The Legacy of Waterfall Product Development
The Waterfall product development process is an artifact of the 19th Century Industrial Revolution. It is a sequential design process that mimics the automobile assembly line. Each phase, Conception, Initiation, Analysis, Design, Construction, Testing, Production/ Implementation and Maintenance, flows steadily in a linear progression. It is also heavily dependent on voluminous deliverables, such as requirements and design documents. It assumes that the user is involved in a meaningful way during requirements collection.
But, since the user may not see what will be delivered until it’s almost finished, there is a very real possibility that the user will be dissatisfied with the final product. There is no provision in the Waterfall process to change direction once the process has begun. This can result in costly applications being introduced to users with little to no acceptance.
There are also lengthy development timelines associated with Waterfall. In many organizations, applications take multiple years to be completed. This may be acceptable for database framework or technology transformation projects, but severely damaging to the reputation and profitability of a digital services company. If a company publishes software, it is, at its core, a digital services company.
The Value of Lean UX for Product Development
Lean UX is not just a mindset or a process, but also a management method. For this reason, there are organizational changes required to get the most benefit out of implementing a Lean UX development process. These will both be empowering and threatening. Lean UX makes it possible to use the power of software and proven methodologies to create a continuous improvement loop that allows an organization to stay ahead of its competitors. It’s this loop that drives organizational agility and allows the company to react to changes in the market at speeds never possible or conceivable.
Comedian Woody Allen is quoted as saying: “If you’re not failing every now and again, it’s a sign you’re not doing anything very innovative.” Releasing early and often, gaining market feedback, and iterating based on a continuous conversation with customers provides two significant competitive advantages: 1) The ability to learn, continuously and quickly, how well products are meeting customer needs, and 2) Raising customer expectations in terms of product quality and company response times to their concerns and feedback. Any outcome short of these will have severe competitive implications for products in the marketplace for years to come.
Implementing Lean UX for Product Development
The shift to an Agile product development process promises to shorten the time-to-market and improve user experience significantly, but only if three fundamental questions can be answered at the start of the project:
- Does the user exist?
- Does the user have a real problem that we can solve?
- Does the user value a solution to that problem?
While we may successfully implement a solution within our planned schedule and budget, if the customer is unable or ends up choosing not to use the solution, then, from the customer perspective, we did not solve their problem in a timely fashion. We need to improve our capability to create the right solutions.
There are multiple factors that create challenges in this area. All these areas are attributes that can occur at times, inconsistently and intermittently. However, all are areas that, if improved consistently, would contribute to better defined problems and more effectively solved industry/customer problems.
- When customers request work, many communicate a solution rather than a problem.
- Most customers are busy and do not focus on evaluating our proposed solution as something they would actually be able to use.
- We have a varied customer base. They vary in their size, sophistication, and strategy. Sometimes there are too many compromises on a solution that it ends up meeting no one’s needs.
- Within the organization, we need to communicate with multiple departments. We don’t always have the correct person/people involved in solving the problem.
- Within the organization, we don’t always have the decision maker who ultimately decides whether a solution, even if it is the right one, would be used, especially if there are costs involved.
- At times, management asks leading questions. We also sometimes focus on validating that there is nothing wrong with the solution rather than is it truly solving the problem. We ‘convince’ the customer rather than listen and observe. We think we know what the customer wants and we don’t turn off our filters. We make assumptions.
- Our Milestone process is focused more on stakeholders than the users that made the initial request.
- In the past, organizations would automate processes and standards that existed. Today, we find ourselves in less clear times where identifying the problem is more difficult and finding the solution more challenging because there is more to uncover and understand.
- We seem convinced that we have a complete understanding of our customers versus working with them to get a clearer understanding of their needs.
- We don’t design products with the customers. We tend to design them on our own.
- Once we think we know what they need, we develop it without always understanding how it will fit. We need to understand their workflows better and how they interact with our and other systems, data, and information.
Here are some strategies to overcome these challenges. Some are more specific and easier to implement than others.
- Create a series of questions on a check sheet, like the Lean UX Canvas, as part of the new business request vetting process that must be answered for each request to define the problem. Don’t move forward without that being reviewed as part of the Discovery process. Validate the assumptions and answers with the requestor.
- Working groups have gotten quite large and there are very limiting budgets for many. Create subgroups within the working group to work more closely and frequently to work thru the solution and then present to the larger working group. In the smaller group use real examples and workflows to walk thru the solution to validate its merit with the customers.
- Clearly define the target markets and the value to each. If the value is different for each target market, select the target market for the solution and make sure vetting is done with that group. Understand the key differentiating needs of the non-selected target markets. Make sure communicate to non-selected target markets, the plan and solution to set expectations.
- Figure out how to better investigate the customers’ full workflows, ask the right questions, and uncover assumptions.
- Determine who the decision maker is from those organizations participating in solving the problem. Validate the use of the solution in their organization. Select several non-involved customers in the target market.
- Add to concept document ‘what success looks like’, e.g., how many need to adopt solution. Include more detail on how adoption will be achieved in the concept document.
- Collaborate more closely and frequently as a solution is being designed. Provide tools that those with frequent customer contact can use to help vet problems, solutions, and help us get to know the customer.
