AHP and Conjoint Analysis

In marketing decisions, people sometimes ask what the differences are between the Analytic Hierarchy Process and Conjoint Analysis. Most of the confusion seems to come from the fact that in each process there are pairwise comparisons. But that's really where the similarity ends. (this content courtesy of Dr. Jerry Wind at Wharton).

Conjoint is the best approach (short of real market experiments) to get consumers to make tradeoffs among the features or benefits of a set of alternative choices, i.e. products. Its major advantages are that it allows in a relatively short interview (typically about 20 -30 minutes) to get enough information on consumers trade offs and, using hybrid conjoint analysis, one can analyze the data and develop three outputs:

a. the relative importance of each feature/benefit
b. the optimal offering
c. through simulators, an answer to what will happen if we change any of the features

Conjoint is especially valuable when presenting the consumer with real products (developed based on a master experimental design). Conjoint is often used to establish price elasticity., segment markets, decide on positioning and design optimal product and service offerings. It has also been used to evaluate different advertising and promotion offerings and distribution channels.

In short it is one of the best consumer research techniques with thousands of application and articles.

The results of conjoint analysis can be very helpful INPUT to managers who use the AHP taking the consumer input and other considerations in determining their optimal strategy.
For example even if I have info on price elasticity from a conjoint analysis study I can use this as a guide to a pricing decision and will be better off including the consumer input as one of the inputs to management deliberations.

Another major difference is that the strength of the AHP is in group decision making (conjoint can not do it)

The major benefits for AHP is the process that allows management to articulate assumptions, structure their decision process, focus on both the short and long term via key scenarios, bring all relevant information (including results of conjoint studies) into the discussion, conduct sensitivity analysis, and get a buy in by all the managers involved.

Conjoint can not do any of these nor is it designed to do it (this is where Jerry's portion of this blog ends).

For example, if you are trying to decide on the best marketing program mix to achieve a specific marketing strategy, you might be considering programs in video, events, Internet, and print. Conjoint has no criteria – just alternatives that you are comparing. You could use it to ask consumers which among a specific set of print campaigns they might prefer (although you won’t have the degree of preference and you won’t have any measure of consistency about that preference), but it doesn’t allow you to compare those alternatives to a set of strategic objectives or tell you how well would very different types of alternatives in different areas would compare, i.e. how a video campaign would deliver to the strategy vs how an event delivers to that strategy vs an Internet program vs. a print campaign, because the features of each of these alternatives are so dramatically different. Aligning the marketing or product mix with the strategy, developing the relative priorities, and allocating resources are not possible in conjoint.

The AHP is a prioritization and resource allocation approach much more fitting to marketing executives who need to align a host of often very different marketing program alternatives to a business strategy. It enables you to generate innovative marketing mix strategies, evaluate the various strategies, and reach consensus among conflicting interests (group enabled). Specific uses include:

- determining the desired target product portfolio and allocation of resources among the components of the portfolio
- determining the desired directions for new product development or introductions (McNeil case)
- generating and evaluating marketing mix strategies

Because AHP uses “ratio scales”, comparing the importance of one criterion against each other criterion, it enables executives to compare and assess the relative importance of very tangible objectives such as “sales growth” with more intangible objectives such as “building brand loyalty”. This is literally a comparison of apples and oranges that everyone thinks cannot be done. AHP is the only approach that can do this because of the ratio-scale pairwise comparison process.

Finally, when you get to the evaluation of the marketing alternatives against those strategic objectives, AHP enables the marketing execs to then compare very different types of alternatives and assess how much each alternative will deliver towards the strategic objectives. Specifically, you can evaluate how a given “Internet marketing” program delivers to a strategic objective vs a print campaign or an event. The Internet marketing program might contribute to the “sales growth” objective very well, but not do much at all towards “building brand loyalty”, whereas an event might deliver much more effectively to “building brand loyalty” but fall short on measureable “sales growth”. The outcome of the AHP process is the prioritized list of all of the alternatives and how much they contribute to the strategy in an exact, structured and quantified way. i.e. the best marketing program has a priority of .96 while the worst is a .07. This tells you that you should be willing to invest 14X as much in the top program compared to the bottom, and enables you to focus on a very innovative marketing mix. The priority of the alternatives then provides a platform to allocate $$ for budgeting to the campaigns that will deliver the most value for the business. You are maximizing the investment towards your strategy, and thus the value delivered from that investment. Very different than conjoint, you know with AHP and Decision Lens exactly how much bang for the buck each marketing program is delivering for you against the strategy.