Developing a Franchise System

by Scott Shane

last updated 23/07/2009

Scott Shane is the Leghorn Career Development Assistant Professor of Entrepreneurship at MIT School of Management. Chester Spell is now Senior Lecturer at the University of Waikato. This article was first published in the Sloan Management Review, Spring

What Successful Franchisors Have In Common

by Scott Shane

last updated 23/07/2009

Scott Shane is the Leghorn Career Development Assistant Professor of Entrepreneurship at MIT School of Management. Chester Spell is now Senior Lecturer at the University of Waikato. This article was first published in the Sloan Management Review, Spring
A new study by Scott Shane and Chester Spell reveals the factors which help new franchise systems succeed

(Editor's note: It is important to emphasise that the following article is based upon the authors' research into franchising in the US alone. Such figures as are available suggest that the failure rate of new franchise systems in New Zealand is considerably lower than those quoted below: however, I believe that many of the conclusions drawn are equally relevant to franchising in this country.)

Entrepreneurs typically establish a new franchise system because they have created a formula which is superior to that of their competitors. If an entrepreneur has built better outlet operations, he or she can earn high profits by establishing a chain of outlets and reproducing the superior profit-generating operations in many locations. However, efforts to establish chains are often fraught with failure.

In our research, we studied 157 companies in 27 industries that first began to franchise in the United States between 1981 and 1983, and examined their performance from 1984 to 1995. We obtained information on 53 different dimensions of new franchise systems, both survivors and non-survivors, for each of the twelve years.

We looked at the causal path between different characteristics of the new franchisors, and measured the effect of these relationships on the probability that the new franchisor would cease to franchise in each year. As a result, we were able to identify a model of what makes new franchise systems succeed.

Our research shows that the average new franchisor in the US fails. In fact, during the twelve year period after beginning to franchise, typically three-quarters of the new franchisors have ceased to exist (see Figure 1).

The attrition rate is quite severe in the early years of franchising; one-third of the new systems stop franchising in their first four years.

The survival patterns differ very little across industries (See Figure 2).

Although the small sample size necessitated categorising the new franchisors into three broad groups – food, other retail and services – similar patterns emerged across the three. New food franchises have a slightly more severe failure rate in the early years than do retail or service franchisors, but the failure rate slows more quickly than for the other two groups.

Twelve-year survival rates are also slightly higher for food franchisors than for service or retail franchisors. Service franchisors have nearly as slow failure rates in their early years as food franchisors, but continue to fail at a faster rate than food franchisors after their seventh year. Retail franchisors appear to have a lower failure rate than service or food franchisors early on. However, they display little decline in the pace of that failure rate as they age.

The high death rate of new systems suggests that franchising is not an easy business in which to succeed and demonstrates the importance of creating a model to explain the survival of new franchise systems.

Are Successful New Franchisors Just Lucky?

One-quarter of new franchise systems survive, and many grow to establish important, thriving chains like McDonald's and Mail Boxes Etc. What do the successful franchise systems do differently? One hypothesis is that they are simply lucky, and that there is little entrepreneurs can do to improve their chances of establishing successful chains.

However, our initial examination of the characteristics of new franchise systems quickly ruled out this hypothesis. The successful new systems were different from the unsuccessful ones in too many ways for their success to be just a matter of luck. In comparison to unsuccessful new franchisors, the more successful ones were larger, offered fewer supporting services, had more recognisable brand names, offered longer-term contracts with franchisees, had fewer headquarters staff people per outlet, and were more likely to be members of the IFA (the US equivalent of the Franchise Association of New Zealand) and to register their franchise systems with state authorities.

Many of these characteristics are related. For example, it is logical to think that the larger a firm, the more recognisable is its brand name. We mapped the relationship between different characteristics of the new franchise systems (see Figure 3). The characteristics ultimately generate a path to system survival.


A fictitious new franchisor, Newfran, is a composite of the successful new franchisors in our study. It illustrates not only the average values for the key characteristics of success, but also shows how the characteristics relate in contributing to success.

Joe Franchisor established Newfran in 1972 and began to franchise ten years later. Newfran was a retail business that sold food, clothing and other retail products in stand-alone outlets that employed seven full-time people. When Newfran began to franchise, it had seven company-owned outlets, Newfran charged franchisees a royalty rate of 5 percent of sales and a 2 percent of sales fee for advertising. The system did not give either direct or indirect financing to franchisees. In the eighteen-year initial contract, Newfran required its franchisees to pay a per-outlet fee of $19,500 and make other franchise investments of $145,000. In return, Newfran supported the franchisee with a headquarters staff of thirteen, and provided 130 hours of training.

What made Newfran more successful than the typical franchise system started at the same time? The model provides some answers. During the course of managing his company, Joe noticed that his stores operated much more productively that those of his competitor, Bill Retail. Joe realised that he had developed better ways to manage retail outlets and make better or cheaper products. While Joe knew his ideas for running a retail business were valuable, he still faced a fundamental problem. Better ideas alone do not provide a competitive advantage. Since his ideas were neither trade secrets nor patentable technologies, they could be copied once they became public. As soon as Joe disclosed his operations to a franchisee, competitors could copy them.

