Post Image

An Actionable Strategy for Data-Informed Design

Angelos Arnis
By Angelos Arnis on 14th December, 2015 Updated on 1st February, 2017

We can all agree that by definition and principle — at least theoretically, design is about solving problems.

Specifically, user experience strives to create a human touch amidst complex technological advances. It is the extra detail that makes a product a bit more fun for everyone, especially for the less savvy crowd.

Many practices can build a good user experience. Those practices differ based on the complexity of your project:

  • For a personal website/portfolio, you can use a very simple approach by asking the clients what she wants to accomplish on her website, then validate with a few quick interviews with employers. Based on that input, you can design a site that reflects current trends and highlights the appropriate skills.
  • To redesign a SaaS platform, you would conduct more field research by visiting customers, listening to their frustrations, then designing some new solutions and observing their reactions/feelings in the A/B testing process.

In this article, we will strive to find a joint force of using data (online and offline), together with human input, user feedback, and designing for the user. Below are the two approaches we will combine into a useful approach with a simple outcome: A kick-ass validated experience.

Human-centered Design (or HCD)

Popularized by the design firm IDEO, HCD is a creative approach to problem solving that starts with the user. 

image04

Photo credit: Angelos Arnis

In a nutshell, it is the empathetic approach to design we all advocate.

  • Identify the core problem with users
  • Brainstorm with all product stakeholders for the best solution
  • Create prototypes for the best experiences
  • Test the more sound concepts with at least 5 users
  • Release a validated experience/solution to the world and keep repeat the cycle for iteration.

Data-driven Design (or DDD)

DDD is a systematic approach to the development and evolution of products and services (digital and non-digital).

image01

Photo credit: Angelos Arnis

As data becomes more accessible, decisions for design should be informed by data. To put it more simply: data complements design.

Quantitative Data-driven Design

Quantitative tactics like surveys help us answer product questions related to:

  • Who are the users?
  • What are the user’s problems?
  • When are the users most active or inactive?
  • Where are the users currently using their solutions?

Qualitative Data-driven Design

Qualitative tactics like user interviews reveal the motivation behind the numbers by answering:

  • How do users behave?
  • Why do they behave that way?

When using data, you must remember to remain empirical and specific.

When we just have data, the ideas we get might become blurry. Through observation and experimentation, we can become specific and the process becomes easier to comprehend and improve.

Fusing the Elements Together

I’ll refer to this immersive approach as a Data-informed User Experience (DiUX).

Let’s dig into the logical steps that we should take for a Data-Informed User Experience.

image02

Photo credit: Angelos Arnis

As an example, we’ll examine Studio by UXPin, the exact resource that you’re currently browsing. Part of the subdomain includes a resources landing page with ebooks. We want to devise an informed decision if we need to make changes and increase conversions when people land on this page.

For this we need data combined with our own instincts. The quantitative data we will use is the bounce rate. For qualitative data we will ask how and why, and then we will brainstorm for better solutions with an A/B testing feedback loop.

Start With Analytics and Observations

The empirical data is a good base for drafting our new combined approach.

We need the resource landing page analytics to tell us what happens when a user comes to this page directly. We also need to understand happens when a user comes to that page through other pages, and what happens after he takes an action to download one ebook.

  • Does he come back after the first visit to explore more of UXPin?
  • Does he visit the page only for the freebies and not the product?
  • Does he arrive by accident and take no action?

Additionally we need qualitative data for our process. This often gets frowned upon because some people cannot visualize these observations as data, but it is crucial to validating the numbers.

For instance, we could follow up our investigation in Google Analytics with 30-minute user interviews that explore:

  • How do the users get to the resources landing page? Why did they decide to click through?
  • Why do some stay for more, and why do others leave right after they completed an action?
  • Do returning users use the page differently than first timers?

As Dave Yeats (Senior UX Designer, Bazaarvoice) points out:

“I’ve come across too many instances of people dismissing qualitative research as ‘anecdotal’ because they don’t understand how non-numerical data is still data.”

After we have gathered this information, it is time to dive into practicing the second part of our new approach which is a bit more practical — and usually more fun.

image00

Applying Our Data

To continue the data-informed process, we create personas based on the analytics and insights from user interviews. The analytics tells you “what” they do (e.g. sequence of pages visited, most popular page), while the user interviews helps you flesh out the persona with “why” they acted that way.  

When you’re done with the persona exercise, you’ve created a clear picture of the user backed up by hard data. The persona is no longer purely qualitative.

We will then share the user insights with the whole team and start to build out prototypes of the most viable ideas. Once we collect feedback on the most logical designs, we will iterate, test with users, and continue refining by releasing the iteration and A/B testing.

Screen Shot 2015-12-14 at 5.24.10 PM

Photo credit: UXPin prototype iterating

Naturally, there’s always more qualitative data to be collected. More observational data and feedback from the users helps us evolve it in the wild the world (note: iterations are always part of the process even after ‘finalizing’ your design). For instance, you should validate the A/B test results with at least a quarterly interviews with 5-7 users.

Key Findings

By combining the empathetic HCD approach with the structured DDD approach, we’re better able to design experiences backed by more than just our gut feelings. We’ve arrived at a holistic design strategy that incorporates business thinking and design thinking.  

We are designers. We are makers. We want to believe we are inventors. But most of all, we are people. Designing a great experience is all about that.

Or as Christopher Murphy puts it:

“The secret to success in our fast-paced industry is, I believe, straightforward: make things, share things and — last, but by no means least — be nice to people. That’s it, really.”

Finding the right balance between data, empathy & experimentation is the distilled result.

Design That Matters: Helping Refugees With Data-Informed Design

Recently a group of refugees in Berlin banded together to map key resources across the city, including counseling, healthcare, German language lessons, accommodation, legal assistance, police and public transport facilities.

Screen Shot 2015-12-14 at 5.20.59 PM

When you examine the project in detail, you see the potential for data-informed design.

The team conducted quantitative research in the form of surveys, aggregated the results, then validated the answers by interviewing refugees. The combined approach helped them understand the most visited areas of the city, and more importantly, why.

Based on the results, the team built an interactive tool designed to welcome those escaping persecution and violence.

This is design at its most human.

At the end of the day, UX design is all about creating meaningful experiences. Designing with people, for people, based upon data we know about people, will always be a repeatable formula for success.

If you’d like to learn more about UX best practices that result from data-informed design (like personalization and meaningful gamification), check out the free Definitive 2016 UX Trends Bundle. The bundle includes 3 ebooks in a single download.

fb-promo-bundle

Still hungry for the design?

UXPin is a product design platform used by the best designers on the planet. Let your team easily design, collaborate, and present from low-fidelity wireframes to fully-interactive prototypes.

Start your free trial

These e-Books might interest you