Abstract
In this article, we seek to apply the insights of recent research on routine to the context of repeated negotiations. To demonstrate the link between both concepts, we introduce an analytical framework in which we identify different negotiation situations in which routine can develop. We distinguish two dimensions of the negotiation process: a problem‐solving dimension and a communication dimension. Our framework for analyzing the role of routine in negotiation is built around these two dimensions. We define those skills that we argue in repeated negotiations can help negotiators manage particular kinds of negotiations depending on the level and type of routinization that type of negotiation involves. Moreover, we demonstrate that our framework is inherently dynamic, which we illustrate with simplified business examples.
Introduction
It obviously makes an important difference whether we negotiate something or with someone only once, or whether we do so a second or repeated time. Various aspects of repeated negotiations have been discussed extensively in the literature. The most essential of these include: negotiation in relationships (e.g., Greenhalgh and Chapman 1996; Sheppard 1995; Sheppard and Tuchinsky 1996), establishing trust over the course of multiple negotiations (e.g., Lewicki and Bunker 1995, 1996; Ross and LaCroix 1996), and the development and importance of reputation (e.g., Ferris et al. 2005; Goates, Barry, and Friedman 2003; Lewicki and Tomlinson 2003). What has not been examined so far, however, is the development of routine in negotiation.
A general perception of the phenomenon of routines can be traced back to the early days of economic analysis (e.g., March and Simon 1958; Schumpeter 1987). Nevertheless, the first systematic approach to routine was its employment as a core element of the evolutionary theory of economic change articulated by Richard Nelson and Sydney Winter (1982). Since then, routine has gained some importance in the research on organizational learning (Cohen 1991; Zollo and Winter 2002) and strategic management (Eisenhardt and Martin 2000; Teece, Pisano, and Shuen 1997). But the findings of the research on routine have not yet been applied to negotiation analysis.1
In this article, we seek to connect these two streams of literature and to investigate what happens when negotiations become routine. We are convinced that routine has an important impact on negotiations, particularly because, in this context, it has a double‐edged meaning. On the one hand, through routine, negotiators develop their capabilities, which can improve their efficiency. But on the other hand, routine also increases the likelihood that negotiators will act in a rote or automatic way, to fall back on familiar patterns and overlook important changes in the negotiation environment and fail to adapt to them. Thus, routine poses the risk of diminishing negotiator effectiveness.
In the second part of this article, we briefly introduce the concept of routine. In the third section, we investigate how routine can be acquired in repeated negotiations. Based on various definitions of negotiation, we distinguish two dimensions of the negotiations process: a problem‐solving dimension and a communication dimension. Each dimension is affected differently by the degree of repetition involved in a particular negotiation situation, depending in particular on the degree of substance standardization in negotiated agreements and on the variability of agreement partners.
In the fourth part of the article, we discuss how routine can be applied in repeated negotiations and discuss the reasons one would choose to apply those routines and the possible consequences of doing so. We extend our framework by introducing the concept of a critical skill set for dealing with the issue of routines in negotiation.
In the final section, we discuss the dynamics and outcomes of our framework. We also trace the development of repeated negotiations within companies, point out the situations in which the kind of momentum our framework describes can and should be reversed, and explain the reasons for this phenomenon.
Defining Routine
Routine can be basically understood as a strategy of dealing with the limits of human reasoning or “bounded rationality” (Nelson and Winter 1982) A necessary precondition for the acquisition and application of routine is repetition (Kesting 2006; Pentland 2003). Before someone undertakes an act, she first has to decide what she would like to do and how she will do it. This decision can be conscious and planned, spontaneous, or even instinctive, but it must be made if the person actually acts. In this article, we call the result of such a decision a “solution.”
When an act is repeated, a person can usually remember both his initial solution and his reasons for taking that particular action, but he also has an opportunity to observe that action's results. Therefore, at this stage he knows whether his initial solution was effective or not. If his deliberation was scarce, however, and produces opportunity costs (because our limited reasoning and analytical capabilities could be used elsewhere), it is rational for this person to apply former solutions to current problems as long as he was satisfied with their results. We call the application of former solutions to current problems “replication.” If a person is dissatisfied with the outcome of his solution, it might be rational for him to invest additional planning effort and partially or completely revise his previous solution (Betsch and Haberstroh 2005; Betsch, Haberstroh, and Höhle 2002).
