Conjoint Analysis

Something that has always fascinated me in the field of pricing is that there is a predictability to human behavior — in particular "purchase behavior." For many B2B SaaS companies seeking to make changes to their pricing and packaging strategies, customer insights research can be an extremely valuable tool for testing packaging concepts and gaining early insight into willingness-to-pay, along with valuable segmentation insights. How should we package our offers and optimally monetize new innovation? Should we bundle or use add-ons? What should the ideal feature thresholds be, and how can we use differences in price, features, and thresholds to drive the most favorable mix across our portfolio?

One of the most widely used tools for answering these questions is conjoint analysis. It helps organizations understand what customers are willing to pay, how price influences demand, and how customers trade price against product features, brand, and other benefits — insight that's especially valuable when developing a new offering, changing an existing price, or evaluating alternative product and pricing structures.

How Conjoint Analysis Works

Conjoint analysis examines how customers make trade-offs among the attributes of a product or service. Respondents review a series of hypothetical product profiles in which features, brands, service levels, and prices are varied, then choose between the alternatives or indicate which they prefer.

The resulting choices allow researchers to estimate:

• The relative importance of each product attribute, including price
• The value customers place on different levels of each attribute
• The combinations of features and prices customers are most likely to select
• The potential demand for alternative product and pricing configurations

Several forms of conjoint analysis are available, including adaptive conjoint analysis, full-profile conjoint analysis, and choice-based conjoint analysis. Choice-based conjoint, sometimes called discrete-choice modelling, is generally the most applicable approach for pricing research because it asks respondents to make decisions that resemble actual marketplace choices — which is also the format used in the interactive exercise later in this piece.

Why Conjoint Analysis Works Well

A key strength of conjoint analysis is that price is presented as one element of the overall offer. This reduces the likelihood that respondents will focus excessively on price or attempt to provide what they believe is the "correct" pricing answer.

Conjoint analysis also provides insight into both pricing and product design. It can show whether customers would pay more for a particular feature, which features could be removed from a lower-priced package, and how a proposed offer might perform against competitors.

When carefully designed, choice-based conjoint analysis can support demand forecasting, product configuration, market segmentation, and pricing strategy for both new and established products.

Where Conjoint Analysis Falls Short

Conjoint studies can be complex to design and analyze. The attributes, levels, competitive alternatives, and choice scenarios must accurately reflect the customer's purchasing decision. Poorly selected attributes or unrealistic product combinations can produce misleading results.

Including too many attributes may also dilute respondents' attention to price and cause the research to underestimate actual price sensitivity. Some approaches, particularly adaptive conjoint analysis, are known to understate the importance of price.

Conjoint analysis is generally less suitable for impulse purchases or situations in which purchasing decisions are highly complex, emotional, or heavily influenced by factors that cannot be represented effectively in a research exercise.

When to Use Conjoint Analysis

Conjoint tends to work best under a specific set of conditions. Before scoping a study, it's worth checking whether most of the following are true:

• The decision genuinely involves trade-offs. Customers are weighing multiple attributes against each other — not just reacting to a single price.
• You can reach a large enough qualified sample. A broad, addressable customer or prospect base gives you room to recruit statistically reliable segments.
• Response rates are likely to be favorable. Existing customer relationships, engaged panels, or incentivized third-party panels make it realistic to hit your completion targets.
• You need to test several hypothetical configurations at once. Conjoint is efficient for exploring many combinations of features and prices within a single study.
• Understanding relative importance matters as much as the price point itself. You want to know how much price matters compared to features, not just which price performs best.

When Conjoint Analysis Isn't the Right Fit

Conjoint isn't always the right tool. It tends to underperform — or simply isn't worth the investment — when:

• The target segment is very niche. If your addressable market is only a few dozen accounts, it will be difficult to field a large enough sample to draw statistically reliable conclusions.
• Expected response rates are low. Hard-to-reach buyers, long sales cycles, or low willingness to participate can make it costly or impractical to recruit enough qualified respondents.
• The real question is a single price point, not a trade-off. If the product and features are fixed and you only need to validate reaction to a few specific prices, a simpler method may be more efficient.
• The purchase decision is highly emotional, impulsive, or otherwise hard to simulate. Conjoint depends on respondents being able to reason through a structured set of trade-offs.
• The timeline doesn't support proper fielding and analysis. A rushed study risks under-sampling and unreliable results.

Key Considerations When Designing a Study

Once you've decided conjoint analysis is the right approach, a handful of design decisions will determine how much you can trust the results.

• Who you recruit matters as much as how many. Testing only with current customers can overstate satisfaction with your existing offer and understate how a prospect might weigh a new one. Blending your own customer base with third-party-sourced non-customers gives a fuller picture of the broader addressable market.
• Sample size should be statistically defensible before you field. The right size depends on the number of attributes and levels in your design, how many choice tasks each respondent completes, and how many segments you need to analyze independently. (See the estimator later in this piece for a rule-of-thumb starting point.)
• Attributes and levels need to reflect real purchasing decisions. Overly theoretical or unrealistic combinations produce answers that don't translate to the market.
• Respondent fatigue is real. Too many choice tasks or attributes in a single study can degrade response quality partway through.
• Screening criteria should match your actual buyer. Loose qualification criteria will dilute the sample with respondents who don't represent your real decision-makers.

Try it yourself. The simplified exercise below mimics a real choice-based conjoint survey. Work through four scenarios, each showing three priced offers built from the same feature list plus a "Choose Nothing" option, then see how the results would translate into attribute utility scores and a recommended packaging structure.

How Many Respondents Do I Need?

Sample size is one of the first questions that comes up when scoping a conjoint study. The estimator below uses a widely referenced industry rule of thumb for choice-based conjoint (sometimes called "Orme's Rule," from Sawtooth Software's research on the topic) to give you a starting figure — not a substitute for a full statistical power analysis, but a useful gut check before you build out a detailed research plan.

Building Conjoint Into Your Roadmap

Conjoint analysis works best when it starts early in the product-development process, rather than after pricing has already been set by adding a margin to cost. An offering priced this way can end up being one customers like but aren't willing to purchase at a profitable price.

Used at the right stage, conjoint analysis can guide feature priorities, product configurations, target segments, and the development of differentiated packages — reducing uncertainty and helping organizations build and price offers that customers actually value and are willing to pay for.