Using a discovery process to reduce project risk — a 3-phase path to success

At Truefit, we believe that great software helps people thrive. To ensure that, we begin each project with a period of discovery before we begin building features. Why? Because the discovery process efficiently delivers insights that greatly reduce risk. We test assumptions early and guide the team to smarter decisions at a time when it is less costly to pivot.

An effective discovery process does more than just ramp up the team's understanding of the business domain and product vision. An effective discovery process...

  • grounds the team in empathy for users
  • provides insights that lead to distinctive value in the market
  • aligns all stakeholders around business and operational challenges that can impact a successful launch
  • identifies clear metrics for success and the design principles to guide effective, focused decisions throughout the development process
  • proves the feasibility of the riskiest technical assumptions and uncovers the true scope of implementation
  • provides early validated proof that the underlying hypotheses driving feature selection and priority are sound — allowing the team to make critical product planning decisions with confidence

At Truefit, the discovery process is broken down into three distinct phases. These can be tailored to accommodate the unique needs of each project while consistently producing effective results. Every project has its own unique risks and demands. After successfully executing hundreds of discoveries over the years for a diverse range of new software products, we’ve found this model to be consistently flexible, allowing teams some latitude to chart their own course.

Truefit Discovery

Phase 1: Tackle the problem space before diving into the solution space

The Understand + Frame phase of the process grounds the team in a clear understanding of what the product needs to do and the problems it needs to solve.

A primary goal of the discovery process is to identify the likely users of the product and effectively build empathy for their present reality. We set out to identify their goals, behaviors, pain points, and underserved needs. We seek opportunities that create value — engaging and retaining users in ways that create competitive advantage.

We interview users to understand—and, when possible, observe—how they perform a given task (e.g., select a primary care physician). When analyzing the insights, we note evidence of each person’s underlying behaviors, motivations, and pains they are trying to resolve at each step of their journey. As we synthesize the insights drawn from multiple interviews, we identify central themes and patterns that tell a larger story.

At the same time, we set out to understand and prioritize the business objectives of each stakeholder with a vested interest in the product (e.g., customer support, sales, outside partners, etc.) This helps the team identify business rules and measurable results that the software will be expected to achieve in order to be considered a success. These interviews also provide valuable insight for the teams’ software architect regarding analytics, scalability, and security.

Informed by these user, market, and business insights, we explore the emerging product vision as a team. We draft storyboards that tell a story about how the proposed product features can potentially address the key user problem(s) we identified. If we need to validate the desirability for these features, we use a survey-based approach (we prefer the Kano Model), that helps us efficiently confirm user satisfaction and priorities.

With these valuable insights, the team can succinctly frame the crucial user problems and identify guiding principles for resolution at an architectural, informational, and interface level.

As an example, we recently worked with an exciting spinout from Carnegie Mellon — DorothyAI. They came to Truefit with a student-led prototype that enabled patent attorneys to better search the U.S. Patent Database using machine learning and natural language processing techniques. However, they needed a superior user experience for patent attorneys to adopt the platform over competing platforms. While confident their technology delivered a distinctive value, they had yet to resolve how attorneys would use this product to support their existing workflows. To achieve this, Truefit led DorothyAI through the discovery process.

“We came to Truefit essentially to get a cleaned-up version of the student project,” explained Curtis Wadsworth, founder of DorothyAI. “What we got was so much more—the discovery process was a unique learning opportunity for us. The user research alone showed us that not only was there a need, the users gave us a wishlist of features that resulted in a product that was very different and a lot more robust.”

Phase 2: Generate and prototype solutions to the riskiest high-priority features

Generate and Prototype

Now that the team is confident about what the product should do for users and why, the next challenge is to determine how the product should work. The Generate + Prototype phase is where the team explores optimal user interface and technology solutions for the highest risk features of the product. We do this before moving into development where it will be far more expensive to pivot.

First, the team comes together to identify the core objects that the user will encounter in the software (e.g., a travel app may include objects like an itinerary, a packing list, a destination map). The team identifies content attributes, calls to action, and nested relationships between objects. This results in an object-oriented UX model that both the designers and engineers can refer to in their individual work moving forward. For a nice overview of how to do this, see here).

Using this model, we generate a diverse range of solutions to each user problem in risk-priority order. We use a “design studio” format that provides the team with a forum to contribute their expertise and openly discuss the advantages and disadvantages of each proposed solution. Together, the team refines and resolves their vision.

From this common understanding, the designers and engineers can work with confidence to resolve high-level information architecture, user flow, and user interface style or interaction details. The end goal of this work is to build a click-thru prototype capable of simulating key user experience details. At this point, the product really starts to take shape and stakeholders begin to get excited.

The prototyping process also illuminates unidentified scope that will help the team more accurately estimate the product backlog before starting the build. In some cases, a coded proof-of-concept is needed to prove feasibility and gain vital insight about usability, development effort, or performance risks.

Since DorothyAI already had their core underlying search algorithms in place, we needed to focus on an engaging front-end user experience that could work across mobile and desktop devices. This required early prototyping using a click-thru prototype to simulate user experience scenarios as well as a coded proof-of-concept using the client’s APIs to test search-processing time and data results.

“After getting educated on what the users wanted out of Dorothy, I had a whole new perspective on how the product needed to work,” said Curtis when the team began walking through the user experience portion of the discovery process. “The early prototype that Truefit created was key in aligning our team around the emerging vision of our service and brand.”

Phase 3: Test prototypes with users to inform difficult decisions in release planning

In the Validate + Plan phase, we test the prototype with actual users who can validate the likelihood of user satisfaction. They can also illuminate any potential blind spots before we go into planning, and which features to build for the first release of the product.

We use lean, qualitative research methods to efficiently validate that we’re on the right path. The resulting insights not only sharpen the design, but provide a vital objective touchpoint for our clients when making difficult decisions about which features to include in the MVP release. The backlog for the first release of the product can now be resolved with confidence. Each included feature solves a high-priority problem for users in a way that will be both satisfying and aligned to your business objectives.

The team is now aligned around a focused product vision and validated release plan that will help them better maintain their velocity in the build sprints.

After testing the DorothyAI prototype, the interviews revealed insights to valuable features they had not considered. A few of these new concepts were prioritized in release planning, while some of the original features were pushed for later release.

“If I had gone to market with the product I started with, I guarantee it would have failed right after launch,” explained Curtis. “The before-and-after of my experience with Truefit proved that success is in the process.”

Get your project off to a successful start

There is an adage, “The harder it is to change your mind later, the more time you should spend making a decision.” In any software development project, the discovery process is a vital way to demonstrably reduce risk and increase the likelihood that the product will satisfy your target users. It requires a skilled team, capable of cross-disciplinary collaboration and grounded in a proven process. When both are present, it is possible to deliver products that can succeed in the marketplace and help people to thrive, from stakeholders to end-users.

Truefit provides skilled, experienced teams with the know-how to build impactful software that engages users and grows your business. Please contact us today using the form below and start a conversation about how we can help you reach your goals.