Against this background, we define routine as a person's ability to substitute deliberate planning and decision making with rote replication.2 These replicated solutions substitute for planning when they determine the execution of the respective act (which is also the very meaning of planning and decision making).3 By substitution we do not mean instant and complete replacement but rather the gradual exchanging of parts or aspects of the act. By ability, we do not mean the mere capacity to replicate the act in any manner and irrespectively of its results but instead replication in a well‐founded manner. The less planning necessary to meet our aspirations, the greater our ability to replicate the act. Likewise, the better the result we can achieve without planning, the greater our ability to replicate the act.4
The core of routine is replication. But, as we argued previously, a necessary precondition for replication is repetition. The more specific the repetition, the more comprehensive the routine that one can acquire andapply. Bearing this in mind, we are using repetition in the sense of similar repetition.
Acquiring Routine in Negotiations
Negotiation has been defined in many different ways. P. H. Gulliver (1979), for example, defined it as an information exchange process established to reach a certain outcome. In Getting to Yes, Roger Fisher and William Ury (1981) defined negotiations as “back and forth communication designed to reach an agreement” (Fisher, Ury, and Patton 1991: xvii). A common characteristic with these and many other definitions (e.g., Cross 1977; Zartman and Berman 1982; Lewicki 1992) is that all of them implicitly distinguish two dimensions when describing each negotiation.
The first dimension is a problem‐solving dimension (“to reach an agreement”).5 In game theory, negotiations are modeled as games in which the players try to select a strategy that maximizes their individual utility by choosing the most appropriate element out of a set of all feasible strategies. This decision is strategic, because the payoff for each player depends on the strategies pursued by the other players, and often cooperative, because to maximize their individual utility the players may choose to coordinate the selection of their strategies. Accordingly, problem solving in negotiation can mean avoiding inefficiencies (Raiffa 1982: 148), identifying “win–win” situations (Fisher and Ury 1981), creating value (Lax and Sebenius 1986; Mnookin, Peppet, and Tulumello 2000), and so on. This problem‐solving aspect is closely connected with the substance of negotiation.
The second dimension of negotiation is a communication dimension (“back and forth communication”). Communication, both verbal and nonverbal, is a critical element of each negotiation (Putnam and Poole 1987). It allows for exchanging offers and counteroffers (Tutzauer 1992), coordinating decisions (Jönsson 2001), and conveying information about alternatives and outcomes (Lewicki, Saunders, and Barry 2006). Negotiations usually have a multitude of possible solutions. The selection of a mutually acceptable final solution is conducted through the communication process between the negotiating partners. As such, the negotiating partners determine this communication dimension.
The concept of routine becomes relevant for negotiation whenever a negotiation is repeated. “Repeating” a negotiation, however, can mean different things. In the broadest sense, one can say that repeating a negotiation means conducting any second negotiation, independent of the context in which the negotiations take place, that is, independent of its substance and partners. But as noted earlier, similarity between repetitions is a prerequisite for routine acquisition. The insights we acquire about one negotiation partner do not necessarily hold true for another negotiation partner. On the contrary, different negotiators tend to act and react rather differently. Moreover, feasible options and optimal solutions to one problem usually cease to be optimal or even feasible in a different problem setting. Nevertheless, unspecific repetition is an important way of building and improving negotiation skills. (We will address the topic of improving negotiation skills through repetition in the following section.)
Negotiations become more similar to each other as more of their elements remain stable in the course of repetition. These elements fall into two major categories: substance and partners.
If the substance (i.e., issues, interests, positions) of the negotiation is standardized and does not vary substantially through repetition, negotiators have an opportunity to improve their specific knowledge (i.e., they learn) about this very substance in the course of repetition:
As a result of the initial planning, negotiators have already identified and characterized feasible alternatives and reached an agreement. With repetition, their familiarity with the issues themselves increases.
With repetitions, negotiators increasingly observe what different alternatives look like and what results they lead to (at the beginning, they were only able to guess). They even have the opportunity to try out different feasible alternatives in the course of repetition and to confront their assumptions within the reality of the market.
