Example #5: Searches Using Attribute-Based Prices

As mentioned earlier, reliable cost information often is not available for conducting profitability searches.  Even though cost information may not be available, firms often have established pricing information (the price they charge buyers, broken down by features, or attribute levels), and these can be incorporated into the product search to help avoid trivial solutions that would be unprofitable. For example, whenever a certain feature is added to a product alternative, its overall price can be adjusted by a certain price delta.

Let’s again refer to the TV data set that we’ve been using. Further, let’s assume we do not have any cost information regarding the various features. Rather, we know a base price, and the incremental prices that certain features (such as picture-in-picture, or a larger screen size) add to the overall price (that buyers pay) for the unit.

The ASM lets you add price tables wherein you can specify prices for one attribute at a time, or multiple attributes at a time, if the prices interact with multiple attributes (e.g. if the price for PIP depends on what screen size is offered). In our example above, there are no interdependencies for prices.

To specify price tables, from the Home tab, click the Revenues and Costs option on the ribbon within the Project Information group. Then, click Price Tables and the Add button to Add prices per attribute level.  Add prices for Screen Size, Sound Quality, Channel Blockout, and Picture-in-Picture as shown in the table below. Do not add a Base Price ($310) to the brand attribute--we will specify the base price in the product specification grid.

Note that given these prices, the lowest possible price that a product can take on is $310 ($310 + $0 + $0 + $0 + $0), and the highest price possible is $440 ($310 + $30 + $45 + $20 + $35).  These prices all fall within the full ranges of price that we measured, $300-$450. If some possible total prices fall outside the range, the ASM will prohibit such combinations during the search process.

In the previous two examples, RCA was interested in creating mid-sized televisions to compete with JVC and Sony, where each competitor already offered two products. We introduced the fixed competition in Example #2. In Example #3, we searched for an optimal configuration for the best single and pair of RCA televisions to offer, relative to the competition, to maximize Revenue. In Example #4, we extended that scenario to include costs and to search based on profit. In this example, we’ll use the same competitive mix, but use attribute-based pricing information instead of costs, to search for the best two products for RCA to maximize share.

Specifying the Products

Simulation Method: Randomized First Choice (click the icon to specify that 50 iterations should be used per respondent, to make things run faster for this example)

Range Behavior:  Grid, optimizing Revenue (click the icon to specify Profit as the single objective)

After several minutes, the optimal 2 products found in terms of maximizing revenue for RCA (net revenue across both products) are given as follows:

Revenue is maximized at $94.8MM by offering a full-featured TV at $440 (undercutting the price for Sony’s identically full-featured TV selling for $450) and another slightly less fully featured TV at $425.