When we think of a car engine, we tend to think of it as a single unit without stopping to consider the pistons, valves, cylinders, and spark plugs that enable it to run. The same can be said for modeling software. As a stand-alone application or as a component of an optimization platform, modeling software should be evaluated in terms of its individual makeup. This is because it has a considerable impact on the ultimate goal of optimizing a company’s strategic planning—similar to how a car engine powers a vehicle to its destination.
So, what are the most important individual considerations when it comes to business modeling software? Here are five that stand out.
-
Can problems be defined with platform technology or a packaged application?
-
Can the software add new constraints and objectives in a fast, flexible manner that goes beyond the initial model?
-
How well does the modeling component support financial inputs and outputs (i.e., are they manually added or available out-of-the-box)?
-
What is the learning curve for corporate stakeholders?
-
Is validation or error checking included, or is it offered as a separate solution that must be purchased?
Question No. 1: Is the modeling software part of a comprehensive platform or a packaged application?
The primary consideration for modeling is how it defines problems. For example, a packaged application that is used for logistics or strategic planning is composed of a hard-coded model. But what if a business has constraints or objectives outside this type of model? This can make it difficult to achieve a feasible/optimal outcome, unless the model can be customized.
Whether it’s a platform that requires the user to write math equations or the River Logic platform that provides a drag-and-drop visual interface, a modeling platform accommodates a business as it is, whereas a packaged solution does not.
Question No. 2: Can the business process modeling software add new constraints and objectives?
As a follow-up to the first question, understanding whether modeling software can accommodate new constraints and objectives beyond the initial model is significant. Important questions may include:
-
What will the demand for Product X be in the following month?
-
How will an input shortage affect profitability?
-
What would it cost to implement a desired feature?
Changes to the model are inevitable, and the technology must have the agility to support future business changes in a fast, flexible manner. Otherwise, you’re forcing your business into a box and going back to the original problem: making poor, uninformed decisions.
Question No. 3: Does the modeling technology support financial inputs and outputs?
One of the essential factors for validating whether a model’s outcome is feasible and optimal is the model’s ability to support financial inputs and outputs. Financial constraints and KPIs are real-world issues that every company has to deal with. Thus, this feature alone is the gold standard that distinguishes modeling software providers as leaders in this category.
Input considerations include costs and revenues such as:
-
Fixed costs
-
Variable cost/unit
-
Variable cost/hour
-
Variable cost/customer
-
Base revenue
-
Revenue associated with discount
Outputs that represent the input data and show working capital, debt-to-equity ratios and net income, as an example, include profit and loss statements and cash flow statements.
Modeling software must show this necessary information in order to answer key questions such as:
-
How will a new product affect existing SLAs and revenue targets?
-
How will a new factory affect profit?
-
How should the enterprise reach its financial goals? What operational decisions need to be made to reach these goals?
Question No. 4: What is the software learning curve for corporate stakeholders?
If the type of model is determined and its ability to incorporate changes is intact, what kind of learning curve is involved for users? For platforms that require users to do the math, a skilled operations research analyst may be needed. But with drag-and-drop visual interfaces, business users with minimal training can perform code-free modeling. This is also true when it comes to packaged applications—they often don’t require a data scientist (unless customizations are necessary, of course).
If the goal is to help the business improve its decision-making, it’s best to steer clear of solutions that require a data scientist. Instead, look for someone who understands how the business flows. Also, consider what happens if a key employee leaves the company; if modeling technology is completely dependent on one or two people for gaining insights and making modifications, the company puts itself at great risk as a single departure could have a profound effect on efficiency.
Question No. 5: Is validation or error checking included in the modeling software, or is that a separate requirement?
The use of modeling software is a futile exercise without the ability to validate the baseline model and check for potential errors. A feasible and optimal outcome is impossible to confirm without the option to drill down into components of the model that don’t line up.
When selecting modeling software, this final feature facilitates and validates the desired enterprise-wide view—a planning model that encompasses all critical operating and financial considerations. A business model example of a supply chain may include:
|
|
When it comes to modeling software, the idiom “hitting on all cylinders” can be achieved when a company considers these five features. Otherwise, the quest to find optimal, feasible answers to old and new questions is never ending!
