# Conjoint Analysis

Kesu01

Find the best combination.

What's the best way to introduce a new product, or change an existing one?

You could just move forward boldly with a new idea and keep your fingers crossed that it works. Or you could reduce risk by doing market research – before you go through all the trouble of creating something customers don't want or don't like.

Getting a product 'right' involves a lot of variables. The most obvious feature is functionality – how it works. However, other things also play a role in the final purchase decision – such as packaging, promotion, materials, and even where a product is manufactured.

For example, people buy cars to get from point A to point B. The type of car they buy is based on many things, including fuel consumption, styling, reliability, and color. While any of these product attributes may be the primary selling feature, people make decisions by considering all the attributes together.

Getting all of these features in the right combination is pretty difficult if you just rely on guesswork. So, how can you evaluate your goods and services by considering their attributes all together, or jointly? "Conjoint Analysis" accomplishes exactly that.

## What Is Conjoint Analysis?

First and foremost, conjoint analysis is a tool that measures buyer preferences. Using statistical analysis, it establishes the impact on the buying decision of one combination of product attributes compared with other combinations. By doing this, you get an understanding of consumer preferences that's much deeper than simply asking consumers to rate individual product attributes.

For instance, a typical preferences survey tells a restaurant that customers rank their priorities as service, price, location, and then cleanliness. So the restaurant makes improvements to service and price, but sales don't increase significantly. They wonder what went wrong… until they try a conjoint analysis, which tells them that the combination of service and location actually ranks higher than the combination of service and price.

Conjoint analysis helps you truly understand consumer trade-offs. What are customers willing to trade if they can't get the perfect set of attributes? To get the warranty they want, will they pay a higher purchase price? To get the 10% discount they want, will they buy a package of eight, rather than a package of six? To get the performance they want, will they settle for fewer color options?

## Conducting a Conjoint Analysis

Conjoint analysis determines the utility (usefulness or desirability) values that consumers attach to different levels of a product's attributes. By showing potential consumers different product offer combinations, and asking them to rank the various offers, you can identify the most appealing combination of attributes. From there, you can make a business decision using parameters like estimated market share and profit potential.

To illustrate this approach, we'll use an example of a consumer goods company developing a new glass cleaner.

### Step One: Identify product attributes to be studied.

For the glass cleaner, five product attributes were chosen:

• Format.
• Price.
• Scent.
• Ingredients.
• Promotion.

### Step Two: Choose options and/or value levels for the attributes.

Our glass cleaner manufacturer thinks that the following options are likely to be the most popular alternatives.

 Formats Spray liquid, Foam, Wipes Prices \$2.49, \$2.69, \$2.99 Scents Unscented, Floral, Lemon Ingredients Organic, Chemical Promotions Buy 2 get 1 free, \$1.00 off coupon

### Step Three: Determine which product attribute combinations to evaluate.

It's usually not reasonable to ask respondents to rank every possible combination of attributes. In our example, there are 108 (3 x 3 x 3 x 2 x 2) different combinations to consider!

The glass cleaner manufacturer chooses 12 combinations that it wants people to rank.

Design Format Price Scent Ingredients Promotion
A Spray \$2.49 Unscented Organic Buy 2, get 1 free
B Foam \$2.49 Lemon Chemical Buy 2, get 1 free
C Wipes \$2.49 Unscented Organic Buy 2, get 1 free
D Spray \$2.69 Floral Chemical \$1.00 off
E Foam \$2.69 Lemon Organic Buy 2, get 1 free
F Wipes \$2.69 Unscented Organic \$1.00 off
G Spray \$2.99 Lemon Organic \$1.00 off
H Foam \$2.99 Floral Chemical Buy 2, get 1 free
I Wipes \$2.99 Lemon Organic \$1.00 off
J Spray \$2.49 Lemon Chemical Buy 2, get 1 free
K Foam \$2.69 Unscented Chemical \$1.00 off
L Wipes \$2.69 Lemon Chemical Buy 2, get 1 free

### Step Four: Determine how to present the product attribute combinations.

You may use a simple grid or chart to present to respondents. Other options include describing the combination in paragraph form, using pictures, and even creating prototypes that people can examine.

Here is how one respondent ranked the 12 combinations.

Design Format Price Scent Ingredients Promotion Rank
A Spray \$2.49 Unscented Organic Buy 2, get 1 free 6
B Foam \$2.49 Lemon Chemical Buy 2, get 1 free 11
C Wipes \$2.49 Unscented Organic Buy 2, get 1 free 3
D Spray \$2.69 Floral Chemical \$1.00 off 9
E Foam \$2.69 Lemon Organic Buy 2, get 1 free 7
F Wipes \$2.69 Unscented Organic \$1.00 off 1
G Spray \$2.99 Lemon Organic \$1.00 off 5
H Foam \$2.99 Floral Chemical Buy 2, get 1 free 12
I Wipes \$2.99 Lemon Organic \$1.00 off 2
J Spray \$2.49 Lemon Chemical Buy 2, get 1 free 10
K Foam \$2.69 Unscented Chemical \$1.00 off 8
L Wipes \$2.69 Lemon Chemical Buy 2, get 1 free 4

### Step Five: Analyze and interpret the data.

Using statistical programs, you can analyze data, and determine the utility of each attribute. Utility is a measure between 1 and 0 (the higher the utility, the stronger the consumer preference).

If you look at the sample data (without using statistical analysis), the top four choices are all wipes, and the top three choices are organic. This consumer is even willing to pay top price for organic wipes.

Design Format Price Scent Ingredients Promotion Rank
F Wipes \$2.69 Unscented Organic \$1.00 off 1
I Wipes \$2.99 Lemon Organic \$1.00 off 2
C Wipes \$2.49 Unscented Organic Buy 2, get 1 free 3
L Wipes \$2.69 Lemon Chemical Buy 2, get 1 free 4

The lowest-ranked combinations (not including the chemical wipes) used chemical agents, regardless of the other factors.

A full conjoint analysis reveals all of the utilities, and helps you determine the attributes that are most important to your customers.

When you collect preference data from a large sample of target consumers, you can estimate the market share that any specific offer is likely to achieve (given any assumptions about competitive response). Your company, however, may not choose to offer the most attractive combination due to cost considerations. If the most attractive option is also the most expensive to produce, you may use the results of conjoint analysis to determine the most profitable option as well.

#### Key Points

Knowing the relative value that consumers place on various product attributes is essential to launching a new product or service successfully. Conjoint analysis is a powerful tool to get this information quickly, and it has become one of the most popular concept development and testing tools used. Conjoint analysis has three basic parts: a designed experiment, the statistical analysis of the resulting data, and the business decisions based on the analysis. Organizations can use the information gained from this to make better, and less risky, decisions than they would have made by using individual factors alone to determine the best combination of product attributes.