If the other party to the negotiation does not change, a negotiator has the opportunity to improve her specific knowledge about this party via repetition:
She learns how the other party acts and reacts in different situations.
She learns about negotiation strategies and techniques the other party uses in different situations and how to properly address these.
Developing this specific knowledge about the substance and partners enables negotiators to replicate parts of negotiations and sometimes even complete negotiations.6
As a result, the development of negotiation routines can be summarized in a matrix that categorizes repeated negotiations according to the degree of substance standardization in negotiated agreements and the variability of agreement partners7 (see Figure One).
We describe negotiations in which varying partners attempt to reach agreements over different issues each time as “no‐routine cases.” Such negotiations offer few opportunities for learning via repetitions because no specific knowledge about negotiation partners or negotiation substance can be acquired that has any relevance beyond that single negotiation. But this situation is not especially common in business. Negotiations involving the initial operations of start‐up companies might qualify as no‐routine cases. The work of consulting companies may sometimes also fall into this category, although consulting companies often repeatedly work in suchspecific areas as restructuring or outsourcing, building long‐term relationships with clients in the process.
On the opposite end of the spectrum we find repeated negotiations in which a constant pool of partners negotiates over very similar types of agreements. In this case the interactions remain very much alike, which suggests that the repetitions may lead to routine development. In business, examples include well‐established business relationships between a producer in a mature industry and his or her suppliers delivering standardized raw material. Collective bargaining negotiations in which the same employer bargains with the same union every three years or so over similar contractual issues would also be an apt example for this case.
In other cases, the parties will remain stable but the issues will vary. Such situations are also relatively common in business. Once a business relationship has been established, it will often be managed by designated employees acting as contact partners for each organization. Over time, the relationship will develop and may change, extended to include new products and/or services, for example, giving rise to new negotiations between the same parties. This may require the renegotiation of the existing contracts or even the signing of new ones.8
The fourth combination shown in our matrix occurs when the substance of the negotiation remains the same but each time involves different partners. If a particular employee is assigned a responsibility to manage thesales of a given standardized product or service, this person will face exactly this situation.
Applying Routines in Negotiations
Let us start with the no‐routine case in which no specific knowledge, applicable to the current negotiation, has been previously acquired. Because both the parties and the substance are new, the current negotiation is being conducted for the first time. The negotiator cannot build on previously acquired specific knowledge, and thus must negotiate in this situation “from scratch.” In this sense, the negotiation involves deliberate planning, which is the traditional approach to negotiations. According to the traditional approach, success in negotiation depends on a set of various skills. Following the general structure of negotiations introduced in the previous section, these skills, again, can be grouped into two different classes.
The first class comprises creative problem‐solving and analytic skills. Proficiency in these areas is a condition for problem‐solving effectiveness. Such skills include in particular:
the ability to determine the interests of all parties, which can be difficult, especially when the parties comprise individuals with heterogeneous interests (von Wright 1987);
the ability to identify each party's feasible alternatives, which depends on one's ability to acquire and process information (Simon 1955 and 1956);
the ability to assess probabilities and the consequences of possible decisions for different possible contexts (Jeffrey 1983); and
the ability to reduce or deal with complexity (Raiffa 1982 and 2002).
The second class includes negotiation and communication skills. Proficiency in these skills is a condition for good performance on the communication dimension, and therefore they have been extensively examined in the negotiation literature, in particular:
the ability to identify and pursue an appropriate negotiating strategy (Pruitt 2001);
the ability to perform and react to different negotiation techniques or so‐called “negotiation gambits” (Dawson 2003);
the ability to manage, recognize, and understand nonverbal communication (Burgoon, Buller, and Woodall 1996; Wheeler 2004); and
the ability to develop relationships and build trust (Thompson 2005).
It is noteworthy that these skills are — at least partly — based on individual experience, which, again, is an outcome of “repetition.” Butcontrary to the development of routine, this “repetition” is necessarily unspecific. The more diverse a negotiator's experience is, the more sophisticated his skills become. The result of this kind of experience is not that one becomes familiar with one particular but rather with the whole range of possible negotiation partners and/or negotiation substance.9
The dynamic is different when a negotiation is repeated specifically, that is, when certain aspects of a former negotiation occur unchanged in a current negotiation. In this case, it is possible for the negotiator to apply the specific knowledge she has acquired about these unchanged elements from the former negotiations. For instance, if certain details of a previous deal, such as payment and delivery conditions, have already been satisfactorily settled with the other party, it is possible simply to apply this settlement in the current negotiation.
