Managing User Experience Performance

Part One: Understanding the Experience/Expectation Gap

Why smart near term choices on improving user experience can be liabilities later. 

Developing complex products today - cellphones, digital music, cars - is challenging because there are many ingredients that must go into providing a satisfying user experience for customers. Everyone one wants to make a compelling, coherent experience, as customers are becoming more sophisticated about expecting them, but few companies have the resources, expertise, budget or time to develop every element themselves - interfaces, controls, web applications, operating systems, retail integration, etc. So you have to make decisions about what you're going to do custom, and what you're going to get off-the-shelf.

These decisions have major impacts on not just how customers will perceive you, but also your future flexibility and growth path as a company. They also have a dramatic effect on profit margins.

User experiences go through lifecycles of improvement. Understanding the lifecycle for your industry is critical to smart development choices.

Here I'm taking an expansive view of "user experience", a notoriously slippery term. I'll talk a lot about Apple and digital music here, but also think about MINI and how they've created a seamless experience with a consistent cheeky personality, from billboard and print advertising, to showroom environment, website experiences with whimsical features and microsites (including a hilarious faux lobbying group), and a line of stylish bespoke accessories that continue the post-purchase experience into other areas of life, or allow wannabe owners partake of the MINI cult.

When developing such a multi-faceted and complex experience, you have a number of options that span two extremes:

  1. Take a number of available off-the-shelf items that affect the user experience (operating system, interface, physical controls, internal components like cpu or memory, web code, etc.) and do more or less customization to provide a coherent experience. This reduces go-to-market time and cost, but removes a lot of control from your hands that translates into brand equity.
  2. Pick key items, such as the interface, and do them from the ground-up internally. This gives you more control over the experience, but increases risk, cost and time.

How do you decide where to place your bets? The answer requires evaluating several criteria, two of which are understanding where the product category lies in a lifecycle of experience performance, and what the extents of the experience system are.

In his well-known series of books, Clayton Christensen talks about how industries transition from vertically integrated systems to modular networks of commodity components as certain levels of performance are met and exceeded. He approaches performance primarily from a functional or technical perspective: speed, reliability, longevity, etc.

User experiences in a product category evolve similarly. There is a predictable lifecycle of user experience performance that a product category goes through as it matures, and as it does it transitions from relying on vertically integrated proprietary systems to provide those experiences, to systems made up of modular components from different vendors, collaborating around established standards.

(Functional and experience performance are often inter-related - better functional performance can allow better experiences. But it's not always true: the early Palm PDAs had terrible functional performance - low res screens, slow processors, minimal memory -- but provided great user experiences. The opposite was true for early Windows CE/Pocket PC products.)

Understanding where your industry or product category lies in this lifecyle should greatly influence how you make choices about development of technologies and components, and in choosing partners that might supply those.

The clusters of technologies and components required to generate user experiences (input devices, software, web interfaces, networking and connectivity, file formats, etc.) typically start out as proprietary, integrated combinations in order to provide the desired level of experience performance. There are two reasons for this:

  1. Performance is sub-par for the majority of potential users. Only early-adopters who seek a key aspect of the system will look past the performance issues and embrace it anyway
  2. Flexibility and high levels of control over the user experience is required, as the first product offerings are often somewhat experimental in nature, as there is not yet a mature understanding of where the product category will evolve, and what the mass market will want from it.

The diagram below shows how the user experience performance improves over time, gradually catching up to the point where the majority of users will be satisfied (the fuzziness of the line is because not all people have the same standards).

ue_lifecycle_graph1.gif 

In the early stages of a product category, there is an experience/ expectation gap - a mismatch between what people find acceptable and what the system can deliver. Everyone of course is trying to close it, because until the crossover of the two lines, the product category is not ready for mainstream adoption. But trying to close the gap with a modules provided by multiple vendors can drastically increase the time it takes to reach that crossover.

Digital music is a case in point. Early products were individual components of the whole digital portable user experience, but until Apple no-one offered an end-to-end solution that stripped away the complexity. BiP (Before iPod) it seemed obvious that features and price were the way to go. This led to combinations of portable players and PC jukeboxes that individually were good, but which had little in common in terms of experience. AiP (After iPod) it was blindingly obvious that it was the ecosystem that mattered, not features of the individual components.

This diagram shows how Apple used the benefits of the vertically integrated system in order to close the gap and shortcut itself to the crossover where user experience performance matches customer expectations for the mass market.

ue_lifecycle_graph2.gifApple learned from its history however, and craftily incorporated externally- developed user experience elements (Soundjam, Synaptics, Veridisc, Portal Player and Pixo)

After multiple companies reach the crossover, the benefits of an integrated system can become liabilities

Apple has now shown Rhapsody and Napster, Samsung and Creative, what an acceptable user experience looks like, and of course they are racing to meet it as well. To achieve a similar ecosystem to Apple’s these other companies must rely on partnerships (e.g. Samsung’s alliance with Napster) and external standards that allow co-development of products based on known interfaces (e.g. Windows Digital Rights Management to facilitate getting music off the PC and onto other devices). Once any one or combination of those companies reaches the crossover with their modular solutions, two challenges will  face Apple.

  1. The profitability of the systems drops dramatically, and Apple loses its ability to charge a premium price for a premium experience. (See the green line in the diagram below.)
  2. The ability of Apple to sustain an innovation lead will be severely tested, as each of the specialized companies developing components for the modular systems will have greater expertise and greater focus, and if history is any guide, will push the capabilities of their systems further in terms of performance, features and cost/benefit than an integrated system such as Apple’s will be able to achieve. In other words, the experience performance curve for competitors will continue up at a steeper rate than Apple's after the crossover.

ue_lifecycle_graph3.gif 

This is why the method used to achieve the user experience has great impact on strategic choices for development and alliances. You want to be very careful assessing not just where your industry is on the lifecycle today, but where it will be when you launch and shortly thereafter. Making the wrong choice will either commit you to a prematurely modular system that, while cheap and fast to implement, won't close the experience/expectation gap (so threatening sales); or will commit you to an integrated system that quickly gets you to the crossover, but which is expensive and in the long term less powerful and flexible (short term gain for long term loss).

This was exactly the situation Apple found itself in in the with the Macintosh as the modular IBM PC's gradually closed the experience/expecation gap to the point where they offered a "good enough" experience for most people, with superior speed and price, a combination that almost proved fatal for Apple. It remains to be seen whether they are doomed to repeat the same cycle again with digital media.

In a Part Two, I’ll have some thoughts on how to deal with the experience crossover if you’re inventing a product category, what qualities to look for in development partners, and how to manage your company through the crossover.