The Benefits of Linkage Analysis

Linkage analysis is the next step from intuition-based or inflexible mental model–based decision making. It takes an organization beyond just using metrics in the form of dashboards or scorecards to understanding the complex relationships among the drivers of business success. In our view, linkage analyses provide an antidote to the problems that are associated with simple, intuition-based mental models. As we noted earlier, organizations are becoming increasingly complex, more geographically distributed, and often dependent on a consortium of loosely linked entities that jointly produce the outputs of interest. In such complex environments, simple decision rules such as mental models, fundamental intuition, scorecards, or rules of thumb may be insufficient or suboptimal in driving strategic choices and business outcomes. In most cases, managers have to deal with several moving parts, and it might not be feasible or easy to comprehend how they move and work together. Therefore, managers might ignore parts of the problem that are difficult to comprehend or for which reliable data might be missing. Alternatively, they might verify some of these relationships on a one-time, ad hoc basis and rely on them over time and across problem classes.

Linkage analysis provides a rational and fact-based alternative to these suboptimal options and allows organizations to draw meaningful relationships among the various moving parts of an organizational system. For example, it allows firms to see if actions designed to boost customer loyalty indeed raise revenues and lower customer acquisition costs. It also allows them to observe whether the nature and size of these relationships is stable or changing over time. Overall, while there are several major advantages of using linkage analysis to support strategic decision making, we believe that some of the critical ones are described here.

Promoting a Culture of Verification-Based Management

One of the indirect but extremely important benefits of linkage analysis is that it builds and promotes a culture of verification-based management. Employees at all levels in the organization understand that decision making is driven by data and verifiable relationships that can be tracked over a period. Over time, the use of a broad swath of data to drive the organization becomes an integral part of it and is implicitly absorbed into its culture. This suppresses gut-feel or beliefs-based choices and tends to drive the organization toward superior decision making and closer to optimal choices.

Formalizing Learning

Organizations that adopt linkage analysis as an integral part of how they make strategic choices tend not to think of these analyses as a one-time, isolated event. Over time, these analyses tend to increase in scope and uncover linkages that were until then not understood and discovered. Second, tracking systems become part of a culture of verification-based management and key linkages rather than key metrics are watched using repeated measures. Both a repeated application of linkage models and the increase in the scope of these models over time provide information through linkages that the organization can learn from. Managers and executives are able to learn about critical linkages that drive firm performance, the best measures for each metrics involved in the linkages, and the evolution of the web of linkages over time.

Working With Causal Webs

In common parlance, when we say cause and effect, we seem to give an impression that the two are directly and closely connected. However, for most higher order strategic problems, the connections are neither direct nor extremely close. Therefore, managers and executives need to be able to conceptualize these causal relationships in terms of a web of interconnected variables that may demonstrate a complex set of mutual dependencies. We find that without the aid of an appropriate set of tools, and the learning that comes from their repeated application, it is difficult for managers to think, plan, and strategize in terms of webs. Outputs from linkage analyses, especially when presented in visual and graphic form, tend to promote strategic thinking in terms of a web of relationships, not merely direct cause and effect. Therefore, managers working with linkage analyses are able to concurrently account for a larger number of factors, issues, and stakeholders in their decision-making process.

Testing and Rebuilding Mental Models

In our experience, we find that most organizations invest little in verifying the mental models that they assume drive the dynamics of their business. And even in the rare case where this happens, it is done in a piecemeal rather than a systematic way. In either case, organizations end up knowing very little about the entire set of strategic flows that drive their performance. Linkage analysis helps build integrated causal chains that enable organizations to connect more dots in their strategic mind space. As a result, when organizations embark upon these efforts, they tend to gain in one of two ways. They either verify their mental models by seeing them laid out and tested in their entirety or learn to challenge them by trying to reconcile them with what the data reveal. In either case, the net result is a model that data have either validated or challenged and that increases management’s confidence level in whichever strategic route it chooses to follow.

Understanding Strategic Flows

As discussed earlier, the impact of pulling a strategic lever is often complicated and cascades through to its ultimate effect through a number of complex relationships. An absence of linkage analyses forces managers to make assumptions about these relationships. For example, an organization might change its innovation strategy from incremental innovation to radical innovation. This single change may have far-reaching consequences and affect the operations of the organization and the performance metrics in a variety of ways. Without the aid of an appropriate tool, it would be impossible for the managers or leaders of the organization to comprehend or measure the effects of changing the innovation philosophy. Linkage analyses provide the set of tools that would help management understand the “flow” from cause to effect through the various intervening relationships. They also measure the direction and size of these effects and help test both intuitive and formal hypotheses.

Exploiting Data Explosion

Because linkage analysis is data driven, it tends to put vast warehouses of data at organizations to good use. As was discussed earlier, many organizations are sitting on a goldmine of data from various sources including customers, employees, core operations, and financial operations. In many cases, these data are residing in various functional silos and few understand the power from integrating them. Linkage analysis provides a platform to break these silos, connect data from various functional areas, build cross-functional models, and discover the true drivers of firm performance. To that extent, linkage analyses help leverage the power of existing data and connect the dots to build strategic pictures on a larger canvas.

Uncovering Strategic Fulcrums

While many relationships or linkages may connect organizational actions to outcomes, we often find that actions directed at a small improvement in one metric or linkage relationship result in a large improvement in the performance of the ultimate outcome. In some firms, the key action is surrounding key hires, while in others it is retaining key core customers. In some, it relates to maintaining the equity of the key brand, and in others at maintaining a cost advantage on a key raw material. We refer to these critical metrics or relationships as the strategic fulcrums for the terminal outcome. Small changes around the strategic fulcrum result in large changes in performance. In other words, we can imagine the organization’s success to pivot around is strategic fulcrums. The correct management of metrics around a strategic fulcrum has a high magnification effect on the organization’s fate. Consequently, these fulcrums should be the locus of strategic activity.

In our experience, however, while several firms intuitively recognize that certain variables or factors matter more than others, virtually none does a complete analysis to discover its true strategic fulcrums. None tracks the movement of linkages to assess how strategic fulcrums evolve over time. For example, when oil prices are high, decisions related to hedging future oil purchases were critical for airlines because the cost of fuel emerged as a strategic fulcrum for relative competitiveness and profitability. The overall strategic impact of the hedge was more than just the cost savings on fuel. The linkages from the decision to hedge possibly related to pricing power, relative customer demand, load factors, revenues, and margins. However, if oil prices come down, the management of fuel prices may not remain a critical strategic fulcrum for any airline. Similarly, lower distribution costs from a direct-to-customer distribution model might be a strategic fulcrum if the price-to-weight ratio of an electronic or a computer product is high but might not remain so when the ratio comes down. We find that organizations get trapped in legacy strategy systems because they do not explicitly identify their strategic fulcrums and do not track the evolution of fulcrums over time. Linkage analyses facilitate the discovery of these strategic fulcrums and the evaluation of the impact of managing them appropriately.

Building a Repository of Strategic Linkages

Finally, linkage analysis helps an organization build a repository of webs of linkages. For example, once the analysis is completed, the estimated relationships among the set of key linkages become a unit of learning that can be stored explicitly as part of a repository. At later points in time, the organization can draw upon these sets of linkages to estimate or predict the likely consequences from actions taken at future dates. For example, if an organization builds a repository of linkages that capture the effects of automating people-driven manual operations, it can better predict the outcome of future automation based on the learning embedded in previous linkages.