To create a competitive advantage for the new franchise chain, Joe needed to build a valuable brand name to protect the chain from imitation. While competitors can copy outlet operations, they cannot copy a brand name. So outlets bearing the name 'Newfran' could not be imitated completely by other firms.

At the time it began to franchise, Newfran was less than 30 percent of the size of the average franchise system in its industry. Economies of scale in advertising allow companies to lower the per-unit cost of advertising to promote their brand names, so when Newfran began to franchise, its per-outlet cost of brand-name promotion was higher than that of its established competitors. Far too small to promote the new franchise system's brand name and attract consumers to the retail outlets as efficiently, Newfran needed to grow.

Rapid Growth

The development of a new system's operating routines may be costly for the entrepreneur, but once developed they can be replicated in multiple locations at low additional cost. By the time he considered franchising, Joe had already developed his superior outlet-operation routines (or systems) while building his seven-store chain. He realised that relocating these routines in additional outlets would incur little additional cost, which gave Joe an incentive to establish outlets as quickly as possible. The faster he could create retail outlets, the greater the chance that his chain would reach a scale to promote its brand name competitively before competitors could copy its outlet operations.

Each of Joe's retail outlets drew customers from a confined geographic area. After all, how far do people want to travel for lunch or to buy clothing? This meant that once Joe had saturated his local market, he needed to expand the system geographically in order to create a large number of Newfran outlets. In the beginning, Joe sought franchisees in forty-one states but did not seek franchisees outside the US.

Newfran's broad geographic expansion imposed great time pressures on Joe. As an entrepreneur, he had two jobs: he needed to sell franchises, and also needed to manage and innovate in his existing outlets. Joe wanted to delegate these activities to others, but he realised that as a sole shareholder in Newfran he had a greater incentive than his employees to explore new products and markets. Joe also realised that he had unique incentives to manage existing operations efficiently because he received all the profits.

Joe did not believe that spending 100-hour weeks on his business was worth the personal sacrifice, so he needed to find a way to build the company without burning himself out. As Newfran spread into different geographical markets, Joe found that gathering information on local real estate, labour markets and customer tastes directly was time-consuming and expensive, since it required his investment in and supervision of employees.

By talking to other franchisors, Joe learned that he could reduce his time constraints and grow faster by delegating many local entrepreneurial tasks to franchisees. Franchisors suggested that Joe could reduce the cost of such activities as selecting sites, negotiating leases on property, opening outlets, and adapting operations to local labour conditions, by delegating them to franchisees who already had this knowledge. In following this advice, Joe provided franchisees with fewer local market services than did the typical franchisor who started at the same time. Instead, Joe looked for franchisees who could operate successfully without field operations assistance.

This choice enhanced Newfran's human resource efficiency. Since Joe devoted fewer people to gathering information about local markets, he could reallocate them to selling outlets and selecting and training franchisees. Joe's strategy made Newfran much more efficient than other new franchisors. Unlike the typical failed new franchisor that could manage only one outlet for each member of its headquarters staff, Newfran achieved a seven-to-one ratio. Consequently, Newfran's per-outlet system cost was significantly lower than that of the typical new system.

Newfran's human resource efficiency accelerated the system's growth rate and enabled it to grow rapidly within a short time, much more so than other new systems. By 1985, three years after it had begun franchising, Newfran had 155 outlets. By 1994, it had doubled its size. Newfran's failed competitors were unable to achieve this rate of growth.

As Newfran grew, its cost of competing with established franchisors to promote its system brand name decreased. Between 1982 (when Joe began to franchise) and 1985, the cost per outlet of promoting Newfran's brand name dropped 93 percent, as the size of the chain increased from 7 outlets to 155 outlets and the amount of advertising Joe undertook grew more slowly. This rapid reduction in per-outlet advertising costs allowed Newfran to build a higher level of brand-name recognition than was the case for the typical unsuccessful franchisor. The increased value of Newfran's brand name also made it more likely to survive by attracting customers in ways that competitors could not easily copy.

Local Franchisees

Rapid growth of the system and Joe's reliance on local market entrepreneurship meant that Joe needed to attract franchisees, a difficult undertaking at first. Franchisees find it difficult to evaluate the quality of a franchise system before investing in it. Moreover, the quality of new systems is very uncertain because outlets may not generate the projected level of sales, and because US research shows many new systems and outlets fail.

To reduce their risk, potential franchisees need to gather as much information as they can about franchisor quality first. Successful franchise systems can reduce franchisee risk by indicating the quality of their franchise. For example, they can subject their systems to the scrutiny of the International Franchise Association (IFA). By joining the association, Joe demonstrated that his system adhered to the trade association's standards, which are higher than those in the general franchising market

(the same would apply to the Franchise Association of New Zealand – Ed.). Surviving new-franchise systems are 11 percent more likely to have joined the IFA than are unsuccessful new franchisors. And, in the critical early years, when the system does not have a reputation, successful new franchisors are as much as 26 percent more likely to join the IFA than are unsuccessful new franchisors.

