A sneak peek at the future of data migration: The DNA of Bluemix Lift


What does dedication to the client experience look like? To create Bluemix Lift, it meant driving from concept to app in 90 days flat. Through rapid iteration, remote collaboration and a focus on user-centered outcomes, one team worked to make data migration delightful. I sat down with Joe Meersman, cloud data sevices design director, Ty Tyner, cloud data services experience design manager, and Miran Badzak, IBM Bluemix Lift product manager, to learn how.

Migrating data is no fun. If you’ve moved homes, you get it: remember sorting through boxes, organizing, packaging, shipping? No one wants to deal with that. In fact, that’s why moving services exist.

Bluemix Lift is one such service—except instead of moving furniture from city to city, it moves secure customer data from on-premises to the cloud. 

Home page for IBM DataWorks Lift

This new service in the IBM Analytics portfolio taps into a large market; every modernizing company needs to move data. To sweeten the deal, existing data management client relationships make this a relatively strategic market. “Data under management is very sticky,” explained Joe. “If you store your data with me, you won’t go to someone else to move it. You’ll come to me, because you trust me, and because I already know your systems.”

Unfortunately, that logic wasn’t playing out. We were losing on-premises clients to competitors like Amazon Web Services. Why? Because user experience matters. The service required people to cobble together several tools, use multiple interfaces, and reference extensive documentation to make a move happen. “Our clients wanted an easy path to the cloud. To meet that demand, we had to create a better end-to-end user journey through our User Interface,” explained Ty.

With numbers declining, Girish Venkatachaliah, VP of Development, and John Murphy, director of offering management, sat down to talk.  They agreed that a new experience was necessary—and soon. By the end of the meeting, they’d set a beta launch date.

 “This experience was a great example of commander’s intent. The mandate was set at top—with a tight three month timetable—and we delivered on it,” said Joe.

But how? Three months is mighty fast, I exclaimed. 

“IBM Design Thinking!” Joe laughed. “It was the only way. The whole thing was done at such an aggressive velocity that we had to rapidly iterate and stay in lockstep across disciplines. IBM Design Thinking made that possible.”

The team leveraged IBM Design Thinking as a framework, not a process. Ty explained: “If the process was ‘ramp up for 6 weeks to Playback 0,’ we didn’t do that. What we did was constant iteration. Instead of making a perfect prototype to show in two weeks, we had an ongoing show and tell.” The team made prototypes at every stage in the process with various levels of fidelity. 

DataWorks Lift Concept Sketches. Photo by Russell Parrish

While designers worked on the user experience and visual designs, engineers worked on the proof of concept. “Because we’re constantly in contact, you can never get too far off track,” Ty said. “Every day you know where you’re going.”  Joe summed it up: “If you don’t let distance creep in, you don’t have to learn how to shout at each other to be heard.”

At the heart of their success was radical collaboration. “We embraced failing fast,” Ty explained. “There was no ego. Just: ‘go, go, go!’”  

I could tell from Ty’s tone that true rapport exists among the design, development, and offering management collaborators on this team. Smiling, Ty described his comrades: “they’re fantastic: very cloud-based, forward-thinking, and start-up like; we’re always in the trenches together.”  What facilitated those relationships? In addition to a kick-off workshop, the team maintained constant remote collaboration via Slack and multiple daily meetings by phone. “It felt like our developers might as well have been in the room,” Joe explained.

Collaboration extended to the inclusion of sponsor users. Despite the tight timeline, the team brought in internal users from IBM Lab Services to validate their work and get feedback. “They’ve performed customer database migrations for years,” said Joe. “Their insights really helped us understand the problem and refine our solution.”

With feedback from internal stakeholders and select IBM clients, we set out to create a fully-managed service on Bluemix,” explained Bluemix Lift Product Manager Miran Badzak. “We listened closely and a pattern emerged: speed, zero-downtime, security and ease of use.” When the service internal-launched, IBM Lab Services started using it and offering feedback to further refine the service. That feedback has now been integrated, laying a foundation for today’s public launch. 

Soon, the Bluemix Lift team will get to test their hard work with their most crucial users: real clients. I’d suggest we wish them luck, but I don’t think they need it. By testing their designs, collaborating effectively, and integrating user feedback, they’ve made their own luck.

 Curious to learn more about Bluemix Lift? Visit this page to get more information.

Follow @IBMBigData

This entry was posted in Big Data. Bookmark the permalink.