Omniata BlogAugust 12, 2015

Motenashi, UX, & Data

Increasingly improving online experiences has raised user expectations, and it has become critical to align user experience with an organization’s data vision and journey.

In a recent article in Fast Company Design, John Pavlus quotes the great Charles Eames, the mid-century designer whose furniture now ubiquitously decorates tech offices everywhere. Eames, Pavlus cites, once said that the “role of the architect, or designer, is that of a very good, thoughtful host, all of whose energy goes into trying to anticipate the needs of his guest”. While no one reading this is designing furniture, many of us do spend countless hours thinking about the digital user experience.

I was recently on a call with my airline, where after about 4 or 5 actions to identify myself, I was transferred to a live operator with whom I had to go through the whole process again. It sucked, but here’s the thing: when calling an airline, I’ve long ago accepted an underwhelming experience as the norm. My expectations are so abysmally low that after 20 minutes of pain, I hung up thinking it wasn’t too bad.

Now, compare that to your experience with Uber and how rude it seems when they have the temerity to underestimate your pick up by a few minutes. Constantly improving digital experiences continue to raise our expectations; accordingly the importance of user experience is shifting from design priority to essential business objective. In pursuit of improvement, many of today’s top companies are working to align user experience with an organization’s data by better planning its data vision and journey.

These are tall orders for any interactive experience, even one powered by scores of servers and artificial intelligence. But that’s just it: because our increasingly context-aware, "smart" stuff has raised the bar on our psychological expectations, these experiences have an urgent imperative to rise to meet them—or risk alienating us in ways that "dumb" objects never could. - What Japanese Etiquette Can Tell Us About Good UX Design

What does Japanese hospitality have to do with your data and your users?

If you’ve ever eaten at a nice Japanese restaurant, the thought and care put into everything is quickly evident. Experiences are always exquisite, if delicately balanced between the minimal and the useful. Quoted above is Pavlus on Kerstin Blanchy’s exposition on motenashi, a Japanese primer for the guest-host relationship, specifically calling out three main concepts: anticipation, flexibility, and understatement.

If users are your guests, how do you become a better host? Big data can provide previously unimaginable insights into how users think and behave. But, where some will use data for better acquisition, engagement, and monetization, others get caught in an undertow of spiraling costs, dizzying complexity, and constant flip-flopping.

The marketing technology space is growing on average 170% every year according to Scott Brinker’s ChiefMartec supergraph; with best of breed-challenging technologies surfacing every 6-9 months or so. It’s expected that this growth rate will continue, which means companies should expect to change or add tools at an increasing rate. As such, marketing stacks can no longer be tool-centric, they instead need to be purpose-centric. An increasing trend is for tools to rely on interoperability and playing well with others, eschewing those that don’t.

In order to create, manage, and deliver experiences that will lead to better-desired outcomes, consider the 5 bullets below to align holistic, long-term data journey objectives to requirements in the short and mid-term.

  • Metrics: You’re measuring basic metrics: ARPU, Session Time, Retention, etc, but beyond just looking at directional indicators or trends, are you working to identify underlying causes?
  • Underlying Metrics: Peel that onion! Your business has numerous unique things that set it apart, consider how certain features or user behaviors impact the above metrics. This isn’t easy, but you can’t manage what you can’t measure.
  • Data Consolidation: It’s not uncommon to find multiple dozen applications running in a marketing stack. In fact, numbers can run as high as over 100 apps. A big concern then becomes the creation of silos. Keeping data separate, user from event, for instance, can haze or limit the complete picture. User data, event data, etc, - no data set is an island.
  • Test and Learn: At this point, you are able to identify trends, underlying drivers, and can separate correlation from causation; it’s time to test your hypotheses.
  • Anticipating User Needs and Wants: You understand your user, have validated hypotheses, know what works and what doesn’t, how do you move beyond automating interactions based on preset rules? You’ve reached the stage where you can anticipate a user’s needs while being flexible enough to recognize how and when to engage. Getting to Understated is for another post.

The above data journey, designed to track and manage the user experience, can consist of dozens of applications, with little alignment or coordination between business users. Rather than considering the user experience/journey holistically, a common pitfall is to break it down into disparate threads and analyze components of the journey individually. A usual outcome of this approach is tangled layers of unaligned tests, overly complex and disjointed user experiences/journeys and randomly distributed data sets. And usually a single marketing team member that holds the map to maze (and your company’s value) in his hands.

Fortunately, the Resistance is growing and more and more folks are demanding openness and interoperability from their vendors. While still a long way off, a unified approach to data platforms is slowly prevailing.

All that said, your data strategy isn’t only about tools or technology, consider the organizational component.

How Airbnb became the consummate host

Airbnb hired their first data scientist while they were still less than a ten-person team, still working out of a founder’s apartment. In a candid piece, Riley Newman, Head of Data Science at Airbnb, tells the story of data science-fueled growth and what it took to grow 43,000% in 5 years. Two things key things Riley shared:

How does your organization think about data?

Data is an opportunity to listen to the voice of the customer. Data isn’t just a stream of anonymous numbers, it’s a record of actions or events that tell a story about a user and their decision making process.

Treating data as an opportunity to better understand your customer helps elevate projects from one-offs like ‘where’s the best placement for this button’ to ‘what should be prioritized in the product pipeline to increase engagement’ and helps create an organic loop that measures efficacy and validates outcomes.

How does your organization use data to make decisions?

In an environment where according to the Economist Intelligence Unit and Teradata, almost 40% of CEOs think relevant insights are being gleaned from their company’s data, but only 19% of VPs agree, there’s an obvious disconnect.

Airbnb closes this gap by not embedding but aligning their data scientists with their business teams in a cross-functional manner, creating a crossover where the data folks are tuned into the end-user and are better able to anticipate the needs of the business team.

This shift is a catalyst for the transformation from a reactive data team that has to wait for requests in order to act, to one that’s proactively looking to solve problems and ultimately improve the product and user experience.

Key takeaways

Being a good host is an ethos well applied to managing the user experience. Learning to provide increasingly better experiences that drive acquisition, engagement, and monetization, is an ever evolving, difficult task. That task is further complicated by what is essentially lack of holistic perspective and planning.

It’s crucial that in addition to baseline metrics that are usually only intended to be directional, you dig just a little deeper and identify underlying drivers for those baseline indicators.

At the same time, avoid situations where you’re chasing the metric and losing focus of the bigger, more important picture. Before you know it, you’ll be chasing all sorts of metrics in disparate, disconnected applications for incongruous purposes, instead of seeing the forest from the trees and managing your data in a unified manner.

Finally, leverage your insights, test, and validate your assumptions early and often; predictive tools are very much reaching a credible state of usefulness, but only when built on a strong organizational foundation.