Empowering Siri’s Designers to use a Data-Driven Approach

CASE STUDY APPLE

 
 

THE CHALLENGE

Empower Siri’s designers to use data to support their design decisions by showing them where to find relevant data, how to get access to it, how to synthesize it, and how to use it to justify their decisions.

ROLE

Design Producer

THE OUTCOME

A detailed document that describes all the relevant data tools for Siri designers. 100% of designers on the team reported that this document is useful for them to justify their design decisions.

 
 

In my first couple of weeks as a Siri Design Producer at Apple, I sat in on many design reviews, where designers would justify their work before stakeholders. One type of interaction happened over and over again:

[After designer presents detailed designs]

Designer: …and that’s my work! I’m open to any feedback. 

Stakeholder: I see that you did [insert feature]. Why? 

Designer: I did that because it’s more intuitive and user-friendly that way. 

Stakeholder: Oh? How do you know? Can I see the data? 

Designer: Um, I don’t really have data to support this right now. 

 

PROBLEM

Siri’s designers were skilled at designing user-centric, seamless experiences. But they struggled in defending their design decisions to stakeholders. Their intuition was on point when it came to user needs, but their designs weren’t always data-driven, making it difficult to justify their design choices.

Their intuition was on point when it came to user needs, but their designs weren’t always data-driven, making it difficult to justify their design choices.

Designers wanted to use more data in their design process, but acquiring data as a designer at an enormous company like Apple was cumbersome — gaining access was confusing, and not all the data available was useful to designers. It was difficult to know which data was relevant, and how to find and understand the data that was collected.

These factors discouraged designers from finding and using data in their design process.

I wanted designers to feel empowered to use data to inform and justify their designs to the rest of the team.

PROCESS

 
 

RESEARCH

I started off doing some research to better understand our designers’ needs.

  1. I set up interviews with 5 designers to ask them how they use data in their design process. 

  2. I identified key pain points for designers when accessing and using data (i.e. hard to access, difficult to know which data is relevant, confused on how to synthesize data).

  3. I sent out a 20 question survey to gauge how many designers were struggling with data-related issues. The survey had questions like:

    • Do you find it easy to find the data you’re looking for?

    • How difficult is it to access the data you need?

    • Is it clear to you what kind of data you need to justify your designs?

    • How often do you use data in your design process?

Survey results showed that designers struggled with understanding where to find data, how to access it, how to synthesize it, and how to effectively use it to defend their designs.

I decided to create a detailed document that served as a one-stop-shop for all Siri designers’ data needs.

Though I can’t share the actual document, here’s an outline of what it looked like:

 

Here’s what it included:

  • All the relevant internal data tools available to designers

  • The kind of data each tool provides

  • Guiding questions to help a designer figure out if a data tool (and data provided) is relevant to their needs

  • Tags to help narrow down the type of tool a designer should use

    • Type: Whether the data is qualitative or quantitative

    • Depth: How detailed the data is

    • Function: What the tool does

    • Technicality: Whether the the tool requires technical know-how to access data

  • Documentation and tutorial links that explain how to use a tool

  • How to access each tool

I presented it to several design teams for feedback so I could iterate.

 

IMPACT

I conducted a post-evaluation of my solution to determine its effectiveness. The evaluation consisted of 5 designer interviews and a post-survey sent to all 10 designers on the Siri design team.

The post-evaluation of my data documentation indicated that 100% of the UX and conversation designers on our team now felt empowered and supported in using data regularly in their design process.

The post-evaluation of my data documentation indicated that 100% of the UX and conversation designers on our team now felt empowered and supported in using data regularly in their design process. Designers reported that the process of acquiring and understanding design-relevant data points became much clearer and easier with the documentation easily accessible.