All Models are Wrong, but Some are Useful

The Kano Coffee Shop Revisited

In my ongoing quest to explore coffee shops, I recently came across another of note in one of the small towns near me.  From the street, it wasn't too obvious, and on closer inspection, it had an eclectic mix of tables and chairs, curious décor, and other bits of furniture placed around the edges of the room - and so looked interesting.  It was certainly buzzing, and smelled like a good coffee shop should.

While waiting to place my order - I noticed a notice - which made me chuckle: 

                    We don't have Wi-Fi

      You'll have to make conversation

                     Pretend it's 1995

 A little while ago, I wrote about another coffee shop and used the Kano model to assess it.

The Kano model says that customer satisfaction can be considered under 5 sets of attributes; for a more complete description of the attributes, you can read my Kano Coffee Shop blog, but here is a quick summary of the attributes:

• Threshold Attributes  are the basic and fundamental attributes that must be met; if you don't have these, then don't even bother. If you are a coffee shop - you need to sell coffee.

• Performance Attributes go beyond those threshold attributes, and are the things that customers hope for or even expect, but can vary between competitors. Perhaps a range of coffees, types of milk, pastries etc.   Wi-Fi would often be considered one of these.  And this is where most companies - regardless of product or service - compete causing clients to weigh options based on availability of features and price.
• Excitement Attributes. To truly shine and differentiate, a product or service should include an exciting or innovative feature.  Invariably, this attribute can be implemented at no or little additional cost.

The Kano model also cautions about 2 other factors: 

• Reverse Attributes inadvertently work against all the other good work you could be doing and result in decreased customer satisfaction. Avoiding these "reverse" attributes is essential to maintaining a positive experience for customers. 

• Indifference is another type of threshold. It is the aggregate effect of the other 4 attributes; below a certain level, customers will feel indifferent.   Above the threshold is where brand and loyalty is built, and a coffee shop can become someone's one and only place to go. 

Take nothing for granted - Challenge conventional thinking!

Wi-fi has been a Performance (and almost a Threshold) Attribute of coffee shops since the early 2010s.   It has significantly changed coffee shops as it has enticed in more and a broader range of people, those people spend longer there, on the whole they do buy more despite the 'wi-fi squatters', and overall the effect has been revenue positive. 

And so, whilst most coffee shops wouldn't even consider not offering Wi-Fi, my new found coffee shop is a great example of challenging conventional thinking.  It's range of good coffees, drinks and food, the quirky design and the ambiance, all combine to overcome indifference.  And they seem to have turned a missing Threshold / Performance attribute into an Exciter!.  As I looked around, everyone did seem to be engaged in animated discussion - just as directed by the notice. 

The Moral of the Story: All models are wrong - Some are useful *

While models like the Kano model are great, they are at best merely generally applicable and certainly not infallible - they are tools to help you think and ask questions. Sometimes, the key to creating a unique customer experience lies in challenging the status quo and understanding the deeper desires of your audience. And innovation, by definition, implies doing something different from the norm. 

By staying attuned to what genuinely resonates with your customers—whether it’s high-tech convenience or a nostalgic escape from digital saturation—you can carve out a distinctive niche in a crowded marketplace.   Use models to help you think round a subject and get most of the way there, but there is no substitute for good old market intelligence and knowing your customer.  Chatting to the owner, she had never heard of Kano, but she did know her customers. 

Graham

*The quote “All models are wrong, but some are useful” is attributed to the British statistician George Box (1976). He was emphasising that statistical models are simplifications of reality and cannot capture all the nuances of the real world, but despite their limitations, they can still provide valuable insights.