[MUSIC] The tool we will use to link financial value to operational metrics is called a ROIC Tree. Because it will visually deconstruct the financial metric that we chose, ROIC, into operational metrics. Let us show you how to build such a tree. We will first provide a general recipe and directly illustrate it with an example for a restaurant chain. We will accompany this session with a set of exercises where you can build a tree for various organizations and ultimately your own firm, department, customer, vendor, or even a competitor. Here's the general recipe for building a ROIC Tree. You start with the objective ROIC on one side and you begin deconstructing a variable into components. How do you decide which branches to deconstruct? You are aiming to identify branches that have impact on the bottom line. And you do so by asking several questions. What are the main cost drivers? Because there is no point of delving into those that are only a small fraction of the cost. What are the strategic levers in the sense that they have long term impact? And which inputs are most likely to change? You continuously expand important branches until you end with well defined, measurable, and actionable metrics that can be tied to operation strategy. Then come the two most important steps. One, you must populate the tree with actual numbers. Doing so makes this a data driven tool, and not just a theoretical exercise. Two, you investigate how the financial value depends on the various operational metrics. This involves performing sensitivity analysis on several metrics, which will allow us to prioritize the most important performance metrics, and thus identify our KPIs. >> To illustrate the process of building a ROIC Map, consider the Mafia restaurant chain in Kiev, Ukraine. Mafia is a modern casual-dining restaurant chain. It offers Italian and Japanese food with high quality service at an affordable price to target middle class customers. Serving both Japanese and Italian cuisines has become a trend for local Ukrainian recently. Mafia is one of the fastest growing restaurant chains in the local market in Kiev and continues to expand rapidly. The restaurant branches grew from 4 to 24 over the last three years. Was serving approximately 3.3 million customers a year. The company is over 1,300 employees. The revenue in 2012 was $38 million hryvnia, the Ukrainian currency, which represents an increase of 56% compared to the previous year. Net income was approximately $7 million, an increase of 44%. Although Mafia's ROIC is high in comparison to the industry's standard the company is facing challenges in further expansion and general efficiency. Restaurant labor is typically divided into two parts. Front of House, FOH labor includes wait staff, floor managers, hosts, bussers, and bartenders. Back of house, B of H labor includes kitchen staff and dishwashers. Mafia's FOH costs are on the order of 3.4 million hyrvnia. While Back of House costs are 16.5 million hyrvnias. Initial conversations with the management indicated that labor costs are higher than in some fast food chains which creates additional risks for future competition. In addition, the absence of a central kitchen to service all branches forces individual branches to build structurally big local kitchens at the expense of more customer seating. In order to understand the impact of labor and better understand the drivers of both revenues and cost, it was important to further develop the trade. >> The main elements that require additional discussion are revenues. These can be determined using two different, but equivalent perspectives. The product view and the customer view. The product view starts with the sales volume, that is the number of orders sent to the kitchen while the customer view starts with the number of customers and tables. We will use the latter, and thus the revenues are determined as the product of the number of customers per year and the average spending per customer. The number of customers per year can be computed as the product of the number of seats and the average turn per seat. The spending per customer is the product of the average price per product, and the number of dishes per customer. Food costs can be determined as the product of the number of dishes and the cost per dish. Labor cost is the cost of managing the 24 kitchens. The cost of each kitchen can be computed as the product of the number of employees per kitchen and the wage per employee. Using the additional operational data we can construct a tree. With the ROIC Tree implemented in a spreadsheet, we have a tool to assess the value created by several operational changes, and to identify key performance metrics by conducting sensitivity analysis. If the restaurant would increase up-selling it could increase the number of dishes per customer from 2.2 to say 2.3, a 4.5% change. The value of increasing upselling or else being equal is reflected by an increase in the after tax ROIC from 24% to 28% per annum, a 17% change. If Mafia simplifies the menu, and thus reduces the time each party spends on ordering at the restaurant, it can increase its seat turns. Increasing seat turns from 2.03 to 2.1, which is a 3.45% change, increases ROIC from 24% to 27%. Which is a 12.9% change. To compare these sensitivities, it is useful to normalize them by considering the value change for a 1% change in the operational metric. For example, a 1% change in number of dishes per customer increases ROIC by 3.75. Similarly, a 1% change in turns increase ROIC by 3.75. Thus, a 1% change in turns has identical value impact as a 1% change in number of dishes per customer. This is because both result in a 1% revenue increase and we had assumed that costs are unaffected. if Mafia could reduce the cost per dish from 1.8 to 1.7, a 5.56% percent change, ROIC would increase from 24% to 26.5%. A 10.2% change. Normalizing show that a 1% reduction in cost increases ROIC by 1.85. Thus if each KPI could be implemented without affecting others, a 1% revenue improvement has similar value impact as a 2% cost reduction. Of course, a comprehensive analysis should include the cost of implementing each change. To summarize what we've done so far, the process of building the ROIC Tree ends with identification of the operational performance indicators as the end node that drives value. Given that these indicators are operational, they can be monitored and controlled in real time, and are leading indicators of financial performance. Sensitivity analysis then ranks these operational performance indicators and establishes a prioritized list of KPIs. Armed with a ROIC Map and a corresponding list of operational KPIs, managers or turnaround specialists can now estimate which improvements in each KPI are feasible per time bucket, week, month, quarter or year, and which growth plan makes more sense. Of course, such assessment requires substantial experience, fact based data gathering, and input from stakeholders. The improvement process can then be clearly articulated in terms of goals and milestones, and tracked and adjusted over time. And finally, we build a narrative to communicate your operational plan and its value to the external investor audience. The last step is to delve deeper into scaling. In a certain sense, we now want to perform a meta sensitivity analysis of how scaling the operating system impacts value. What is new here is that an increase in demand will impact multiple KPIs simultaneously. Once we understand which, and how KPIs are affected by growth, we can build those relationships into the ROI Tree. We then have a powerful tool to test the scalability of the operating system, but investigating how does financial value depend on scale? To help you in that process, we will introduce scaling laws in the next section, that predict how increased demand impacts various operational metrics. [MUSIC]
