Conjoint Analysis – Pricing Through the Eyes of Your Customers
When customers buy, they rarely look at price in isolation. Instead, they weigh up features, benefits, and the value they perceive., and they largely do this subconsciously. Even in B2B markets, where there is more conscious analysis of decisions, there are multiple factors that customers balance when making a buying decision.
A laptop with a faster processor, more memory, and a larger screen is likely to command a higher price than one without those attributes – but exactly how much more? And which of those features are the ones driving the higher price? That’s where conjoint analysis comes in.
Conjoint analysis is a research technique that helps businesses uncover how customers make trade-offs between different product features and price. Rather than asking customers directly what they value – which often leads to unreliable or unrealistic answers (see my blog on the challenge of customer research: https://www.davidabbottspeaker.com/blog/researching-what-customers-value) – conjoint analysis looks at what they do when faced with choices. By observing the options people select, we can reverse engineer their decision-making process and see which features truly drive preference, and by how much.
It’s an immensely powerful tool for pricing, product development, branding, and positioning. Let’s explore how it works.
What is Conjoint Analysis?
Conjoint analysis is based on a simple premise: customers evaluate products as bundles of attributes. Each attribute has different levels – for example, for a laptop, “screen size” might include 11”, 13”, and 15”. Customers then consider different combinations of attributes, such as a 13” laptop with an i5 processor, 12Gb of memory, and a £1,500 price tag.
By presenting these combinations to customers and asking them to choose between each combination, we can use statistical analysis to identify the relative importance of each attribute and the value customers attach to different levels. This allows us to estimate willingness to pay, predict demand, and optimise product design.
Conjoint analysis is not new – it was first introduced in the 1960s – but it has grown in popularity because of its versatility and its ability to mimic real-world decision-making more accurately than asking customers what they say they want.
Types of Conjoint Analysis
Over time, different methods of conjoint analysis have emerged, each with its own strengths.
Traditional Conjoint (Full Profile)
Respondents are shown full product profiles, each with a complete set of attributes, and are asked to rate or rank them. This works well when there are only a few attributes, but quickly becomes unmanageable as the number of features grows.Adaptive Conjoint Analysis (ACA)
Here the survey adapts as it goes along. Respondents first indicate which attributes matter most, and the software then focuses on the most relevant trade-offs. ACA is especially useful when there are many attributes to test, though it can be more complex to design.Choice-Based Conjoint (CBC)
The most widely used today, CBC presents respondents with sets of product alternatives and asks them to choose one, just as they would in a shop or online. This method closely resembles real buying behaviour and tends to produce more reliable insights.MaxDiff (Best-Worst Scaling)
A related technique where respondents pick the most and least preferred option from a set. This is particularly effective for ranking things like product claims, messages, or branding elements.
Although the mechanics vary, the principle is always the same: break products into their components, combine them in different ways, and analyse the choices customers make between the combinations to see what they value.
What is it Used For?
Conjoint analysis is one of the most versatile research tools in marketing and pricing.
Typical applications include:
Pricing Research – Estimating willingness to pay, identifying price sensitivity, and understanding how price interacts with other attributes.
Product Development – Determining the optimal mix of features before committing investment. For example, does adding 16Gb memory deliver enough incremental value to justify the cost?
Branding – Assessing the premium customers place on brand names compared with technical specifications.
Product Claims – Identifying which claims or messages resonate most strongly, and which can be dropped without reducing appeal.
Market Segmentation – Revealing distinct customer groups that value different things. Some may prioritise price, others performance, and others brand – enabling tailored propositions.
In essence, conjoint analysis allows you to move from guesswork to evidence, basing key decisions on what customers actually value.
How Does it Work?
The process begins with defining the attributes and their levels. For a laptop, attributes might include screen size, processor type, memory, and price.
These attributes are then combined into hypothetical product profiles or bundles. Rather than showing every possible combination (which would be overwhelming), researchers use experimental design techniques to create a subset of profiles that still captures the necessary variation.
Respondents are then asked to choose between profiles. Each choice provides a data point, and when analysed across a large enough sample, the data reveals how much value customers attach to each attribute level.
The output includes:
Part-worth utilities – numerical values showing the contribution of each attribute level.
Relative importance – a ranking of which attributes matter most.
Simulations – predictions of customer choice for any given product configuration, including demand and market share estimates.
It’s rather like solving a puzzle – by piecing together how people choose, we can reconstruct the hidden rules behind their decision-making.
