Use Kano Analysis to Prioritize the Product Backlog

Photo by monstroDeciding which features to put into the next release is one of the more agonizing decisions that a product manager has to make. If you are like most, you’ve got a list of desired features which far exceeds the time and resource constraints under which your next release will be created. Leaving some of these features on the futures list can feel like abandoning a child. So, how do you decide which will make the cut? In this article, we’ll take a look at how to use the Kano Model to prioritize the product backlog.

Kano Analysis is one of several techniques you can use, not only determine what goes into the release, but also to explain your decisions to stakeholders. The Kano Model was published by Dr. Noriaki Kano in 1984, for which he was later awarded the prestigious Deming Prize for making an outstanding contribution to the study of TQM or statistical methods used for TQM. Though this sounds like heady stuff, it’s really quite simple to grasp.

Kano Analysis allows you to classify product (or service) features according to their impact on customer satisfaction. The model divides product features into three categories: fundamental, linear, and exciters. To be competitive, your product must satisfy all fundamental features, optimize the linear features, and include the “right” exciters. As you might guess, the juice is in the exciters.

The model is depicted graphically by plotting a given feature on two dimensions:

For example, if the feature is “portability,” then your product would be at the far left end of the x-axis if it weighed 50 pounds and had to be plugged in to a wall socket to function. The product would be on the far right end of the x-axis if it fit comfortably in your pocket and came with a rechargeable battery. Determining the y-axis value is a little more tricky. If portability is the norm for your product (like a cellular phone) then even with the best implementation, you might not get much higher than midway up the y-axis. If, on the other hand, portability is not the norm for your product (like a microwave oven), then even the poorest implementation could get you to the top of the y-axis.

With some relatively easy customer research, you can plot features on this graph. Once graphed, features can be placed into one of the aforementioned categories based on their position on the graph as illustrated below.

Kano Analysis

Fundamental Features

Fundamental features are “must haves” for the product. Think: brakes on a car, remote control for a TV, A/C in a house in Arizona. You aren’t going to win a customer with any of these features, but you’ll almost certainly lose them without. Though they are typically not very interesting, you must have a sound understanding of the baseline features for your product. This is more difficult than it sounds, because customers won’t often talk about them. Missing just one can be disastrous, so be diligent.

Linear Features

These are often called “performance features” because more is generally better with them. Think: gas mileage for a family car, screen size for a TV, and square feet for a home. Talk to customers about your product, and these are the features you are likely to hear about.

Linear features represent the world of standard competition. The more you have of them, the better your product is in the mind of the customer; and the more customers are (usually) willing to pay. The inverse is also true – the fewer you have of them, the worse your product, and the less customers are willing to pay. While your product must offer the right performance to price ratio, guard against getting hung up on these features.

If your product has been around for a while, you almost certainly have a good grasp on the linear features of your product. The trick here is not so much understanding what they are, as to understanding where the sweet spot is for the competitive environment.

Exciter Features

These are the “oh cool!” features of your product. These are the features that your customers, though they weren’t expecting them, love. These are the features that help you to create brand loyal customers.

Unlike baseline and linear features, the perception of your product is not damaged by the absence of an exciter feature. Because your customers aren’t thinking about these features, they don’t gig you for not having them. This aspect of exciters provides one of their most interesting attributes; they don’t have to be perfected in order to have a positive impact. Think about the first iPod Shuffle – it was hardly perfect, but it’s small size was exciting to customers. Consequently, the iPod Shuffle captured 58% market share in less than 2 months on the market. Exciters are indeed exciting.

It’s not easy to create exciter features, which is why so many companies choose to compete on price and linear features alone. It’s unlikely that your customers will help much here. They won’t be talking to you about these features until after you deliver them. This is where innovation, design, and market intuition come to bear. Invest here – it is worth every penny.

The Kano Model in Action

By now you might be asking yourself, how do I go about gathering the data used to plot the features on the Kano Model graph? Customer surveys are a great method, and all you need ask are two questions per feature. The first question measures the customer’s satisfaction with the presence of the feature, the second measures the customer’s satisfaction with the absence of the feature.

Kano Analysis Questions

  1. What do you think of the product if it includes feature X?
  2. What do you think of the product if it does not include feature X?

There are three valid responses for either question: “I like it,” “It doesn’t matter to me,” or “I dislike like it.” I personally prefer to ask customers these questions verbally, and then categorize their answer into one of the three valid responses. Doing so allows you to gather additional information, as well as ask follow-up questions. Follow-up questions concerning price can really help in making trade-off decisions for linear features. Free form verbal responses also provide you with a better idea of the strength of satisfaction, allowing you to better place that feature on the y-axis.

The answers to these questions determine the Kano categorization as follows:

If you receive any other combination of answers, you should dig a little deeper on the issue. Perhaps your customer is indifferent to the feature; in which case you would be wise to omit it from the release. Perhaps you’ve stumbled upon an anti-feature; a feature that the customer would prefer not be present, but which they have learned to live with (THINK: Clipit or DRM). This is a pleasant discovery if you can remove the feature, as it will have a beneficial impact on overall satisfaction with your product.

Once you’ve categorized the features on your backlog, you can begin to think about the next release in terms of the implications on customer satisfaction. Note that there is no “right” mix. Your mix of features will vary depending upon market conditions and product life cycle state. New products will include many more fundamental and linear features, whereas mature products should include more exciters.

In addition to a useful new framework for assessing and prioritizing the backlog, you now also have (somewhat) objective data to discuss with executive management. In my experience, this is a very good thing.


If you enjoyed this post, please consider to leave a comment or subscribe to the feed and get future articles delivered to your feed reader.


Nate – Great piece. Interesting, succinct and informative… ranks high on my Kano Analysis.


Thanks Matt, glad you liked it.

[…] in relation to customer satisfaction. Scott Sehlhorst breaks it down pretty well, but I prefer how marketada puts […]

A variation on these two questions would be:
What do you think of the product if it solved problem X?
What do you think of the product if it does not solved problem X?

The challenge of talking about features is that people think they know what the feature is and it’s hard to overcome their expectations of the implementation… just look at the post-announcement commentary on Apple’s iPad. Lots of anticipation for a feature set driven by hysteria, then after, lots of moaning and groaning because feature x wasn’t included.

Any system based on customer preference rather than employee opinion wins my vote.

Thanks for sharing your approach.

Aftyer ? originally commented ? ?eem to hav? clicked the -Notify m? wben ne? comments aare
a?ded- checkbox ?nd f?om no? on w?enever a comment is added I receive
four emails w?th the exactt ?ame ?omment. Perhap? there is ann easy metod you can remove me from that service?
Thaks ? lot!

Leave a comment