In a time where advanced analytics techniques and applications are buzzing, it is important to prioritize the purpose and the value to drive these efforts. Simulation in analytics is the concept of “approximat[ing] complex systems in which its properties and behavior mimic the actual system of interest,” according to Gartner. This definition emphasizes the tactical function of simulation but the purpose and value behind it provides even greater possibilities: the ability to shift decisions from being reactive to proactive. Understanding the inputs to and repercussion of decisions before making them gives organizations the leverage to build well-informed, modeled long-term strategies, supported by analytics. This shift empowers analysts and business stakeholders alike to get creative in elements of their business that can change, knowing their decision-making is augmented through simulation. This piece will take you through four points, discussing the value simulation in analytics can bring to organizations.

Simulation in Analytics

1. Understand System “Levers” That Can be Pulled: 
   Simulation in analytics models are meant to be robust due to their reflection of the real-world as accurately as possible. Take a client example: we assisted a food manufacturer  in simulating their P&L, encompassing items across the whole supply chain, based on a model predicting cases of product sold. By nature of simulation technique, we ran through a variety of changes to budget items affecting the P&L, ultimately driving towards the best net profit the changes provided. In the case of increasing the trade budget by $3,000, we observed an increase in net profit of approximately $12,000 – an example of a relatively small monetary change with a huge effect, that may not have been recognized without simulation and the ability to run through a multitude of situations and see results.
   Consider another example of digital twins, a digitized version of a given system, which also reflects the live changes and updates that take place in the real system. Given the real-time updates, the user can understand decision making points on the most-up-to-date reflection of the system, without having to make a physical change. Ultimately, simulations will allow organizations to be proactive in understanding the ways in which strategic decisions might affect their entire systems and operations.
2.  Leverage Human Factors and Constraints: 
   Simulation is like other analytics methodologies, in that human expertise is necessary to develop a sound model of the system at hand. Only after incorporating business Subject Matter Experts (SMEs) does simulation truly live up to its potential to create useful insights and meaningful change. The SMEs can contribute integral knowledge about factors that may be neglected from the data, which could be misinterpreted by models, thereby jeopardizing its legitimacy. For example, qualitative and often-times subjective factors such as reputation, influence, and various network effects may have substantial impacts on real-life outcomes. SMEs can spot such missing variables, include them into the simulation, and improve the accuracy of the projections.
3. Broaden Use Through Visualization: 
   Companies can also leverage simulation in analytics to build their visualization capabilities. Whether it is creating a simple report that summarizes the results of a simulation in an organized manner or using a more advanced simulation software with live enactments of the modeled system, we see branching into this space as a representation of an organization’s willingness and desire to innovate. Simulation provides a cost- and time-efficient tool that can expand an organization’s outlook and accelerate its digital journey. The visualization component can help with adoption across levels, with Gartner highlighting the importance of “usability and positioning [tools] to be consumed by less technical users,” and “easy-to-engage visualizations” for buy-in at management level.
   Take this client example, where we built a Power BI dashboard that allowed sales representatives of a manufacturer to tinker with volume and price across all brands at a given retailer. This dashboard enabled the reps to show multiple scenarios to retailers to prove out the best option, empowering them with data-driven negotiations to maintain share and profitability.
4. Think Strategically, Shifting Away from Metrics and Towards Best Results, While Aiding in Change Management:
   Buy-in for an emerging field like analytics can be difficult across organizations, as we see analytics maturity vary widely across, and even within, our clients. Building or using a simulation in analytics tool is a great vehicle to catalyze adoption. Optimization of a system or problem can only be achieved when SME knowledge is incorporated, in addition to the modeling taking place. This symbiotic relationship shows analysts that the intention is to enhance their ability to uncover insights, not to remove their say.

Additionally, we see our clients increasing in advancement and breadth of the data they’re collecting through all aspects of business, like sales and marketing, supply chain, and consumer engagement. These facets aid in making a simulation be the best representation of the real-life system and all of its constraints. In making tweaks to the components of the system then running the simulation, SMEs can choose how they want to evaluate success (e.g. maximizing gross profit in a P&L, minimizing waste on a manufacturing line) then consider the scenario parameters with the best results, enabling more creativity and flexibility than driving towards a set metric.

You can find more information on Clarkston Consulting’s analytics capabilities here.

Contributions by Andrei Volkov, Maggie Seeds, and Elise Watson