- Get some training, whether formalized or in books, to arm our team with better tools to get to the root problems that need solving as well as the implications of those problems. Need to improve critical thinking skills.
- Streamline the milestone process to make it easier to provide responses.
- Update the milestone questions to ensure we have the right person answering, right questions being asked and make sure we are getting responses from our target market
- Utilize the UX team and their processes to be more objective and less leading.
- When prioritizing with the industry, don’t move forward without an advocate working with us to design and validate the solution—follow user centered design principles.
- Along with IT, review our planning and execution process so that we don’t end up in the situation where IT is ready with coders and the solution isn’t fully vetted. We are bringing IT into the process earlier, which is a good thing, but these earlier stages need to be more senior level with only a few less experienced for learning.
- Have more 1:1 or small group vetting at our customer events, including Global Conference, Product and Services, working groups, and roadshows.
- Find communication tools that have regular customer 1:1 contact, e.g., marketing directors, consulting, training, to get feedback from customers.
More specifically, there are five guidelines for implementing Lean UX in an Agile or Waterfall development process:
Build a team of collaborative, cross-functional people, including industry stakeholders.
Work with your organization to build a team of people who are naturally curious about other disciplines, work well with others, and are flexible. The team should include everyone needed to get the project done – product owner, developers, analysts, quality assurance, UX specialists, designers, etc. Include industry stakeholders, both supervisory and hands-on users.
Agile is, by nature, a highly collaborative approach to project execution, and therefore requires team members who enjoy offering up their specialized insight into all aspects of the project work. Cross-functional teams provide multidisciplinary perspectives on all project ideas, which lead to better design solutions.
Bring UX researchers and designers onto the team from the very beginning of the project.
Be sure to bring UX specialists and designers onto your team from the very beginning so they can help establish and test the big picture vision for the product, and help keep the team’s eye on this vision as the project unfolds. Bringing UX specialists onto your team early helps the team work together in setting up work processes that maximize collaboration and efficiency between the testers, designers, and developers. In addition, it gives the UX team members time to establish their own testing/design protocols and develop study participant lists, which is difficult to do in the thick of sprinting. (In the Agile world, a “sprint” is a set period of time during which specific work must be completed and made ready for review, usually a matter of a few weeks.)
Take a collaborative, not a siloed, approach to doing the work.
While team member specialists are still responsible for delivering their pieces of the work, the difference comes in how Agile teams approach the work. Teams that successfully integrate UX into their process do so by involving all team members in the usability studies and design work, and by iterating the design based on solutions agreed on by the team. UX specialists and designers, in turn, work hand in hand with analysts and developers as they build the design into a working product.
This type of close collaboration helps foster a shared vision for the product, build trust, and increase team efficiency. In addition, it helps ensure that the design is constantly accounting for the realities of development, while development is always informed by the user needs.
As Jeff Gothelf, in his book, Lean UX: Designing Great Products with Agile Teams, points out, “Permission to fail breeds a culture of experimentation. Experimentation breeds creativity. Creativity, in turn, yields innovative solutions.” Because there is no one-size-fits-all Agile methodology, it’s important to make sure the team has the freedom to experiment with their processes, deliverables, and sprint rhythm. Establish a culture where taking risks and progressive elaboration of team processes and deliverables is encouraged – after all, you’re unlikely to get it right the first time, but if you experiment, eventually you will.
Make sure user experience activities are focused and actionable.
Working on Agile time scales necessitates modifying the timing and granularity of UX activities that are performed. For example, your team may need to:
- Streamline study planning processes.
- Develop smaller-scale studies that focus on a few key design elements.
- Test on low-fidelity mock-ups instead of polished prototypes.
- Use methodologies, such as the Rapid Iterative Testing and Evaluation (RITE) method, to test and iterate designs more quickly.
- Supplement formal usability testing with quick-turnaround online tests, or through informal tests using in-house participants (especially when testing early designs).
- Think about faster ways to recruit members of your target audience to participate in more formal studies (for example, creating a bank of participants available for studies).
To be sure the UX activities meet your team’s informational needs, all team members must be involved in articulating the study questions for each round of UX testing. Once the results are in, the whole team should go through the findings and come up with realistic, actionable solutions to the issues identified.
Findings should be succinct and prioritized by the team so everyone buys into the findings and knows which problems to focus on first and which should go into the backlog. In Janet Six’s interview with Carol Barnum, an award-winning UX author and speaker, Carol describes her team’s process of debriefing together at the end of each study day to “agree on the findings and prioritize the issues the development team needs to fix. The developers, when present, walk out with their list of issues to address in the next or a future sprint.” If done properly, light-weight reporting is by no means synonymous with lack of quality; it is simply a different, and often more effective, way to communicate your results.
Gothelf, Jeff and Josh Seiden. Sense and Respond: How Successful Organizations Listen to Customers and Create New Products Continuously. Harvard Business Review Press (February 7, 2017)
Gothelf, Jeff and Josh Seiden. Lean UX: Designing Great Products with Agile Teams, 2nd Edition. O’Reilly Media; 2 edition (October 20, 2016)