How to Apply a Data-Driven Mindset in an Agile Framework
By Dylan Hebb
Web development is currently going through a revolution at the hands of data science.
As many development teams work within the framework of the Agile methodology, teams are scrambling to integrate data science into their framework. With so many core principles of the Agile methodology present, teams struggle to understand where it fits.
First, let’s review some of the most important principles of Agile. Through attending several training seminars and fluency reviews from industry leaders, the most important takeaways have been:
- People over processes and tools
- Collaboration over negotiation
- Responding to change over sticking to a plan
- Continuous improvement through iterations
In essence, this means that you build teams of dynamic people to collaborate and adapt to change on a given project to produce the best product possible. These values are known widely but rarely executed fluently.
To continuously improve means to seek excellence through trying new things, learning new techniques, and taking calculated risks as a team. For some of our teams in recent years at Content Bloom, this has meant taking a data-driven approach to web functionality, business processes, and user experience changes on Agile projects.
Why Should Agile Projects Be Data-Driven?
Prioritization of the backlog will always be driven by the perceived business value of a backlog item.
What better way to help guide business value than leveraging data?
When product owners, business stakeholders, and development teams get together to determine the work to be completed in upcoming months, teams will utilize a number of different prioritization techniques. Sometimes labeling items in the backlog as must-haves, should-haves, could-haves, and can-waits.
Data can greatly impact these decisions by providing full context of backlog items to differentiate the must-haves from can-waits. Not only that, but how should the user experience be built? How the customer wants it. There is no better way to get a full understanding of your customer than to analyze how they interact with your website.
In this article, I’m going to review some strategies of incorporating analytics into your Agile process to help you make informed decisions and, as a result, to create better end-products.
Now, let’s unpack this.
1. Incorporate Data into Your “Definition of Ready”
The first step in incorporating data into your daily Agile practice is including it into your definition of ready and/or definition of done. Each scrum team has its own team-agreed upon definition of ready and definition of done, which are the minimally required components of a user story before it is placed into a sprint or marked complete.
This could include rules such as:
- Shared team-understanding of the story
- Acceptance criteria exists
- Story is estimated
- Story is sprintable
Now, to include data into the definition of ready, you:
- Identify how to measure the story (if measurable)
- Identify how often the story will be measured
This should be discussed during backlog refinement. To begin integrating data into your team’s daily Agile practices, start small; identify a few stories that you’d like to measure, per sprint.
Use analytics tools (such as Adobe Analytics, Google Analytics, Heat maps, etc.) to collect and share the data with the team.
Sit back, and watch.
2. Create a Culture of Data-driven Decisions
When reinforcing culture in any context, it begins with a team-shared knowledge and understanding of the vision, goals, and motivations.
It is not enough to embed analytics at the root of a business, it needs to be integrated across all structures to be effective, including on the development team.
The success of an analytics program is regularly measured by your team’s ability to develop a well-rounded view of your customers.
At a high level, here are some common motivations of integrating analytics into your practices to focus on, to begin to form that view:
- Quality-driven: Identifying bottlenecks and friction to improve processes
- Value-driven: Informed decision making based on user interaction
- Predictability: What do our customers want in the future?
By identifying and analyzing these three themes, your team will gain a competitive advantage in providing true value back to the business and your customers.
3. Data Needs to be Self-Serve
Large organizations are beginning to really lean into the self-serve model. Meaning an internal or external customer can go through the full cycle of acquiring access, upgrading their accounts, getting training, or cancelling their account without ever requiring another person.
This is due to companies becoming increasingly interested in empowering users and internal teams to make the best of their resources by reducing bottlenecks, increasing automation and removing as many external dependencies as possible.
By adopting self-serve in the context of internal analytics or business intelligence tools, everyone on the development team needs to be able to pull data themselves. By doing this, you empower every Agile team member to get access, learn the tools, and leverage data for any user story, any idea, at any time.
If you plan to adopt data-driven insights into your development teams across your organization, self-service is increasingly not becoming a choice.
The Agile framework has been described as simple, but not easy.
Teams spend numerous years to reach Agile fluency. To add layers of complexity such as data-driven user stories and decisions, it can seem overwhelming at face value.
But – it doesn’t have to be. Data should make your life easier.
Less hypothesizing on what the customer wants, more collaborating on what the data shows.
The end-result of incorporating data into your prioritization and development discussions is the delivery of customer-focused quality, value, and predictability.
If you’re an Agile team that currently does not leverage data and you’re considering integrating these methods across your organization and development teams, keep these guiding principles and techniques in mind. I encourage you to try these practices, adapt and modify them to your own team’s needs to deliver maximum value, as any Agile framework would approve.