On Monday, April 29, 2013, OLC attended NYC Web Design Meetup’s event, An Introduction to Lean UX featuring Jeff Gothelf and Josh Seiden, Managing Directors of Neo and co-authors of Lean UX: Applying Lean Principles to Improve User Experience.
Josh Seiden introduced himself and Jeff Gothelf as two founding partners at Neo. They wrote a book together, titled, Lean UX: Applying Lean Principles to Improve User Experience. He wanted to set the stage for context that is important to consider. “Software is continuous,” he said. “Think for a moment about this concept. I started in the ‘90s as product manager. I worked with designers and engineers. We adopted a model from the print model. Similarly, the engineering model is like a business model. We have to figure out what we’re doing, or people die [or companies fail].”
“The new reality is that software is distributed through the web now. It’s using continuous employment. Amazon deploys new software every 11.6 seconds. Software is basically consuming the world now,” Seiden said, naming book publishing and the music industry as examples. He went on to say that the automobile industry was becoming more software-centric, as third-party mechanics couldn’t repair cars without the software diagnostic tools that only dealerships have. “General Electric pledged $1 billion in software COE [Common Operating Environment]. This is because we’re all in the software business. GE recognizes that. The very nature of software is changing the way we work,” Seiden said.
He explained continuous methods, which he put as “no new models, just one big continuous system. Instead of lots of little things, it’s one continuous thing.” Seiden said that the only chance we have is to build it, ship it and figure it out if it works. “This is a flip from the original model, where we figure the product out first, built it and then ship it for use.” Seiden briefly mentioned the Agile Manifesto as continuously driven. “The Agile Maniesto was developed, but it is painful to work with it. Designers struggle until they adopt Agile UX. And when they do, business people struggle, so their response is lean startups. It’s basically an iterative design cycle,” he said.
The Lean UX cycle is when you state a desired outcome, declare assumptions, hypothesize, design the experiment, build the Minimum Viable Product, ship it, learn from it and do it all over again.
After Seiden explained this, Jeff Gothelf took the helm and explained the details of the Lean UX cycle.
“Tactics to use in Lean UX range from selecting your team to learning from failure. First, you need to have a small team to work your ideas. There are certain qualities that make the transition to Lean UX easier. The ‘Two Pizza Team’ is one of them. The point is to keep a team small, so you can get a clear sense of what people are doing. You have better conversations that help the building process,” Gothelf said. Small teams, at least, according to Gothelf, are dedicated to projects, need to be co-located and they need to be cross-functional—in other words, small teams need to be self-sufficient.
On infrastructure, Gothelf explained that it is difficult to integrate with legacy technology. "There's a lot of work to be done in organizations for your team to run efficiently," he said. An example he used was on General Electric spending millions of dollars on a front-end system. "Large companies are spending money in infrastructure so that design and production teams work on the same project," he said. “Instead, build a structure that supports rapid iteration. The goal is to minimize the transition from idea to prototype.”
“The biggest challenge in Lean UX is to focus on outcomes. By changing the conversation from output to outcomes. This way, teams are empowered to hold rapid iteration cycles to see if ideas hold water or not,” Gothelf said. Because output leads to impact and many companies focus on output, the team assigned to that task might not have any value to the company. To focus on the outcome, the team assigned to the task is able to figure out what features are necessary for the product to succeed in rapid iteration.
And to hit the intended outcomes, small teams need freedom to iterate.
Gothelf warned the audience that if they were to gather objective data and the data doesn’t move the needle in a direction that they need to go, they should immediately “kill the idea and move on to something else.” However, if they felt like they were on to something, then they should pivot. “When you finally hit on something that’s working well, double-down and scale,” he said. To reiterate, Gothelf said to run small, low-risk experiments often, refine the hypothesis, reward teams on learning from failures and to make decisions based on objective observations.
Regarding continuous making and learning, Gothelf said to push iterations constantly to get feedback from the market. “It’s about capturing feedback and reacting accordingly,” he said. "Make a design decision and get feedback from the market. The market is the judge—you can't control it. But you can control how long we spend on the product. It can be for months to just hours.” The point of Lean UX is to reduce inventory, risk and waste. It’s to reduce time wasting. A system built on continuous learning is taking small chunks of data and testing it. “So to recap, instrument your software to measure usage, built a process that includes the customer and combine the what, which is quantitative, with the why, which is qualitative, for a holistic view of your project.”
From here, the floor opened to questions from the audience.
“What do you have to say about protecting brand image when going through repeated failures?” an audience member asked.
“It’s imperative to protect the brand,” Gothelf answered. “One way is to not use the brand name at all on the website. Another is, although it’s not proven, to use the marketplace as a pricing experiment. Just be creative to protect the brand.”
“The Nordstrom Innovation Lab tries new concepts in-store. They make it very clear that they’re trying new experiments and it makes them look very innovative. But on protecting the brand name—it’s a case-by-case thing,” Seiden answered.
“How do you deal with uncertainty?” a member asked.
“We do a lot of work with enterprise clients that don’t have consumer-facing products. The Lean UX method is great at experimenting towards a goal. There’s a degree to which you’re dealing with uncertainty. Are you trying to figure out what you’re doing? You need to be wary of stuff like that,” Seiden said.
“What are the skills necessary for the ‘Two Pizza Teams’?” a member asked.
“For us, they need to be competent. The issue is that if one person is unable to carry their weight, it’s going to bring the team down. A competent team means a cohesive team,” Gothelf said.