Oracle: Infinity Streams Redesign

UX Design Intern, Customer Experience Team

My Role

Jane is a marketer who wants to increase sales for a product. She wants to target customers in real-time on the company’s website who have it sitting in their cart, so she can send them an exclusive email offer. While building these rules in a software to target these customers, she realizes she has no idea how many people have that product it in their cart, which products are more popular, and if anybody is even on the website!

With Streams, a separate application, she can write a complicated query to gather all this data and analyze the resulting number with no graphs or visualizations, increasing the time she spends analyzing data to create her campaigns. She doesn't feel confident that her campaign would increase sales, and if she is targeting the right product.

Infinity Streams allows marketers to view real-time behavioral data from their customers across multiple platforms (websites, apps) in real time. When making rules and filters in a separate application to capture this data, marketers are unable to see this data. How can we bridge this gap?

The Challenge

The Design Process

Before uncovering this gap, I conducted research utilizing different methods and sources to learn about pain points, user experiences, and recommendations from stakeholders about Streams currently.

Through initial research, we discovered a problematic gap between Streams and its’ application in another software- marketers cannot see the real-time Streams data when it is most needed.

Competitor Analysis

I analyzed similar competitor products to compare usage of data.

Data Analysis

I gathered key metrics about the most used features and how long users spend on Streams.

I condensed all this qualitative and quantitative data into insights:

1. The current user experience when building rules is problematic - checking for data requires going to Streams and writing queries to drill down on data and then switching back to the rule-building application. They cannot see this data when it is most needed.

2. Data visualizations could be helpful for marketers to quickly derive insights from user data.

3. Competitors address the gap of not being able to see data - but only with historical data

4. Some type of preview of data would make the current experience more marketer-friendly.

   For example, how can we visually show how many users are browsing from a certain city?

We discovered the core issue was that the Streams data is not visible to marketers when they most need it during the rule-building process.

With these insights, we wanted to define our purpose

Our solution should enable marketers to understand user behaviors to feel confident about the decisions they make to capture data, which they are unable to do so by the identified gap.

By synthesizing research findings into key insights, my mentor and I realized there was a need for marketers to understand user behaviors to feel confident about the rules they are building.

Marketers need to understand user behaviors to create personalized experiences

This user goal statement allowed me to ideate solutions and create sketches to represent them.

Guided by research insights and the primary user goal, my ideas and sketches addressed the need for powerful data visualizations, a marketer-friendly interface, and integration of Streams within the application for a seamless user experience.

I created numerous sketches that would address stakeholder and user needs.

I quickly learned how iterative this part was, as I improved upon my ideas with each round of feedback with stakeholders and other designers. After many, many sessions, we really started to narrow down our solution here.

The idea was to have data visualizations created with data from Streams integrated within the rule-building application (rough sketch below).

Moving forward with this idea, I moved on to create a higher fidelity prototype to present for one more round of feedback. I utilized key feedback points that I would implement into my final design:

1. Have the data preview integrated on the same page when building the rules

2. It would be helpful to see real-time data from Streams AND historical data. Depending on the situation, at times a marketer may need to see historical data to better predict the impact a certain behavior or filter can have.

3. Display relevant data. Streams requires users to write queries to drill down on data - how can we provide the data that marketers care the most about?

Finally, I was able to move on to create my final high-fidelity prototype that demonstrate my final solution.

The final solution consisted of implementing data visualizations showing real-time data and historical data to allow marketers to know the potential impact of the rules they are creating. Additionally, marketers are now able to see data on these rules and specific behaviors prior to even building rules with a data-rich dashboard implemented on the landing page. Marketers now have a connected experience between Streams and its application when building rules. They can make data-informed decisions quickly, helping them deliver personalized, digital marketing experiences.

What does Jane’s experience look like now?

Jane is excited about using the newly designed rule-building application and its’ integration of Streams.

When launching the rule-building application, Jane now can see a dashboard view of data visualizations, showing her what behaviors and scenarios are trending, and where live users are happening. She is able to see right away there is an increased number of customers viewing a new product, with almost 400 people viewing it in real time. She decides she will target these customers viewing that product and send them an exclusive discount code to their email.

When actually building out the rules, she can see live data with pulsating animations, showing that there is live data steaming in for the filters and behaviors she was adding. Additionally, she can see historical data for the past day by default for specific user behaviors. She sees that yesterday, there was about the same number of people viewing it, giving her the insight that the product is fairly popular. When she wants to see all live and historical data for behaviors and filters combined, she can click a preview button that allows her to see it all in one place. She quickly scans the data and knows that her discount code will reach hundreds of people and that she is targeting the correct product.

The problematic gap and user needs were addressed by utilizing Streams’ powerful real-time data and integrating directly into the software that used that data, with the addition of historical data and a dashboard view that allows marketers to zoom in and zoom out.

After presenting my final solution and prototype to the Oracle design leadership, my peers, and stakeholders, there was some key feedback received. Additionally, I wanted to also include what the next steps should be after my time at Oracle.

1. User testing with marketers.

   Time constraints did now allow direct user testing to happen. User testing is essential so that we can understand how users feel about the solution and their pain points, and find areas of improvement.

2. Further, refine the solution and create more iterations

   After talking to stakeholders, I would’ve evaluated the data we presented on the initial landing page. What types of data could we show that would be more helpful to marketers? Additionally, I would’ve had the preview of all the data stay on the screen consistently, enabling marketers to see the data update without having to click the preview button.

3. Update the Streams to match all the new components and templates of the Oracle design system

4. Integrate the solution across other Oracle products. The historical and live data visualizations could be powerful across other applications.

What’s next?

I will always be grateful for my time at Oracle. Having a background in web design, and little experience in UX Design, I was nervous about taking on this role. Additionally, working with enterprise applications and having to learn the ins and outs CXM and its products seemed daunting at times. However, the guidance of my mentor, my fellow interns, my team, and others, helped me push through when I had serious doubts.

• It’s always ok to ask questions (no matter how silly they may seem!). Learning something new can make doubts arise, but it’s important to take the initiative to push through by asking questions.

• Storytelling is an essential skill to learn. It helps your audience understand the landscape and your solution. I learned how to use a story to present my findings and solutions.

• The design process is always iterative. Being able to take in feedback, test, and apply findings is an iterative process that allows you to improve upon your idea or solution.

Reflection and Takeaways

The Solution

By integrating Streams’ real-time data and historical data visualizations into the existing rule-making application, marketers can better understand their users and the impact their rules can have.

Additionally, before even starting to build rules, marketers have access to a dashboard of data to understand how their previous rules and filter are performing.