Membership of the IFA requires franchisors to adhere to US state and federal regulations. By imposing a higher standard, registration enhances the new franchisor's chance of survival. Successful new franchisors are 22 percent more likely to register with state authorities than is the typical unsuccessful franchisor.

Its high initial fees and investment meant that Newfran needed IFA membership and state registration as a way of establishing its legitimacy. The greater the projected sales volume of the retail outlet that a franchisee is buying, the greater the size of his or her investment in the franchise system. The higher the outlet cost, the greater the franchisee's downside risk. The greater the franchisee's downside risk, the more information that franchisees want on the quality of the new franchise system. Therefore, franchisors are more likely to signal the quality of more costly systems through registration.

Franchisor Trustworthiness

Joe needed to convince potential franchisees that he would not take advantage of them. Because Newfran's contracts reserved the right of termination, many potential franchisees feared that Newfran would use termination to take back high-performing outlets before their investment had paid off.

Until Joe had established a reputation for ethical behaviour towards franchisees, he needed to find a way to reassure them. Joe reduced this problem by establishing longer-term contracts with franchisees, six years longer than the contract of the typical unsuccessful franchisor. With a longer-term contract, the franchisee's investment in the system was more likely to be amortised before Joe had a chance to terminate the franchisee and control the outlet.

Successful Systems

Successful franchise systems change over time, largely because of the size and distribution of the system rather than because of changing franchisor policies. From 1984 to 1995 Newfran expanded, registering and establishing outlets in new states and increasing the outlets in the states in which it already operated (see Figure 4).

Successful new franchisors also expand into other countries. The percentage of successful new franchise systems looking for franchisees overseas increased from 39 percent to 59 percent during the study. By 1995, Newfran had established a significant presence in Canada and other foreign countries.

Successful franchisors also change their system size and characteristics. Over time, Newfran established more franchised outlets. It had a relatively stable number of company-owned outlets. The system grew significantly from 1984 to 1995, primarily through the establishment of additional franchised outlets.

While Joe changed his operations and the terms of his contracts over time, his initial choices in pricing, services and operations had powerful consequences for Newfran's growth and overall success. His early decisions about contracts continue to influence his operations; his contracts with franchisees established in 1984 are still in force. His correct choices in 1984 are a big reason that Joe is still in business.

Stability Of The Successful System

Was the Newfran model consistently successful over time, or just in the initial year? To make sure that the model was consistent, we recalculated it for different years; our results showed the same model (see Figure 3). In addition, the model predicts new franchise system survival across both time and new franchise systems.

Why does the success model continue to predict franchise system survival even as the new franchisors mature? The answer appears to lie in the consistency of many attributes of franchise systems. Our examination of the successful new systems showed that they changed few of their basic policies during the twelve years of the study.

For example, the average royalty rates and advertising fees were 5% and 2% respectively of sales during the time of the study. The average franchise fees fluctuated between US$19,488 and US$28,385 but showed no apparent trend up or down. Similarly, the amount of franchisee training, the size of the initial investment, and the amount of the cash investment varied, but without a trend. The term of the franchise agreement and the length of the renewal period also remained relatively constant during the period. There was virtually no change in the support services that successful new franchisors offered to their franchisees.

This stability of policies meant that if the new franchisors initially established franchise systems in line with the success model, the franchisors were able to maintain the positive web of policies and increase their likelihood of survival over time. The contract between the franchisor and franchisee lasts a relatively long time, is subject to some regulatory scrutiny, and makes it difficult for the franchisor to change policies and fees. A new franchisor, for example, cannot easily change the contract provisions regarding the support services it provides. It cannot treat new franchisees differently from existing franchisees by establishing new provisions for them, or it risks law suits from either the new or old franchisees. Consequently, if the new franchisor does not get the system model right initially, it is not likely to become successful over time.

Joe established the right franchising policies initially. Since the policies were self-reinforcing, Newfran's success fed on itself. Joe's colleagues, who chose wrong policies, could not extricate themselves and eventually failed as franchisors.

Summing Up

Once entrepreneurs have decided to develop a new franchise system, they should expand rapidly. Initial policies should be established in line with the success model, and not altered. Many local entrepreneurial tasks should be delegated to franchisees, reducing services and the number of headquarters staff required. Membership of the Franchise Association is encouraged, and where appropriate registration with the authorities is important. Finally, new franchisors also need to show a trustworthy nature and high quality to potential franchisees.

Like most new businesses, only a small number of new franchisors are likely to survive over time. This high failure rate makes it important for entrepreneurs who start franchise systems and entrepreneurs who buy into them to identify the policies that enhance the long-term survival. While the policies identified in this study do not guarantee success, they increase a system's prospects.

Scott Shane is the Leghorn Career Development Assistant Professor of Entrepreneurship at MIT School of Management. Chester Spell is now Senior Lecturer at the University of Waikato. This article was first published in the Sloan Management Review, Spring
Order a Print Copy
Order a Print Copy