Lately, we’ve had several requests to do conjoint analysis. It’s not something we usually see, and in most cases, the requests were to research products where conjoint just wasn’t appropriate. In a recent case, we began the process of setting up to do the research with a partner, but as we drilled down on the product and competitive landscape, we found that it just wasn’t the right method. As a result of these requests, we have some advice on when to use and when not to use conjoint analysis. My hope is to get a few comments from others so we can expand the discussion. For those of you who are strong believers in the technique, I promise to publish your comments along with the rest of them.
Conjoint analysis identifies relevant product/service attributes and compares their value to price with a weighted scale or by seeing how respondents select from pairs of choices. It is an “attitudinal” measure of value in that respondents weight their perception of a feature’s value, relative to price. It is fast and easy to implement and is fairly inexpensive, relative to other methods. The resulting analysis gives the appearance of statistically significant precision. Also, it permits the building of complex models, which predict market share moves given changes in price and feature sets for the product.
This method works very well for complete consumer products where respondents understand and have a lot of experience with the product. By complete, I mean for things like cars, stereos, phones, PCs, etc. It works well when the results are going to be rolled out through retail sales organizations that don’t have a lot of price negotiations or pricing flexibility.
It doesn’t work well when one or more of the following conditions are present:
- It is a part of a more complex product or system—for example, a part for an automobile or machine, a process control valve in a chemical plant. The focus of the research needs to be on the performance of the part and how it impacts the broader machine or system. The real value doesn’t come relative to the price of the part but relative to the price of the machine. More than that, it impacts how it helps a firm earn more profits. Conjoint doesn’t reveal that information. We know of a high tech firm that went bankrupt and had to be sold, because they dramatically underestimated the value of their solution and saw a massive decline in price and profits. They used attitudinal measures for components of a complex system, and it failed.
- It is being sold through a field sales force. They will need to have a very specific understanding of the real mechanics of the value in use of the product or part—attitudinal results just don’t do this, no matter what the researchers say.
- There is a likely competitive reaction which negates any pricing moves. This gets down to the basics of pricing strategy—tough to model in a simulation. It should be modeled in a series of scenario planning exercises.
- The results will be used to convince a discount-oriented sales force and senior executive to stop their behavior.
I tried to keep this list simple,but you can see that most BTB products and services fall into these categories. That is why we tend to use Business Case Analysis (BCA). Often called ROI analysis, BCA is the process of doing interviews with customers to see how the product has a direct impact on their business—in terms of increased profits and reduced costs. You need to look at both of these, as sometimes the increased profits from capacity improvements can make the cost reductions look quite small.
BCA requires depth interviews, which are often positioned as exploratory and qualitative, which at the start of the process, they are. But after an amazingly few interviews, the purpose evolves to confirmatory and quantitative. That is, they confirm what we have learned and use those confirmations to build a very quantitative model of how the firms use of the product or service improves their throughput, profits, and reduces costs.
The advantage of this method is that the resulting model (What we call CASE ROI—ask for the white paper) becomes a selling tool which a) forces the salesperson to “ask the right questions,” b) subsequently shows what the specific value benefit is to each client, by plugging the answers into the sales tool, and c) is used to minimize the damage of price negotiations by providing a value frame rather than a “lower competitive price” frame the buyers often use.
My intent is to start a discussion on this sticky topic. Yes, there are firms out there that do a terrific job with conjoint, but we have found that “when you give a kid a hammer, everything becomes a nail.” That is, they always want to use it, without giving consideration for the downstream results. I’d love to hear what you think, and as I said at the start, I welcome all comments.