Identifying the magnitude of different influencing characteristics

Asking the right questions in the best way

As researchers we are frequently tasked with asking the right questions in the best way to ensure the data we return to a client, helps them make the best decisions for their organisation.

Often a right decision is dependent upon understanding the priorities that exist within a market and different types of market research. However, we know buying decisions involve a long list of different influencing factors. What we need to know is how important are each of these factors, and which are the factors that are most prioritised within the audience.  This is as important in content marketing research as for traditional research briefs.

Issues with standard approaches

There are several ways to approach questions of importance or priorities in research, many of which you might be familiar with. You can ask survey takers to tell us their strength of feeling of each item on a given list, or to select out of a list which items are most important to them, or even to rank the whole list, from high to low.

However, each of these methods runs into difficulty when it comes to asking customers to make trade-offs which we need to do if we’re to understand priorities, especially when we are looking into the full range of potential buying influences on a customer journey where there are a long list of items to consider.

What is MaxDiff and how can it help?

MaxDiff is a questioning technique that can be included in any survey where we need to understand how individuals in a market make a buying decision. This could be if we want to understand which features to include when developing a new product or service, if we want to understand how behavioral attitudes are impacting a buying journey; or even if we want to know which pizza’s to include on a menu. We’ve also used this marketing research method successfully in campaign planning, really getting to the crux of the message components that will have the largest impact.

Say a client gives us a long list of answer options to one such question, MaxDiff allows us to split this list into a series of questions, each with a varying combination of just 4 or 5 items each time. Survey takers then go through this series of questions, selecting the most and least influential to them on the list each time.

Why should you consider MaxDiff?

This technique has enormous advantages, firstly we can test many more factors whilst maintaining the quality of responses as respondents are only considering 4 or 5 factors at a time, not a lengthy list that is much more difficult to process. Secondly, by forcing respondents to make trade-offs we can much more clearly discern how our audience truly prioritises in decision making. It further eliminates the scale bias that can emerge in scale questions where respondents simply select every factor as important to them, leaving us as researchers with limited power to distinguish between factors and respondents.

How to use MaxDiff

When using MaxDiff we will take the long list of items you want to test and return data in terms of scores that can either by percentages or indexes that tell you both the how the factors rank in importance and how much more important each factor is to each other in the list, for example a factor with a score of 10 is twice as important as one with a score of 5. 

You can also take MaxDiff a step further and use the data in a persona analysis where we segment your audience based on similarities in the preferences and priorities determined through MaxDiff.

So, thinking of using MaxDiff….

All in all, MaxDiff is the perfect research tool in a crowded high-noise market when incremental insights are needed to set you apart.  It allows us to drill down on any number of buying triggers for our clients whilst returning detailed and strategically relevant insights. Once the priorities in your market are known you can really improve your conversion by communicating what your audience most needs to hear.

Click here to see where it’s used and how it’s done.