In this context, applying routine means nothing more than using specific knowledge about former solutions to solve current problems. It is rational to do so for two reasons. First, doing so can be expected to improve the efficiency of the result. In the course of repetition, negotiators learn about the consequences of their actions and learn to avoid mistakes. It is rational to utilize this knowledge in the course of repetition. This is the “learning by doing” aspect of routine.
The second reason, however, is that applying routine helps us save scarce planning capacities and time. Applying the solution to a former problem means that we do not have to go through the whole planning process again to solve the current problem, or, as the cliché goes, we can avoid “reinventing the wheel.” In this way, routine has the capability to substitute deliberate planning.10 Against this background, routine can indeed be understood as a very effective strategy to deal with our given bounds of rationality.
Consequently applied, developing routines can help negotiators develop best practices. In general, a “best practice” is a comprehensive preexisting answer to a certain problem. For our purposes, a best practice is a “program” for acting in a specific negotiation situation that may not necessarily lead to the optimal solution but to a solution that, at least, meets the negotiator's aspirations (Feldman 2000; Kesting 2007). Best practices take into consideration all accumulated deliberations that have been conducted and experiences that have been made in the course of repetition. Therefore, it is possible that the use of best practices can create greater efficiency — even more than is achieved by the research and planning that goes into the first negotiation.
The use of best practices provides a sensible solution to decision problems in cases where we expect that further planning would not pay off. (Additional planning and strategizing entail accruing opportunity costs because our limited reasoning and analytical capabilities could be used elsewhere.) In this situation, best practice has the potential to reduce the effort that goes into planning to a minimum.
The development and implementation of best practice guidelines is of fundamental importance for each organization seeking to benefit from the opportunities that negotiation routines offer. It becomes especially important when negotiation systems are designed and/or restructured. For instance, in 2002 Motorola Inc. decided to implement a new procurement system, the Motorola Internet Negotiation Tool (MINT). Between 2002 and 2005 the company negotiated more than $16 billion online, saving more than $600 million.11 Motorola managers credited such key success factors as the company's devotion of resources and time to train the purchasing staff and suppliers prior to as well as after the rollout. In other words, Motorola made sure that the best practice guidelines were established and conveyed to those who needed to master them.
But using routines also has a downside. Routine‐controlled acts in general, and best practices in particular, are seldom fully adapted to context changes for two reasons:
Using routines to achieve efficiency in thinking and planning (i.e., “not having to think about it again”) means that, once they are established, solutions are not questioned anymore. This systematically excludes any comprehensive proof of whether the current repetition really works within the current context.12
Those who study routine generally assume that highly routine‐controlled best practice has the tendency to become automated (or rote, or “patterned”) and subconscious (Nelson and Winter 1982; Zollo and Winter 2002; Kesting 2006). This not only shifts the attention further away from best practice, but furthermore, it makes one's established best practice inflexible and resistant to change.
The consequence of both is that best practice can turn out to be insufficiently adapted to context changes and, hence, always carries the risk of becoming ineffective. This is the case when routine becomes a trap (or a “box,” as in “thinking inside/outside the box”).
Let us apply the outcome of our analysis to the framework we introduced in the previous section. The resulting matrix (Figure Two) illustrates critical skills sets in repeated negotiations defined as a class or as classes of skills a negotiator should possess in order to deal efficiently with a particular negotiation situation.
We will first have a look at the no‐routine case in the top‐left cell of the matrix. Professional negotiators often confront this situation. Even a negotiator who is equipped with a full set of negotiation skills will find it difficult to develop and apply efficiency‐increasing best practice in no‐routine cases. In business organizations, negotiations in no‐routine cases often call for the involvement of highly skilled employees who are able to cope with high variability on both matrix dimensions. Accordingly, scarce and costly high‐end management capacities may be required in these cases.13
The need for critical skills diminishes when one of the matrix dimensions, either the substance or the partners, becomes fixed or standardized. Let us first consider the case of repeated negotiation interactions with the same partners over various issues, represented by the bottom left cell of the matrix. Once a business relationship is established with the intention of sustaining it long term, the representatives of each business will likely encounter each other repeatedly at the negotiating table. As their interactions repeat, they will develop routines with each other, and their whole process of communication is likely to become routinized to some extent. In this situation, we believe that creative problem‐solving and analytical skills will be more critical than interpersonal and communication skills because the relationship is already established.
