Good/Better/Best – Product Bundling
Business scenario #1: You have a series of tiered products within your category. You would like to prioritize the features and benefits (value drivers) and the associated willingness-to-pay to better serve key market segments and optimize revenue generation.
You want to optimize your existing product bundles
Business scenario #2: You currently hold the dominant market share in a category for a premium product and want to capture more share of the category by introducing mid and low level products. You would like to better understand which product features and benefits best align for each tier (good/better/best). You are also concerned about cannibalization of the premium product.
You want to build off of existing success in the market and capture more overall share of the category by adding products at lower levels.
Business scenario #3: You offer a subscription-based service. Since your company’s inception, you’ve offered your subscription at one price point. You would like to introduce a tiered subscription strategy and would like to understand which features and benefits create the most unique subscription offerings. Additionally, you would like to understand which bundling strategy would lead customers to choose the “Better” or “Best” options over the entry level subscription (“Good”).
You want to “unbundle” your current offering and assign product features and benefits to specific subscription levels – with the goal of enticing customers to choose higher level subscriptions.
Do any of these business scenarios ring true for you? Well, the good news is the same core research strategy can be used for all three scenarios. Conjoint Analysis is such a powerful research methodology that we believe it deserves its own series on our blog. In the following weeks we will be sharing many of the powerful business applications of Conjoint Analysis. You won’t want to miss any article from this series so we encourage you to subscribe to our blog.
FIRST let’s look at how Conjoint Analysis works and then we’ll look at how it was used in the three business scenarios mentioned earlier. Today we’re going to focus on one particular “flavor” of Conjoint called Choice-Based Conjoint (CBC).
Choice-Based Conjoint (CBC) methodology is designed to derive the underlying motivators or drivers of consumer decisions. CBC is effective because it doesn’t ask individuals to try and articulate the relative importance of the variety of factors (price vs. features) which influence their purchase decisions. Rather, respondents express their preferences by choosing products from sets of concepts rather than by rating or ranking them.
Choice Based Conjoint (CBC) strengths include:
- It indirectly measures feature/benefit “utilities” so that respondents are not overly sensitized to individual features
- It simulates “real-world” decision-making by forcing respondents to make tradeoffs on multiple product features and price points, similar to everyday experiences
The CBC methodology allows us to control the inclusion and exclusion of specific levels of each attribute, which helps maintain an internal good/better/best logic.
The analytical output includes attribute utility values (which measure the individual contribution of each attribute to concept acceptance), and most importantly, are used to generate simulated preference share, enabling comparisons in the appeal of specific bundles of attributes to each other.
Let’s jump back to the business scenarios we mentioned earlier. First let’s look at how CBC is used to “optimize revenue of existing product bundles.”
The first step is finalizing the “attributes” and “levels” involved in the study. An attribute is simply a product feature or benefit. For example, a CBC study for an automotive manufacturer might include:
- Seating material
- Miles per gallon
Levels are the alternative statements under each attribute. For example, levels for the attribute “Warranty” might be 1 year, 5 years, 10 years, etc.
Once you finalize the attributes and levels, we design, program and field the study to obtain the consumer preference data. During the analysis, we create a baseline scenario for each bundle. Sticking with the same automotive example, the baseline scenarios for the good/better/best bundles might look like this:
Now that we have our baseline scenarios we can systematically vary the attributes and levels to simulate how consumers’ choices change under a range of good/better/best options. The best way to understand this is to picture the baseline scenarios and then, while holding every other variable constant, switch out one variable at a time.
For example, let’s imagine the baseline scenario above. An alternative scenario might list the exact same attributes and levels EXCEPT the price for the Good vehicle would drop to $12,000. By systematically varying the attributes and level included in simulation, we can begin to tease out the relative importance of each variable.
So how does this maximize revenue of existing product bundles?
By analyzing which bundles of attributes shift share between each tier (good/better/best) we can begin to formulate the maximum revenue scenario. The price sensitivity of the different bundling scenarios plays an important role. For example, let’s look at the baseline scenario we started with. Let’s assume the relative market share for the baseline scenario is:
- Good: 25%
- Better: 45%
- Best: 30%
If we calculate the revenue in this scenario for 1,000 units sold, the revenue breakdown would be:
- Good: $3,750,000
- Better: $9,000,000
- Best: $9,000,000
TOTAL REVENUE: $21,750,000 (Simulated revenue is calculated as: preference share of vehicle X 1,000 units X price point)
Now let’s adjust the price for the Better vehicle to $18,000 and assume these changes:
- Share for the Good vehicle dropped to 20%
- Share for the Better vehicle increased to 55%
- Share for the Best vehicle dropped to 25%
Now let’s recalculate the revenue under these new market share conditions.
- Good: $3,000,000
- Better: $9,900,000
- Best: $7,500,000
TOTAL REVENUE: $20,400,000
So, as you can see, the second scenario resulted in $1,350,000 less revenue than the baseline scenario. Even though the share for the Better vehicle increased, it resulted in lower overall revenue. By systematically altering the attributes and levels (including price) in a series of simulations, we can identify the right product bundles that optimize both market share and revenue.
A similar process would be used as for scenario #1. However, we would include key competitors’ brands and attributes in the choice exercise so that we can identify the set of features and benefits (and prices) our new product entries should offer to grab market share from the competition without cannibalizing our existing premium product.
Again, the overall process remains similar. The key difference here is that you would focus on identifying the most unique offerings, and not necessarily the offering that maximizes revenue. This is called the maximum differential scenario.
The scenario that maximizes the difference between the three vehicle offerings is identified by running simulations to find the attributes and levels which accentuate the uniqueness of each product.
The goal is to create a good/better/best offering that reduces cross-consideration and influences customers to purchase products/subscription packages that best fits their most desired features and benefits.
Need help optimizing your product bundles?
If you need help getting the most out of your product bundles, please contact us today.