The life cycle of successful product features

The life cycle of successful new features

What you once found amazing is now boring. You can use that to your advantage.

There was a time when car navigation systems were rare. “As if I can’t read a map!”, my dad used to say when that technology became more present in cars.

But that changed quickly. He soon started to see the value of a car navigation system. To many people, a navigation system became an attractive quality of their next car. And soon after, as more and more cars were fitted with navigation systems, performance expectations increased (more colours! voice control! auto-updating maps!).

Today, a navigation system is an expected feature of any car. (“As if I’d buy a car without a navigation system!”, my dad says now). But that can soon change again, with smartphones taking over the role of navigating the driver.

Buyers are becoming indifferent about a built-in navigation system. Some even do not want (to pay for) a navigation system in their cars anymore.

It’s not hard to think of other examples that follow the same path as the car navigation system.

Instagram’s photo filters for example went the same route. In 2010, when more and more people started sharing pictures online, Instagram launched its photo filters. “Real” photographers turned up their noses: “Who would want their photos to look old?”. Photo filters were an Indifferent or even Reverse feature.

But the value of the filters soon became clear to Instagram users. Instagram introduced photo filters as a response to the problem of crappy photograph quality on smartphones.

People who started using the filters also noticed that the attractiveness of their photographs increased. One study shows that “filtered photos are 21% more likely to be viewed and 45% more likely to be commented on by consumers of [a] photograph” (Bakhshi, Saeideh, et al, 2015). So photo filters became an Attract feature.

As other apps started integrating photo editing features, the sophistication of these editing tools further became a driver of user satisfaction. Photo editing as a feature had moved to the One-Dimensional, “the more, the better”-category.

By now, as photo filters have become an expected feature of any photo app, customer satisfaction no longer increases with the editing tools’ performances. It's a Natural feature.

You’re starting to see a pattern here, aren’t you?

Dr. Noriaki Kano thought so too. He researched features such as TV remote controls over a span of 15 years. His (Kano, 2001) and other research (Löfgren et al, 2011) shows that successful products follow this life-cycle:

Indifferent ⇒ Attract ⇒ One-Dimensional ⇒ Natural

• A new feature is met with indifference. Users don’t know what to do with it

• Once users start realizing its value (through use, seeing other people use it or through advertising), the feature can become an attractive quality of the product;

• Expectations rise as competition kicks in: the feature must increase its value to the user to maintain satisfaction levels;

• The feature plateaus and becomes an expected feature. Increased performance no longer contributes to higher satisfaction.

A new feature’s life cycle starts in the Indifferent or Reverse category.

People have no idea what to do with a new feature or what to use it for. It’s not hard to imagine the indifference of someone back in 2010 who was told about a photo sharing app “that makes your pictures look old”. Why would anyone want to do that on their brand new smartphones?

Once people discover a feature’s value through using it, seeing other people use it or through advertising, the feature can begin its journey to success.

New features that begin life in the Indifferent category can be destined for success. But mature features in that category were either always there or came from somewhere else. These features are (becoming) redundant; you should not give them further attention.

Don’t jump to conclusions too soon whenever you see a Kano analysis where one of your novel features is in the Indifferent category. That feature could very well grow into a big success. Users may simply have no idea what to use it for yet. You will have to help your users discover the value of the new product feature.

Using the life cycle to stay ahead

Successful features follow a predictable path across Kano model categories. You can use this knowledge to your advantage. If you know if and how a feature will begin to impact customer satisfaction differently, you can plan ahead and optimize for user satisfaction.

The Kano model cannot help with the very first stage, the ideation around new features. Imagining new features is part of another process altogether.

Instagram wanted to help its users make their photos look good. Filters were an answer to a latent user requirement.

You won’t find latent requirements by asking users how they think you could improve your product. A fish crawling out of the primordial soup wouldn’t say “I need legs!” when asked what would improve its life. By observing the fish and what it wants to achieve, you might come up with the idea of giving it legs.

It’s just as likely that Instagram users wouldn’t have said “filters!” when asked what would improve their online lives.

You can discover latent requirements through observation. Instagram’s photo filter feature was a flash of genius, but it’s also thanks to:

• Instagram’s keen observation of user behaviour and photo production. The team could just as well have glossed over the crappy photos, think users didn’t care or blame the hardware;

• Their good problem solving and engineering skills (thinking of the filter and developing it);

• The rapid evolution in mobile phone technology. Increases in performance allowed for more resource-hungry and complex software.