The partnership of Wal‐Mart and Procter and Gamble is a good example of such a negotiation situation. In the 1980s, both companies realized the importance of sustaining a good long‐term relationship with each other and implemented joint teams to manage their supply chain. This operation allowed both companies to routinize communication flows and concentrate on solving joint problems.
A similar, if opposing, dynamic is at work when repeated negotiations take place between varying partners over standardized issues. The top‐right cell of the matrix illustrates this situation. These cases often involve a standard form of agreements or even contracts of adhesion (“take‐it‐or‐leave‐it” contracts that offer little or no negotiating room), and only minor modifications reflecting specific partner needs are allowed. Precisely because of this, creative problem‐solving and analytical skills are less critical to success in this kind of arrangement. Interpersonal and communication skills, on the other hand, could help a negotiator establish trust and could persuade the other side that the proposed deal is highly attractive, despite the low level of possible customization.
The sales operations of automobile dealers are excellent examples of such situations. Although some degree of customization is available, the sales items (cars from a specific automaker) are generally relatively standardized and so is the substance of the negotiations. But for dealers, the parties with whom they negotiate vary greatly, comprising a wide swath of the general public. Successful automobile sales people can be expected to possess strong communication and negotiation skills but have less need for advanced analytical skills or substantial creativity.
When the substance is standardized and the parties are stable, it is easier to develop routines for both of those dimensions, routines that can then be formalized in the form of best practices. Once best practice guidelines are formulated, their strict execution will maximize efficiency. In these situations, routine allows us to reduce the complexity of repeated negotiations to simple sets of procedures that can be followed even by relatively unskilled negotiators, so comparatively, negotiations of this type demand fewer costly resources from a company. Negotiations involving media companies (television and radio stations, newspapers, and magazine publishers, for example) and the media agencies that buy advertising from them fall into this category. The universe of potential negotiation parties is small and usually stable, and the negotiation's substance highly standardized, for example, advertising time or space. Relationship management is, for both parties, a largely administrative job carried out based on best practice rules established in the past. In this case, the entire negotiation effort can often be reduced to a minimum, consisting only of brief agreement adjustments to recent changes (e.g., in schedule, rates, etc.).
Model Dynamics and Their Implications
The position of any given business organization within the matrix will vary depending on the particular negotiation. This inherent dynamic affects both how these organizations develop routines on the one hand, and how they react to context changes on the other.
Routines require time to develop. At any given organization, the first repetitions of a negotiation process will resemble the situation depicted in the top‐left cell of the matrix that we call the no‐routine case. At some point in time, both the substance and the partners are new, and even for experienced negotiators, application of a routine developed over the course of previous negotiations can lead to misunderstandings and prove to be ultimately ineffective. In these cases, the negotiator with the most highly developed sets of skills will be most effective.
On the other hand, many business activities are naturally stable and many negotiations become repetitive, with the same substance and/or the same partner (e.g., involving products manufactured in a more or less standardized series or in established long‐term relationships with suppliers, customers, partners, or even the government). Moreover, researchers have repeatedly pointed out the benefits to be gained from process standardization.14
The first step toward standardization of processes is to establish stable relationships with an optimal number of business partners. After relationships with suitable business partners have been established and maintained, routine building can begin. For example, Ford Motor Company has recently announced a plan to offer fewer key suppliers long‐term contracts and to work with these suppliers cooperatively on design and cost reductions (McCracken 2005). The downward movement from the initially occupied top‐left cell of our matrix illustrates this process. Very often, if the overall number of negotiation counterparts is relatively small and stable, the task of managing relationships with a particular party or parties is assigned to a particular person (e.g., a key account manager in a sales department). This person becomes responsible for all issues connected with that business relationship, including negotiations of various deals.
