Overview

Conjoint analysis is an approach for measuring customers’ preferences; it is particularly useful for analyzing and predicting customers’ responses to new products and new features of existing products. With conjoint analysis, companies can decompose customers’ preferences for products and services (provided as descriptions, visual images, or product samples) into the “partworth” utilities associated with each option of each attribute or feature of the product category. By recombining these partworths, companies can predict customers’ preferences for any combination of attribute options, determine the optimal product concept, and identify market segments that value a particular product concept highly.

Conjoint analysis also helps firms answer such questions as: 

• How much are our customers willing to pay for an extended warranty?
• What factors drive customers' choices?
• If we must choose between two different features to introduce in the next generation of products, which one would have the most impact on customers’ choices?
• In our market, how many customers are price sensitive? How many are quality-driven in their purchase decisions?

Getting Started

Creating a conjoint study template

Step 1 Creating a study design template

A conjoint study is a multi-step process. In Step 1, you will enter your attributes and levels.  In Step 2, the Enginius software will use those attributes and levels to create profiles of product combinations to be rated by your customers. Enginius uses an Addelman design to create the minimum number of profiles required to extract the partworths correctly (for more details, see Conjoint Technical Note at https://www.enginius.biz/index.php/instructors/teaching-resources/technical-notes/). Note that if you change any of the product profiles or remove any product profiles, the software may no longer provide accurate results (or any results at all). 

An attribute is a general property or characteristic of a product category that you can use to build and describe alternative products or services. ”Color,” “price,” or “quality” are examples of attributes. 

After you have described the attributes, you must enter levels for each in the next step. Whereas an attribute represents a characteristic such as color, price, or warranty, the levels are the particular values that an attribute can take, such as red, $20, or 1-year warranty.  Each attribute requires at least two levels.  The screen capture below shows the dialog box that results from using Enginius Templates (Conjoint Analysis).  Note: Generating a Conjoint template is a 2-step process.  Both steps need to be completed before any analysis can be done.

The screen capture below shows the Attributes and Levels for each attribute from Step 1 of the template process.

The above matrix should be modified with the desired attributes and levels.  Once done, Step 2 of the template process can be completed, as shown below, with all of the options selected and explained in the following section of this tutorial.  Your particular study may not need all of the options presented below.

In the above dialog box, please review the selections available via the drop-down arrow in the Preference Partworths field:

Preference partworths can either be estimated from ratings (provided in “short” or “long” format), or “Provided as is” (a data block of existing partworths).  

Short format ratings are used when all respondents answered identical surveys (profiles). Because of the similitude in survey format and content, the characteristics of each rated profile only needs to be specified once for the entire analysis (as opposed to once for each respondent) hence the data is less verbose and easier to input. However, the constraint of similar profiles for all respondents may limit the user’s options. As above, the ratings are then used to infer respondents’ partworths (preferences). The short format ratings will consist of two data blocks. One block will contain the profiles and one block will contain the ratings.

In this example, each respondent was ask to rate the same 13 combinations of attribute and levels as described above.

Conjoint ratings (short format): Each respondent rates several identical profiles.

Long format ratings are used when each respondent rated different conjoint profiles (shown below). These ratings are then used by Enginius to infer respondents’ partworths (preferences). With this option, every respondent can be exposed to different profiles and even different number of profiles (i.e., each respondent has his or her own customized or randomized questionnaire). The long format ratings requires one data block with each row containing one profile that the respondent has rated. Note that the Enginius template generation process will generate identical bundles for each respondent. If you would like to use different bundles for different respondents, you would need to develop these bundles yourself. 

In the example below, each respondent was asked to rate a different set of profles.

Conjoint ratings (long format): Each respondent rates several different attribute combinations

Provided as is is used when the respondents’ partworths are already provided by the user. Typically, these partworths have been computed from a conjoint survey that is not accessible anymore to the user.

The Simulations section of the dialog box allows you to further customize the conjoint template that is generated.  

In this section you can limit the number of product profiles (placeholders) that exist in the current market, choose to include or exclude incremental revenuefigures, include new product profiles and restrict usable levels in the resulting template.

Incremental Revenue allows you to specify incremental revenue for each attribute/level specified in the Conjoint design restriction table.  Incremental Revenue is an optional table which will enable you to run simulations based on contribution.

Provide new product profiles allows you to specify product options that could be introduced to the market. Each attribute must contain levels from your existing conjoint design.

Restrict usable levels allows you to restrict attribute levels available within the product when running a simulation with optimal products. This feature is useful when one wants to further test which attributes are more attractive to respondents. For example, at a given price, is warranty more valued than features?

Step 2 Creating a data collection template

Data collection for a conjoint study usually requires a data collection instrument that allows respondents to either rate an attribute/level combination or rankeach combination where the most preferred has the highest rank.  In the above template generation, Enginius has generated the needed product profiles to the study design, and has populated example (generated) product ratings.  The two tables are shown below.

The above tables should be used to create the needed data collection vehicle to gather respondent ratings for each of the generated profiles.

Step 3 Entering your data

Once the respondent’s ratings have been collected, they can entered/imported into Enginius.  The table below shows an example to the product ratings table populated from respondent data for the above-described study. 

Step 4 Estimating preference partworths

Respondents’ partworths represents the results of converting respondent ratings or rankings to partworths.  In the analysis of the raw respondent ratings or rankings data, partworths are identified for the factor levels such that each specific combination of partworths equals the total utility of any given bundles or product profile. Partworths must be calculated for each respondent. Respondents’ preference partworths can be interesting to analyze in and of themselves: What are the most important attributes (or features), what are the most preferred levels (or options), and so forth?