How to Use Excel to Do a Conjoint Analysis
While specialist software such as Sawtooth, Qualtrics, or Conjointly is designed for conjoint analysis, you can run a simplified version in Excel.
List Attributes and Levels
Define the features and levels you want to test.Generate Product Profiles
Use an approach called fractional factorial design (available through online generators) to create a manageable set of product combinations; or pick the combinations that you think are the most likely to be popular.Collect Responses
Ask respondents to rank, rate, or choose between the profiles i.e. ‘which of these would you buy?’.Code the Data
Use dummy variables in Excel to represent each attribute level. This is more simple than it sounds – it just means scoring ‘1’ if a feature is present and ‘0’ if it isn’t, and then by adding the numbers together in the next step Excel can get an estimate for the ‘part-worth’ the attribute contributes to the customer’s choice.Run Regression
Use Excel’s Data Analysis add-in to run regression, estimating the part-worth utilities for each level.Interpret the Results
Examine the coefficients to see which attributes are most important and what trade-offs customers are making.
Although Excel has limitations, this approach can be a good way to learn the basics before moving to more advanced tools.
How to Interpret the Output from a Conjoint Analysis
The numbers from a conjoint analysis may seem abstract at first, but they translate directly into actionable insights.
Part-worth utilities tell you the relative value of each feature level. For example, moving from 8Gb to 16Gb memory might add +2 utility points.
Attribute importance comes from comparing the ranges of utilities. If price has a wider range than processor, then price has more influence on choice.
Market simulations allow you to test “what if” scenarios. You can model how customers would respond to a new product, a price change, or a competitor’s launch.
The real power comes from combining these insights. You can identify the sweet spot – the product configuration that maximises customer value while still delivering profit.
Step-by-Step Example – Choosing a Laptop
Let’s bring this to life with the laptop example.
Attributes and Levels:
There are four attributes we are interested in – screen size, processor power and amount of memory, plus price. Each have multiple ‘levels’.
Screen Size: 11”, 13”, 15”
Processor: Intel i3, i5, i7
Memory: 8Gb, 12Gb, 16Gb
Price: £900, £1,200, £1,500, £1,800, £2,100
Step 1 – Create Profiles
We generate a selection of laptop profiles mixing different combinations.
Step 2 – Gather Responses
Customers are asked to choose between sets of laptops.
Step 3 – Analyse Data
Suppose the regression gives us these part-worth utilities:
11” screen: -2 | 13” screen: +1 | 15” screen: +3
i3: -3 | i5: +1 | i7: +2
8Gb: -2 | 12Gb: +1 | 16Gb: +2
£900: +4 | £1,200: +2 | £1,500: 0 | £1,800: -2 | £2,100: -4
Step 4 – Interpret
This tells us that customers prefer larger screens and more memory, while processor improvements matter less. Price sensitivity is clear – demand falls sharply beyond £1,500.
Step 5 – Simulate
We can now simulate choices. A 15” i5 laptop with 16Gb memory at £1,500 would score highly, while an 11” i3 with 8Gb at £1,800 would be highly unattractive.
This kind of analysis allows you to decide which configurations to offer and at what price points.
Pros and Cons of Conjoint Analysis
Like all research methods, conjoint analysis has advantages and limitations.
Pros:
Captures real trade-offs rather than abstract preferences.
Quantifies willingness to pay and feature importance.
Supports product design, pricing, branding, and segmentation decisions.
Enables simulations to test different market scenarios.
It avoids the need for complex and multiple sets of research where each individual option has to be tested in isolation.
Works well for both B2C and B2B.
Cons:
Designing a good study requires expertise – poorly chosen attributes yield poor insights.
Surveys can be long and demanding for respondents.
Advanced analysis often needs specialist software and statistical skills.
As with any research, results are only as reliable as the sample used.
Quite high numbers of respondents might be needed to get statistically significant information.
Final Thoughts
Conjoint analysis is one of the most insightful tools in the pricing and marketing toolkit. It helps us answer a deceptively simple question: what do customers truly value, and how much are they willing to pay for it?
By uncovering the trade-offs that customers make, conjoint analysis allows businesses to design better products, set smarter prices, and position themselves more effectively. Whether you’re a global technology company or a start-up launching your first product, it can provide the evidence you need to make confident pricing and product decisions.
And at its heart, conjoint analysis reminds us of something fundamental about pricing: it’s never just about the number. It’s about the entire bundle of value, as seen through the customer’s eyes.