The second approach to processes standardization is to establish a well‐defined and stable negotiation‐based process for such functions as sales and procurement, with the consequence that the substance of potential deals with business partners becomes standardized (even though the set of partners might vary). Google Inc. is a good example of an organization with a well‐defined product portfolio. So far Google's strategy has been to focus on delivering fast and reliable tools for searching and retrieving information from the Internet. Unlike such Internet portal operators as Yahoo or Lycos, Google has largely generated revenues from the sales of a limited number of search‐related products.15 This process could be drawn as a migration from the top‐left cell of our matrix to the top‐right cell. Whenever such a process takes place, the negotiator finds herself in a situation in which routines involving actual substance are fully developed in such a way that the negotiator will more likely utilize communication and negotiation skills, with less need for highly developed analytical skills and creative problem solving.16
In the initial repetitions of negotiation, the set of critical skills the negotiator will need to call on will by necessity be relatively inclusive (communication and negotiation skills, as well as creative problem‐solving and analytical skills). This indicates that at this stage, the negotiator with the most comprehensively developed set of skills will be best able to deal with the high degree of complexity inherent in this type of negotiation. If with repetition, any of the dimensions becomes more fixed and/or standardized, the need for the negotiator to have such a comprehensive critical skill set diminishes. This dynamic has obvious ramifications for the processes of task delegation and assignment as well as training and even hiring. The critical skills required for different types of negotiations should obviously be considered when determining the best candidates for taking on particular kinds of negotiations.
Clearly, as a particular negotiation relationship becomes more routine, responsibility for managing that process is likely to move around within an organization, often being delegated further down the “corporate ladder,” from, for example, the chief executive to a sales director, then from a sales director to a key account manager. Of course, best practice guidelines can be documented in writing (guidelines, company policies, manuals), but because routine has a significant tacit element, it is unlikely that such guidelines could capture the relevant knowledge in its entirety. Hence, organizations confront a challenge in developing the appropriate channels for transmitting skills and knowledge developed through routinization throughout the organization.
As we noted earlier, however, repetition presents the risk that routine acts will become increasingly subconscious and patterned, and that negotiators who develop such routines will become inflexible and ineffective when changes on any of our matrix dimensions occur. Such a situation could occur for instance, when a new negotiator is introduced to represent the interests of a long‐term customer. Routine acquired through repetitive negotiation with the person's predecessor may thus become obsolete, especially if the new person demonstrates a completely different approach to negotiation.
An unanticipated change in the substance of an agreement can also present problems for a negotiator who has fallen into a routine. For example, if a negotiator tacitly assumes that established conditions with a particular supplier (payment and delivery, for example) are given and fails to continue to negotiate them (or negotiates them only superficially), he may be surprised and unprepared when the other party proposes an unexpected change. Another risk is that both parties will become so comfortable in the existing arrangement that they fail to develop additional opportunities to create value by altering the terms of their agreement as relevant circumstances outside the agreement (externalities) change.
Negotiators must deal with routine very carefully. The kind of negotiation changes that we illustrate with our matrix are most likely to occur when:
new counterparts are identified (e.g., as a result of geographic expansion or an extension of the product portfolio);
the identity of the parties becomes unstable and relationships diminish in duration;
new negotiation issues are introduced (e.g., through innovation‐based extension of the product or service portfolio);
the elements of the negotiation's substance become highly customized (e.g., as a consequence of intense market competition); and
the external environment changes (regulatory changes, changes in cost and availability of resources, consumer‐market driven changes, etc.).
Whenever any of these changes are identified, the previously described dynamics should be reversed (as represented by moving to the right, the top, or to the top‐right on our Figure Two matrix). In such situations, depending only on routine and on resulting best practice rules established over time may produce highly ineffective outcomes. Therefore, to assure its permanent usefulness, routine needs to be regularly monitored and occasionally modified or even established anew in order to help obtain desired efficiency gains in new conditions.
The challenge for businesses is to recognize promptly which cell of the matrix most accurately describes a particular relationship with a particular business partner and to assign to that relationship the employees with the most appropriate set of skills to deal with them and/or to develop the needed skills in those employees already involved in those particular negotiations.
