Page Tools

Company Profile

Sawtooth Technologies Consulting Group

Northbrook, IL
P: (847) 239-7300
E: consulting@sawtooth.com
W: www.sawtooth.com/index.php/consulting/

We are experts in applying conjoint analysis to product development and pricing issues, leading to better informed, customer driven decisions.

» See all content shared by Sawtooth Technologies Consulting Group

Page Bookmark

Article:

Conjoint/Discrete Choice Model Output: What’s the Share About?

by Sawtooth Technologies Consulting Group

When business decision-makers look at conjoint and discrete choice model output, how should the share results be interpreted and used? In this post, we share our thoughts about preference share and market share, based on decades of practical experience.

If you used a conjoint-based (or discrete choice) simulation model and saw the following output:

you might assume the figures represent market shares. But, the output of a conjoint simulation model is known as a “share of preference.” In this post, we will discuss the important differences between preference share and market share, and provide our thoughts about how to interpret and use these figures to help with business decisions.

In previous posts, we have described the basics of conjoint analysis and simulation modeling. In this post, we pick up at the point of using output. [But, if you don’t have time to read the prior posts, work with this: A conjoint model compares respondent preferences for the building blocks of a decision – such as which TV to buy – with the offerings in the market. The output predicts the level of preference any product might achieve.]

Most model users might prefer to simplify and assume that preference share equals market share. That would seem to make life—and revenue projections—easy. But, conjoint and discrete choice simulation models are based on important assumptions, some of which are basic to market research (e.g., representative sample, accurate answers, stated preferences will translate into behavior) and others which are unique to conjoint: All products have equal levels of distribution; Consumers are fully aware of and understand the specifications of all products; All products have reached their equilibrium shares (that is, full adoption has occurred); All relevant attributes that contribute to share have been included.

What’s a Researcher To Do? Take a close look at share of preference AND market share.

So far, this post may seem focused on caveats. Let’s switch gears. The good news: Shares of preference are very powerful. They help us understand what a product’s potential is, how well-liked it is versus competitive products, and what aspects of the product (such as brand, a particular feature, or a competitive price) are most contributing to preference. In addition, by simulating changes to the product (such as changing the mix of features, or raising or lowering price), we can determine the relative benefit of our strategic options. We can see which changes most increase or decrease preference, and which have only slight impacts.

But, the inevitable question arises: Why don’t these shares look like our market shares? Skittish researchers will apply external effects, which are simply multipliers that can make preference shares look like market shares. In our experience, applying external effects robs the business decision-maker of valuable information. We recommend comparing preference shares to market shares. Why? Because there is information hiding there!

If our product has a preference share that is lower than its known market share, what might be going on? Let’s start by realizing this suggests that the product is selling at a higher level than “preference” would indicate it should be. Possible underlying causes:

Distribution or awareness of our products is particularly strong. Therefore, we may be vulnerable. If competitors gain distribution or awareness, it would appear buyers will prefer their products.
A market shift is occurring and other products have not yet reached their potential. Again, we’re in a vulnerable position. But, with the conjoint model, we can see what actions are needed to stave off competition.

If our product has a preference share that is higher than its known market share, what might be happening? Start by realizing that our product is not selling up to its potential. Possible underlying causes:

Distribution or awareness of our products is low. Therefore, we have an opportunity. We may have been considering costly product changes or price decreases, but we need to allocate our resources to letting the market or channels know about our product.
A market shift in our direction may be coming. Perhaps existing products have been capitalizing on consumer loyalty or inertia. In this case, giving consumers the chance to experience or sample our product may be needed.

When preference share does not equal market share, the conjoint model won’t necessarily tell you whether distribution, awareness, loyalty, or inertia are at work. But, you will have important information about whether your product is in a vulnerable position (preference is less than market share) or whether you have an opportunity (preference is greater than market share), even before any product or price changes are made.

What Else Can Be Done? Look at conjoint share results over time.

As noted above, time can be a factor in modeling… the conjoint model assumes equilibrium shares have been achieved. To see how time may figure into the equation, two approaches can be used:

Track results over time and compare predicted to actual. Let’s say the conjoint model predicts that a new product will have 20% share of preference. After the product is introduced and some time has passed (perhaps one year), compare the 20% model prediction to the actual share achieved. This will result in an approximate “discount factor” to apply to future model runs.
Model a historical action and compare to the model results. Let’s say your product was priced 10% higher one year ago, and everything else in the market was the same. The conjoint model can be run as if it were one year ago (with the higher price). Then it can be run at the current, lower price. The resulting predicted change in share of preference can then be compared to actual change in market share. As in the previous example, this will result in an approximate “discount factor” to apply to additional model runs.

Bottom Line: Use the power of shares produced by conjoint analysis.

Instead of approaching preference shares with trepidation (“will they look like market shares?”), we suggest conducting a thorough and open exploration into the differences between preference and market shares. There’s information that can lead the way to vital understandings and profitable strategies.

-June 2010

This content was provided by Sawtooth Technologies. Visit their website at www.sawtooth.com.

Newsletter Signup

Article:

Conjoint/Discrete Choice Model Output: What’s the Share About?

by Sawtooth Technologies Consulting Group

Other content shared by Sawtooth Technologies Consulting Group

Conjoint Analysis: Product Line Strategy

Conjoint Analysis: New Product Selection

Conjoint Analysis: B2B Customer Service Strategy

Conjoint Analysis: Product Portfolio Optimization

Conjoint Analysis: Staff Training

Understanding Conjoint in 15 Minutes

Finding the Sweet Spot of Competitive Positioning

Knowing when the Price is Right

MaxDiff – maximum power to act on market preferences

Predicting Choice with Conjoint Analysis and Discrete Choice

Segmenting for Success

Discrete Choice Analysis Makes Comparing Apples to Oranges Possible

Predicting Revenue from Conjoint Results

Aligning Your Line: Using Conjoint to Manage Product Lines

MaxDiff: A New Way to Prioritize Benefit Statements

Asking Discrete Choice Questions When the Answer is, “It Depends”

Using Discrete Choice and MaxDiff to Spend HR Dollars Wisely

Why I Still Love Conjoint

Resources

Community

Company

Contact