Not all new product features will be a success. However, the life cycle implies that when a new feature moves from Indifferent to Attract, it’s on its way to success.

Instagram promoted the use of photo filters by giving them a central place in the app. In a way, it was giving the feature a push from Indifferent to Attract, by making it very easy for users to discover the feature’s value.

That was a courageous move. It’s not unthinkable that some people in the team would have thought filters were a gimmick and should be hidden under three layers of UI.

Once photo filters took off, they became an Attractive quality. Other apps started integrating photo filters too. In 2012, Ars Technica reported

Both Twitter and Flickr have recently updated their iOS apps, adding the ability to apply one-touch, retro-style filters to mobile photos uploaded to their respective services. The features seem like me-too additions after the spectacular rise of Instagram as a top social network.

The mere presence of photo filters was no longer enough of a competitive advantage. Instagram doubled down its efforts on developing more and better filters. It introduced more sophisticated ways of editing photos.

Photo editing moved to the One-Dimensional category and user satisfaction was now related to the feature’s performance.

This stage has all the trappings of an arms race. Competitors are coming up with more and better versions of a feature. This is the moment to take a step back and evaluate the feature. Is it worth improving on every aspect of this feature? Does every performance increase result in increased customer satisfaction? Users expect some parts of the feature to be present, like adding text to a photograph, but maybe being able to choose from hundreds of fonts does not increase user satisfaction.

Subcomponents of photo editing migrate towards the Natural category, while others remain in the One-Dimensional category. Smart product developers realize that. They pour efforts into the subcomponents that increase customer satisfaction and move the ones that don’t into maintenance mode. There’s no need to try and beat the competition with Natural features. Keeping these subcomponents stable is enough.

Splitting the feature into its components and allocating efforts accordingly frees up resources. These can then be allocated to do more impactful work on the product.

The final stage brings new product ideas

After the Natural phase, it is my contention that features either stay there, migrate to Indifferent or become Reverse features.

Look at the history of the Instagram filter. Because the quality of smartphone cameras steadily improved, the filters lost their original function and appeal. Some users even rebelled against filters, and “capturing something that required #nofilter was [...] a badge of honor”.

After a successful feature has gone through its lifecycle, its users have evolved too. New needs are emerging that cannot be solved by increasing performance alone.

Instagram influencers have started creating their own custom filters using specialized tools like Adobe Lightroom. Their goal is to make themselves uniquely recognizable by the tone of their pictures. Some influencers even sell the custom filters they have developed.

This behaviour reveals a completely new user need. You shouldn’t only mourn a successful feature that nears the end of its lifecycle. Look at what it has done to your users and what new behaviours have come up. Use that as a cue for new product features or even a new product.

Summary of actions to undertake per stage in the life cycle

Life cycle stage and required actions for successful new features

Feature Category	Current competitive advantage	Action to move feature along
Indifferent		Promote usage
Attract	Differentiation	Increase feature performance (make it more sophisticated, add bells and whistles, …).
Perform	Feature performance (better, more, …)	Measure satisfaction of incremental changes. Split up into subcomponents. Some will become Must-Be sooner than others. Don’t waste effort on improving these, only improve subfeatures that increase satisfaction.
Natural	None, except if competition does not have the feature	Invest only in feature maintenance.

At the end of a successful feature’s life cycle, observe user behaviour and find out if new segments have emerged (power users, …). Develop new hypotheses and introduce new features.

Detecting evolution

It is impossible to say at what frequency you should survey customer to detect shifts in feature categories. “It depends”, the consultant said. Only by doing and analyzing surveys will you be able to know whether you’re surveying at the right rate.

The presence of features with no clear winning category is a possible indication that these features are moving in one or the other direction. (But it's also an indication of segments in your surveyees). Use these findings if you have only done one survey and want to convince management to invest in more surveys.

The two major parameters that dictate the rate at which features move between categories are:

• Domain. Social media mobile app features move through categories faster and have a shorter life span than features of large ERP systems. Types of users, the domain's acceptance of change, its ecosystem, interdependencies and competition levels, … are all contextual parameters that influence the speed of a feature’s trek through categories;

• Zoom level. The more general the feature, the slower its course through the lifecycle. The life cycle of the ability to pay for a product online moves slower than the ability to pay with Apple Pay.