It is also possible that a particular company may be confronted with a situation that falls into more than one of the categories in our matrix. Such situations should be reflected in the organizational structure as it is, for example, in the case of Lycos Europe where two units — sales and business development — were established to deal with two such different types of negotiation relationships. The responsibilities of the sales department are limited to establishing initial contact with potential customers and offering them packages consisting of the standard Internet advertising packages (such as banners, buttons, key words, etc.) for deals up to a certain value. The members of the business development group then negotiate all deals exceeding this threshold, as well as those that include nonstandard advertising solutions and strategic partnerships. The sales units are located close to major customer clusters in major cities, whereas business development is located at the European headquarters in Germany, where a small group of people are responsible for large pan‐European deals. In terms of skills, business development managers can be safely considered “complete negotiators” whereas in recruiting key account managers for the sales department, the emphasis is on communication and negotiation skills. The size of both units varies over time, but the sales unit has always had a larger staff.
Conclusions
In this article, we have applied the findings of routine theory to the investigation of repeated negotiations. We argue that negotiators can acquire routine in the course of repetition with respect to two different dimensions of a negotiation relationship, the substance of the deal (interests, positions, etc.) and the identity of the parties themselves. On this basis, we have developed an analytical framework to illustrate the acquisition and application of routine in negotiations and the inherent dynamics of this process.
What practical advice can be derived from our research? Routine theory assumes that through planning and performing an act, a person acquires specific knowledge about this very act. We call this learning acquisition of routine. In this sense, explicit and implicit knowledge is the substance of routine. This learning is the engine of all changes, making repetition also relevant for negotiations.
According to routine theory, repetition matters. The analytical framework we developed allows the analysis of some of the ways in which negotiation can change in the course of repetition. So far, the analysis of repeated negotiation has been largely restricted to such aspects as reputation and trust. In contrast, we argue that examining negotiations through the lens of routine offers insight into all relevant aspects of repeated negotiations by allowing us to model the development of knowledge as it is applied in repeated negotiations.
We argue that it makes an important difference if we negotiate a particular issue for the first time or if we do so a second or repeated time. Repetition has an important effect on how negotiations are carried out and, thus, becomes a relevant factor for negotiation analysis.
These theoretical insights have important practical consequences. We consider the following to be some key lessons for practitioners that can be derived from our research.
It is important to analyze the structure of conducted negotiations and, in particular, to check how standardized the substance of these negotiations is and how stable the set of partners (Figure One). This analysis will indicate whether these negotiations display a potential for developing routine.
Routine increases efficiency and saves resources. Therefore, it is useful to standardize processes, to actively develop routine, and to establish best practice guidelines for dealing with those negotiation situations that display stability in repetitions.
The organizational structure should be adjusted to the types of negotiations (according to the dimensions of our matrix) conducted within the organization. Often it is advisable to form specialized organizational units to deal with different types of negotiations.
Routine is capable of substituting skills. Even complex negotiation settings do not necessarily require fully skilled negotiators if sufficient routine has been acquired in the past. Therefore resources can be saved if employees with matching skills are hired and/or assigned to conduct appropriate negotiations (Figure Two).
Changes in the environment can make routines obsolete and ineffective very quickly. Therefore it is crucial to implement control mechanisms monitoring existing routines and assuring that they do not become dysfunctional due to overlooked changes in the environment.
Although the study of routine has been around for some time, applying it to research in negotiation theory is a new and promising undertaking. Most of the concepts we note in this article, such as the link between routine and organizational structure and the development of routine‐control mechanisms, are new, and additional research into these areas would be worthwhile; and few empirical studies have been undertaken on these topics.17 We hope, however, that because of the practical relevance of this topic, these research gaps will soon be closed.
NOTES
Repetition of strategic interaction has also been a vast topic in game theory literature. Some of the aspects discussed there include: solution concepts in repeated games (e.g., Abreu 1988; Fundenberg and Maskin 1990), emergence of cooperation (e.g., Axelrod 1984; Kreps et al. 1982), reputation (e.g., Abreu and Gul 2000; Fundenberg, Kreps, and Maskin 1989), and information and learning (e.g., Blume and Easley 1995; Fundenberg and Levine 1998). Because of its underlying rationality assumptions, however, game theory is of limited relevance to our analysis.
