Background

This product was built for Veritas to enter the Small Medium Business market by offering a hyperconverged data protection product that supports use cases from data center to the cloud.

Product Strategy: The legacy data protection products had some pain points - users had to build their own infrastructure, perform many manual operations, perform forklift upgrades and incur high running cost. Veritas Iris aimed to solve these by consolidating data protection into one form factor - Compute, Storage, Network, Application.

Competitive Landscape:

Competitive offerings in the marketplace, such as Cohesity and Rubrik, offered hyperconverged products. At first glance, their user experience seemed simple - clean design, lots of spacing to clearly understand information, colorful visualizations. However, on closer examination, we learnt it had significant problems. For example, in order to finish key flows, the user had to bounce around various different parts of the product on their own, with no guidance. The ideal solution here would be one seamless flow or at least a flow that guides the user through the entire experience.

User Personas

Pivot to Generalist: Veritas’ own flagship data protection (NetBackup) catered to large enterprises where multiple user personas perform various silo-ed operations - Backup Admins, Workload Admins, Security Admins, etc. Iris targeted a new kind of user for Veritas, “generalist”s who perform many diverse tasks using many diverse products and tools in a SMB. They want to ensure their data is protected, but don’t want to specialize in data protection or build that skillset. Hence, we had to design a user experience so simple and intuitive, that a generalized IT admin in an SMB could use it without any specialized training, efficiently finish their tasks, move on with their day, and finish all the other tasks they have to perform.

Major jobs to be done:

• Protect - I want to see and be able to search all the assets (physical, virtual, cloud) in my environment that are relevant to me and perform essential protection tasks, including backup my data, take snapshots, define and adjust protection policies, and recover data so that my data is adequately protected based on SLAs, is stored in the right place, and is available to the right people.​

• Recover - I want Instant Access (fully functional point-in-time copy) to my snapshotted VMs so that my RTOs are within seconds.​

• Monitor - I want to monitor and manage my appliances and my organization’s storage use so that I can plan for the future, understand how storage optimization capabilities have resulted in data reduction, understand the level of protection and vulnerabilities and when something goes down so that I can take action on unprotected data, failed backups, and assets to minimize business disruptions.​

• Security - I want to grant restricted and customized access to users based on roles, so that they can perform tasks that make sense to them and ensure their data remains secure.​

Requirements gathering and Concepting

For this project, I designed four features and the flows associated with these features.

• Reporting

• Licensing

• Active Directory Integration

• Disaster Recovery

Below, I will walk you though the design process for one of the features - Reporting. At the start of the project, I sat down with product management and hashed out requirements. We did a lot of whiteboarding, discussion, collaboration and competitive analysis. I used all this information to create some high level concepts and storyboards to validate with PM and Engineering before moving forward.

 Research Round 1 - Concept testing

After I created some concepts, the researcher conducted the first round of research on these concepts. We wanted to assess the landscape of our customers and get a better understanding of their environments, challenges, use cases, needs, etc. We also wanted to run our concepts by them and get their feedback.

5 users - VMWare Admin, Infrastructure Manager / IT Admin, Sr. Network Engineer, Sales Engineer (2)

Learnings:

• We learned what types of reports they are most interested in. For example, reports around job failures, capacity, connection, etc.

• What aspects of the concepts did they like, so that we would know what to keep. For example, Job by Status card, Top Job Failures. Download buttons,

• What aspects of the concepts did they not understand or appreciate at much so we would know to improve or drop these. For example, Policy details table, Jobs Completed Per Hour

• We got to hear what else they would add here or how they would improve these concepts. For example, from the Job by Status card, give them the ability to drill all the way down to see what host failed and why

 Research Round 2 - Usability Testing

With more fully flushed mockups, the researcher conducted a usability testing study to gather feedback from users

7 users - Lead Server Operations Engineer, System Engineer, Sr. App Developer, IT Director (2), System Admin (2)

Similar to the earlier round, we listened carefully for what to keep, remove and improve upon.

High level recommendations:

• Provide ability to set up e-mail alerts​

• Consider removing Data Ingestion from the Backup jobs Report​

• Provide ability to drill down into details using legends, charts, etc.​

• Re-evaluate Filters for clarity and usability – consider surfacing Filters on the page instead of buried in the panel​

• Consider changing terminology – protection and classification​

• Provide ability to set their own thresholds for the Storage Utilization warnings​

• Provide ability to view predictions on storage utilization​

 Other features

I also created the experience for Licensing, Active Directory integration and Disaster Recovery. Here are some artifacts from those flows.

Design Leadership

About midway through the project, I stepped into the position of Design Lead due to attrition. In addition to designing and owning the four features above, I was also the Design Lead for the entire project (36 features). I provided strategic and design direction to a team of 6 designers, 2 researchers and 6 technical writers to deliver this entire product release.

• Review designs, give feedback, inform design direction

• Manage designers’ timelines, act as an escalation point, remove obstacles to enable progress

• Coach designers on process, collaboration with cross functional peers, communication strategies, and tools to use for making the case for design

• Work closely with Product Management, Research and Infodev to inform the product direction

• Provide executive updates on behalf of CX (or just design)?

Created Fit and Finish process

In my earlier companies, engineers and designers worked closely together to review and revise engineering builds. However, at Veritas projects followed an inconsistent process after handoff to engineering. In most projects, once we handed it off to engineering, we did not get to see the designs until a very public demo with a large audience or upon release. It was too late to do anything about it.

I worked with engineering to create a fit and finish process whereby

1. Engineers provide frequent builds to designers.

2. Designers review and provide feedback to engineering via JIRA tickets or a confluence page.

3. Engineering informs the designer or updates the JIRA ticket once they’ve addressed the feedback.

4. Designer would review the build and either close the item or keep it open and work with engineering to resolve

It seems like a simple process but it became really powerful and collaborative when after a few weeks of doing this, engineers realized that the most efficient way of following this process is to frequently review the in-progress builds live with the designer via a meeting. This allowed the designer to offer feedback early and often, the developer could address the feedback on the spot and tweak according to designer feedback. This greatly minimized the back and forth over JIRA/confluence and saved everyone’s time.

Research Round 3 - Alpha with Sales Engineers

In this round of research, research team collected feedback from internal Veritas folks - Sales Engineers, Sales, etc. They organized a multi-day workshop and observed the participants installing the hardware (for some use cases) and use the Iris software. They collected some very valuable feedback. After we collected the feedback, a cross functional team (Design, Research, Engineering, Product Management, Sales Engineer, Partner reps) triaged it into Bugs, Feature Gaps, and Enhancements. We also worked together to assess the priority of these items. For all items for the Design team, I assigned it to the various designers and helped them brainstorm ideas to address as needed.

Research Round 4 - Beta with Customers

The research team also conducted a Beta program with Customers such as DataEndure to get feedback on the product, customer testimonals, as well as offer an educational and pre-sale opportunities. Again, as the feedback came in, the cross functional team (Design, Research, Engineering, Product Management, Sales Engineer, Partner reps) triaged it into Bugs, Feature Gaps, and Enhancements. We also worked together to assess the priority of these items. For all items for the Design team, I assigned it to the various designers and helped them brainstorm ideas to address as needed.