If you have collected rating data from your respondents, Enginius will produce the partworths as shown in the table below. Note that the web report option will only show an excerpt from the full respondents table. To see all respondents, generate your report in Excel format.  After partworths have been generated, you may use partworths for your conjoint analysis without using the ratings again (unless you add additional respondents to your original data).

Existing Competitors describe the attributes of each of the existing products that were included in the study. Some options currently exist in the market, such as products or services offered by competitors or your own company. You must describe these existing products if you plan to study the market potential of new offerings, which are gauged with reference to what already exists in the market, or to analyze cannibalization effects of your new product on your company’s existing products in the market.

Running Analyses

To run Conjoint analysis on the data you have loaded/prepared, click on the RUN CONJOINT ANALSYSIS button on the left side of the Enginius dashboard.  

The above dialog box will allow you to specify the parameters for the analysis to run.

• Conjoint Design allows you to select the data that describes the design of the conjoint analysis to be run.  Note: Enginius allows you to define and name many data blocks, so you could have more than one “design table” in your analysis. 
• Preferences will allow you to select the Estimation method (when estimated from product ratings) or if partworths are already calculated, “Provided as is”. When estimating from short format ratings, you will need to specify the product profiles being rated data block and the product ratings data block. When estimating from long format ratings, you will only need to specify the product ratings data block as this data block will contain both profiles and ratings.
• Simulations allows you to select what simulations should be included in this particular analysis.
  • Existing product profiles specifies the profiles for products already in the market place.
  • Current market share allows you to specify the current market shares of existing products, if known.  You can infer a more precise relationship between preferences (preference partworths) and choices (market shares), which enhances the predictive value of your simulations. Of course, you must also know exactly what alternatives already exist. 
  • Decision rule allows you to select the rule to be used in this analysis
    • First choice rule: Each respondent selects the product that provides the highest utility among competing products and a specific new product concept being evaluated. If customers buy products in the product category infrequently and/or are highly involved in the purchase decision (e.g., house, car, expensive computer), the maximum utility rule is the preferred option.
    • Share of preference rule: Each respondent’s share of purchases of a particular product is a function of his or her preference for that product, compared with the total preference for all products in the competitive set. This analysis option is most suitable for products that customers buy frequently and/or for which they are less involved in the purchase decision (e.g., beer, toothpaste, restaurant).
    • Logit choice rule: The share of each product for each respondent is a function of the weighted utility for that product, compared with the total weighed utility for all products in the competitive set. The weighting uses an exponential function. This analysis option provides an alternative to the share of utility model. The Logit choice rule is only available when current market shares is selected.
    • Alpha rule: A weighted combination of the maximum utility rule and the share of utility rule, this method chooses a weight (alpha) that ensures the market shares computed in the simulation are as close as possible to the actual market shares of the existing products in the market. This option is available only if you provide information about the market shares of existing products in the segment to which you are targeting the new product. The Alpha rule is only available when current market shares is selected.
  • Incremental revenue (advanced option) allows you to select incremental revenue values associated with each attribute and levels.  The values specified is the increase or decrease revenue amount for each combination.
  • Type of simulation allows you to select what products will be included in the analysis
    • Existing product profiles which simulate performance of the existing set of competing products, assuming customers are familiar with all the products and the products are equally widely available for customer purchase.
    • New products from set you have defined. In this case, the simulation introduces one new product at a time into the market along with all existing products in order to compute the market shares of all products, including the new product.
    • Optimal products which test all possible combinations of new products and keeps those that lead to the highest market shares (or highest revenues, if you have checked that option), after taking into account existing product profiles in the current market. This analysis helps you identify new opportunities, or “holes,” in the market.
  • Restrict usable levels allows you to select which attribute levels will be included in the analysis (controlled in this example by the Conjoint design restriction table).

Depending on the data you have available and the type of analysis you want to perform, you may choose different options than the ones shown in the above selections.  

Make the desired selections for the above data blocks and click the Run button.  The Conjoint analysis will be run with the chosen selections and the analysis report will be generated.  The analysis described below was created with the selections shown above.  When analysis is complete, the following dialog box will appear:

Click “Show Report” and the Conjoint Analysis report will open in a separate tab within your browser.  

Interpreting the Results

The following section explains the output produced from the following Run Analysis selections, using the OFFICESTAR: CONJOINT ANALYSIS (FROM PARTWORTHS).

The Preference partworth section contains tables, a bar chart and pie chart that show the importance of each attribute and level included in the analysis. 

The table and bar chart below show the closeness of location, large assortment, office furniture and including both software and computers are important attributes. 

The pie chart below shows the importance level for each attribute.  This chart shows the following order of attribute importance:

The Simulations with existing products section show the Office Equipment store has a significant advantage over the Department Store included in the analysis.

The Simulations with optimal product section show the optimal products and how those five products would perform based on the respondent data. As selected within the Conjoint design restrictions data block, only a location within 5 to 10 miles was an available choice.

If you selected the Incremental Revenue (advanced) option, you will also see a chart depicting the Pareto frontier and a table that shows Weighted revenue for each optimal product, as shown below.

The above tables show that optimal product 1, located within 5 to 10 miles, having a large assortment of office supplies, no furniture, and offering software and computers would enjoy 32.8% of the market while Office Equipment and Department Store would drop to 53.0% and 14.2% respectively.

The report produced will also show a pie chart of the market shares with each optimal product.