An alternative would be to define routine on the physical level of action taking as a “pattern of activity” (Dosi, Teece, and Winter 1992; Feldman 2000). But this approach makes it difficult, if not impossible, to investigate the interplay of deliberate planning and subconscious repetition in action taking (see also the discussion in Kesting 2007).
Such solutions usually materialize the results of former planning and experiences. But they can also be acquired by observation and imitation.
For a detailed elaboration of the routine concept see Nelson and Winter (1982), Winter (1986), Cohen et al. (1996), Dosi, Nelson, and Winter (2000), and Kesting (2006 and 2007).
Problem solving can also be considered one of the negotiator's strategies (Pruitt 2001). It is demonstrated when negotiators try to identify options that satisfy both parties' goals. This strategy has also been labeled as collaboration (Thomas 1976), integrative bargaining (Walton and McKersie 1965), or creating value (Lax and Sebenius 1986; Mnookin, Peppet, and Tulumello 2000).
It is, however, important to note that learning in general and routine development in particular is rather difficult in negotiation settings for at least three reasons. The first of these is the incompleteness of information and the absence of structured feedback. Revealing certain information in negotiation might be highly disadvantageous (e.g., reservation point), and, therefore, the parties usually release such information in a selective and strategic way, if at all. Second, truthfulness in negotiation cannot be assumed. Although truthful revelation of interest is one of the fundamental preconditions for value creation in negotiation, very often it is not met by the negotiating parties. Third, revealed information is often unintentionally misinterpreted. Three decades ago, Amos Tversky and Daniel Kahneman (1974) demonstrated that in many situations our perception and interpretation of received information might be highly distorted (cognitive biases). Additionally, we also need to remember that the repetitions of the negotiation process might occur in significantly long time intervals and therefore what we learn in the current negotiation setting might be partially or completely irrelevant in the future ones.
Other related aspects such as the variability of the process or changes in the environment are also incorporated in the framework because they only become relevant for negotiations when they affect one or both of its dimensions: substance or partner.
A very interesting and important aspect of our framework, which is not discussed in this article in detail, is the process of joint (interactive) learning and routine development in negotiation. Our framework looks at selected aspects of repeated negotiations from the point of view of a particular party. It is, however, clear that once the set of partners becomes stable, the learning process along this dimension will be highly interactive for all involved parties. This means that during consecutive repetitions of a negotiation process, the parties learn from each other, which then leads either to establishing joint routines or to close coordination of their individual routines. This type of learning is completely different than the learning with respect to standardized substance, which usually does not display interactive characteristics. We believe that this topic is extremely interesting and highly relevant, and deserves to be analyzed in more detail in a separate paper.
Moreover, it is generally assumed in the literature that individual skills are also outcomes of systematic training (Meerts 2001; Winham 2001). Training of negotiators is usually based on unspecific repetitions in artificially created contexts (e.g., simulations).
Rackham (1980) describes empirical research results demonstrating similarities and differences between skilled and averaged negotiators with respect to planning. These results, although very interesting, do not consider either of our framework dimensions and therefore cannot be directly compared to ours.
See Metty et al. (2005: 7).
Of course, it still allows for building in some kind of controlling procedures, indicating when it is necessary to turn away from routine control and back to conscious action planning. Finding reliable mechanisms, however, often proves difficult.
In this and the following paragraphs we assume that the respecting negotiations are not minor but have a significant importance for the company. Otherwise, such a narrow attribution of responsibilities appears unnecessary.
A pioneer in the field was the Boston Consulting Group, which articulated the idea of the learning curve in 1972 in their client engagements.
Such as, for example, Google AdWords and Google AdSense. See http://www.google.com/ads/index.html, January 2006.
Sometimes it may be useful to speak also about the standardization of the negotiation process itself, especially in the context of designing or reengineering it (Doctoroff 1998; Sander and Bordone 2005). However, as mentioned earlier, for simplicity reasons we decided not to introduce process as a separate dimension, but rather treat it as a part of the existing ones.
Several psychological experiments illustrating the existence and the implications of mindlessness are quoted by Ellen Langer (1989). Although the concepts of mindlessness and routine seem to be related, they display significant difference in terms of their central focus. Mindlessness concentrates on lack of situational awareness and conscious control as well as automation in task performance whereas routine can be also consciously acquired and applied.