How to Put Continuous Learning Back in the Backlog

“We all need to be agile. But it’s simply not enough to push the process forward. It’s not enough to just move from iteration to iteration. Between each iteration lives a fertile gap where innovation grows – that is, if you have the right type of information. Make sure you’re getting both iteration and innovation.”


In his recent post, “Are You About to Make a $100 Million Mistake?” Filter Experience Design Director Michael Hinnant describes how working from the wrong sorts of insights can lead your product teams to solve for the wrong problems – and lead your product far off the path to success.

This risk is easily magnified, to a potentially significant degree, when you are working in a continuous delivery environment, or have invested in DevOps or implemented other extensions of agile development methodology.

In many cases, this happens because companies focus so much attention on greater output, and very little on better outcomes. They may be highly adept at delivering code at an ever-faster pace, but they are not nearly as effective at identifying valid user needs while doing so. They build rapidly and efficiently — they just build the wrong thing.

I would assert that organizations need to direct these investments not only toward engineering, but also toward building out their user experience resources. After all, the market forces driving brands towards DevOps, continuous delivery (CD) and agile methodologies are ultimately user-centric, not engineering-centric: all these practices, processes and systems are deployed in the service of delivering more value to user (to in turn boost user engagement, sales, loyalty, and other outcomes of a better user experience).

UX, then, must not be viewed as merely a secondary supporting function. And in the continuous delivery enterprise, traditional UX practices cannot simply be bolted on to this new development methodology.

They must be designed for and prioritized within the rapid iterative processes of continuous delivery. They must be baked into each cycle, and into the mindset of product teams and the culture of the organization.

 

Where to start

One key place to put this into practice is the backlog, a core component of agile development and continuous delivery. Remember, continuous delivery is about continuous learning; when addressing your backlog, the fundamental question isn’t “what do we need to build?” It’s “what do we need to learn?” The days of far-ranging, 500-page requirements documents are over, and so are traditional UX practices built on extensive requirements capture to ensure that deliverables are as detailed as possible at the start of the project. Today, the notion that a product backlog will paint a complete and comprehensive solution well before the next sprint begins is antithetical to the practice of continuous delivery.

At the same time, development needs quality UX specs and requirements to succeed in delivering value. Among CD-related practices, the ability to capture user requirements was the number one priority identified for improvement in a survey of more than 3,500 software professionals by Outsystems for its 2018 report on The State of Application Development.

What this means, then, is that product managers need to deliberately work more UX discovery and design cycles into their continuous delivery backlog. And not just with lip service: at both a philosophical and operational level, the team and the organization need to make sufficient room in the process for continuous learning, making sure not penalize this work. This means getting comfortable with the fact some iterations may drop down in the schedule from time to time to accommodate more robust learning activities.

Jeff Gothelf, the author of Lean UX: Designing Great Products with Agile Teams, offers some specific recommendations on how to make this happen. He highlights the periodic use of “experiment stories” within your backlog just like any other story—it goes on your backlog and gets prioritized against your user stories. The experiment story is staffed and estimated by your team just like a user story, and holds two key directives: what you need to learn (what is the hypothesis), and how you’re going to learn it (the tactic: A/B tests, customer interviews, etc.).

The key is to plan ahead, determine how to prioritize it, and decide what stories fall below it because they may be fundamentally changed by the learning from this experiment. Again, a feature may fall out of the sprint to make room for the experiment story, which may seem anathema to teams focused on speed. But from a product-value perspective this isn’t a loss, because the point of an experiment story is that your team didn’t have enough information at the beginning of the sprint to properly inform the iteration in the first place. The key for product managers is to be honest and transparent with yourself about what you don’t know—about what your team needs to learn to deliver real value to your product and your customers.

Once the experiment story is complete, bring the learnings to the full team and other stakeholders; capture the findings in your wiki, follow up with appropriate specs and UX design artifacts, and assess the impact on your backlog—up to and including re-prioritizing or writing new experiment stories. Leaving room for this kind of learning builds a better a better platform for shared understanding into your continuous delivery process. By being transparent, identifying your knowledge gaps, making room for learning and incorporating this into your subsequent sprints, you transform the culture over time from an emphasis on output to a focus on outcomes. The process of managing and prioritizing against your backlog becomes more and more efficient and effective, which in turn drives more value and helps you deliver better products.

Highly experienced UX specialists, adept at agile methodologies, play a key role in maximizing the value from this process. Beyond the ongoing design collaborations in any given cycle and the ongoing fixes listed in the backlog, they bring a level of expertise and insight that helps product managers address some of the most critical decisions when determining priorities: What scope of advancements are needed to distinguish the product from its competitors? Where does novelty and innovation matter the most, and where is it acceptable to be on par? How do backlog prioritizations intersect with the interests of other stakeholders like analysts and market influencers? How does UX create a clear and compelling brand identity that also accommodates dynamic business imperatives? And how can a product team process all of this into an incremental roadmap that remains agile and sets a strong vision?


Filter is helping product owners, teams and companies optimize UX practices for the rapid pace of continuous delivery, DevOps, and agile development. Discover how we can drive greater success for your business: click here to contact Filter and let us know more about your UX needs.