments that stay with the new product from start to finish. Such teams usually include people
from the marketing, finance, design, manufacturing, and legal departments, and even supplier and customer companies.
Top management gives the product development team general strategic direction but no clear-cut product idea or work plan. It challenges the team with stiff and seemingly contradictory goals— "turn out carefully planned and superior new products, but do it quickly"—and then gives the team whatever freedom and resources it needs to meet the challenge. In the sequential process, a bottleneck at one phase can seriously slow the entire project. In the simultaneous approach, if one functional area hits snags, it works to resolve them while the team moves on.


New-product development: The development of original products, product improvements, product modifications, and new brands through the firm's own R&D efforts.

Idea generation: The systematic search for new-product ideas.

Idea screening: screening new-product ideas in order to spot good ideas and drop poor ones as soon as possible.

Product concept: A detailed version of the new-product idea stated in meaningful consumer terms.

Concept testing: Testing new-product concepts with a group of target consumers to find out if the concepts have strong consumer appeal

Business analysis: A review of the sales, costs, and profit projections for a new product to find out whether these factors satisfy the company's objectives.

Product development: A strategy for company growth by offering modified or new products to current market segments. Developing the product concept into a physical product in order to ensure that the product idea can be turned into a workable product.

Commercialization: Introducing a new product into the market.

Test marketing: The stage of new-product development in which the product and marketing program are tested in more realistic market settings.

Sequential product development A new-product development approach in which one company department works
to complete its stage of the process before passing the new product along to the next department and stage.


Related Content: MGT301 - VU Lectures, Handouts, PPT Slides, Assignments, Quizzes, Papers & Books of Principles of Marketing