At its best, MaxDiff can help an org understand which product features or attributes will resonate most with core constituencies by forcing participants to make tradeoffs between what can seem like equally desirable product dimensions, ultimately compelling a 'top' choice among many.
The result is a clear signal about relative preferences, which can be especially valuable when your product partners are staring down a long list of potential features to ship. But here’s the catch: Timing and intent matter, a lot. Running a MaxDiff study at the wrong moment can make the whole exercise a waste of time. So let's first review when we should absolutely not run a MaxDiff.
When MaxDiff Doesn’t Add Value
- The roadmap is already set. If product managers and leadership already know which features are getting built in the next cycle, MaxDiff won’t change the outcome. At best, a MaxDiff will show support for features that would be forthcoming anyway. At worst, evidence showing certain in-flight features won't appeal to consumers will be dismissed or written-off, and the messenger (that's you) promptly fired out of a cannon. (Not really, but you get the idea)
- Features or attributes are complex. The descriptions do all the lifting in a MaxDiff, so the more complex the idea described, the harder it will be for respondents to prioritize items in a set. A good MaxDiff uses universally understood product features or attributes. For a car brand, that might involve simple descriptors like 'fast', 'reliable', 'good gas mileage'. It's harder for respondents to prioritize concepts that are complex. For example with a car line, a 'hydrostatic continuously variable transmission (CVT)' is not something that's readily testable with a typical audience.
- Features or attributes are nascent. To answer a MaxDiff sensibly, respondents need mental models for understanding the features or attributes described. Revolutionary or emerging features shouldn't be used in a MaxDiff, as respondents may not be able to provide valid answers. Imagine describing the first iPhone in a MaxDiff. In that case, a new/revolutionary idea may be dismissed because respondents will gravitate to ideas they already understand.
- Design-level answers are needed. MaxDiff works at the level of “Feature A or Feature B?” and not “What’s the best placement for this control?” It helps with broad prioritization of features or attribute, not execution. And without qualitative feedback, you won't know details about why certain features seem appealing and why some don't.
 |
| Having consumers prioritize doughnut toppings against each other is easy, having them prioritize their chemical ingredients, not so much |
When MaxDiff Really Helps
- You're still early in the process. MaxDiff is most useful when you’re at the idea stage, when the list of potential features is still long, and no one’s sure what general direction to move going forward. That’s when it can help eliminate the “nice-to-haves” and surface the “must-consider” items. Take it from me: If a stakeholder asks you to conduct a MaxDiff late in the cycle, after product managers have already laid out the roadmap, you should decline. Instead, you should think about concept testing and other approaches that will get you detail on how to improve features.
You need to provide broad direction. Think of MaxDiff as a way to get a directional map of consumer or user interest. If certain groups of attributes or features resonate with users, they'll emerge toward the top of the MaxDiff preference rankings, and you'll be able to give stakeholders an idea of the type of feature they should consider developing and shipping. It won’t tell the org how to build something, but it will highlight which feature or attribute areas deserve more attention. For instance, if you see 6-7 features dealing with reliability in your automotive MaxDiff, you've got a general idea of what customers want.
You have a team that's open to input. MaxDiff works best when stakeholders are willing to let user preferences shape the conversation. If there’s a genuine appetite to hear what users are interested in, MaxDiff can provide a clear, evidence-based perspective. But again, timing matters. The earlier in the planning process, the more willing product managers and leaders may be to hear contrary evidence. And contrary evidence that shows consumers do not prioritize certain pet product ideas will emerge!
If there’s a genuine appetite to hear what users are interested in, MaxDiff can provide a clear, evidence-based perspective.
The Takeaway: Timing Is Everything
The main lesson is simple: MaxDiff is most valuable when you use it early enough in the product development window to influence decisions. It’s far less effective if the roadmap is already locked, or if you’re trying to answer questions that require a more detailed research approach.
Done at the right time, MaxDiff can quickly clarify broad areas of user interest, giving your team confidence to move forward. Done at the wrong time, a MaxDiff will produce yet another piece of research that doesn’t get utilized in product planning.
So before you launch a MaxDiff, you should ask: Are we still exploring options? Are stakeholders open to letting user input shape direction? If the answer is yes, then MaxDiff can be a great tool. If not, save the budget for other methods that will actually inform the work ahead.
Comments
Post a Comment