Conjoint Simulator: Demand Varies by Price

How much should the company charge to get the most profit?

It's an important business question. Knowing how much to charge for a product/service can make the difference between profitability and bankruptcy. After field experiments (which can be ultra expensive) a discrete choice study is the best way to inform pricing strategy. Because you get an understanding of how much consumers like different product features, this is the only way to truly do value-based pricing.

Demand estimates are dependent

In a addition to the price charged, the units forecast in the simulator is a function of the other features that make up the product. We borrow the term utility from economics to describe how much people like an attribute. The forecast also depends on the other options that are available to the consumer. For example, the odds of choosing a Snickers bar depends on if a Reese's is also available.

Beyond units

Units forecast is nice, but what decision makers really care about is profit. Once you have the units forecast getting to revenue and profit is pretty straightforward.

Revenue is Units * Price

Profit is Revenue - Costs (fixed and variable)

Sometimes it makes more sense to look at Contribution Margin instead of profit. Contribution Margin is Revenue - Variable Costs.

It is important to note that the profit-optimizing price point is the same as the contribution margin-optimizing price point. Having fixed costs in the calculation just shifts the price-profit curves downward and the peak is still in the same place.

Automation

Changing the prices $1 or $.01 at a time can get boring fast. Programmatic automation tools allow us to zip through all of the price points quickly. It will store the price, units, revenue and profit numbers in a table and produce a quick chart for visual inspection.

Author - Jake Lee

Jake is the founder or Red Analytics. He has been working in Marketing Science and Advanced Analytics since 2005. His top interest is in using data/analytics to help managers make better decisions. He has primarily focused on custom market segmentation and discrete choice (conjoint analysis, choice-based conjoint) modeling. He frequently presents new ideas at the Advanced Research Technique Forum (ART) and the Sawtooth Software Conference. When not working, he likes grilling cheeseburgers in the backyard.