Modern Backlog Grooming Demystified

by July 16, 2018

Visual Requirements Modeling is Key to Modern Backlog Grooming

If you’ve been paying attention to the new methodologies around devops and agile testing, chances are you’ve wondered (at least once) how to go about doing modern backlog grooming. You might have even heard about model-based-testing or visual requirements modeling before. But why would you consider doing modern backlog grooming or testing for that matter in this way to begin with? Perhaps you’ve created flows in the past with tools like Microsoft Visio. Some common uses for flow-charting tools are creating project timelines, customer service talk tracks, or pulling information in from outside sources. But does this way of thinking translate to achieving a more modern backlog grooming approach or more broadly, streamlining a software development lifecycle? In this blog, we’ll discuss not only the benefits of visual requirements modeling as part of the broader model based testing practice, but why it is a game changing strategy — and a key building block to modern backlog grooming.

Humans Are Visual

Before we dive into technical specifics, it’s important to consider why visual requirements modeling for modern backlog grooming works to begin with. Recent studies have shown that the majority of humans are visual learners. Meaning that most people learn better from what they see in graphical, pictorial or video content. The rise of social media and mastering the talent of creating a pristine online appearance has only enhanced this already prevalent concept. Realizing these truths as the most effective way to communicate across different personas and occupations, can change the way people chose to do business. The reason creating a collaborative model is so successful is because it capitalizes on the visual learning tendencies across our race as a whole.

One Stop Collaboration Shop

Now that we understand why creating a visual requirements model is valuable, we can go into the specific of how it applies to a software development team as they venture into doing modern backlog grooming and other app development and testing activities. Traditionally, software teams are often siloed and have minimal contact with people outside of their immediate team. The business analysts or product owners write what the application needs to do, which typically gets passed to the development team in some kind of text format. The developers take their interpretation of that document and integrate code changes to reflect the new requirements. At the bitter end of the sprint, the quality assurance team, who may or may not know exactly how the application is supposed to perform, rush to test the final product. With this method, the majority of problems originate from lack of communication and product being tossed over the wall without much explanation. In fact, it is estimated that 56% of defectsstem from ambiguous requirements. This problem can be solved by utilizing a visual model so that every team can view and discuss one object without having to decipher written content.

Versatility is Key

Not only does your application model have to be something that is easily interpreted by different teams, but making modifications to the model has to be intuitive. Not all applications are equal, so your modeling tool should be able to accurately demonstrate the differences that exist between a banking or gaming application.

Modern Backlog Grooming


There can also be dramatic differences in related software within the same company. One development group might have a completely different set of tools and practices than another across the hall. So when choosing a tool, it is important to make sure it has all the integrations a company needs. This will minimize the work of transferring information and reduce the likelihood of artifacts being lost from manual steps.

Taking it to the Next Level

Agreeing on what your application should do is great, but can you make your model do more than just eliminate ambiguities? Because you already have all the different scenarios represented in a model, it is possible to define test cases in parallel. A model like this allows you start defining test cases as soon as the model is complete at the beginning of the software development lifecycle. In addition, mathematical algorithms can be run in the background giving the minimum amount of test cases with the maximum amount of coverage. So not only does spending the time defining a visual flow allow testers to fully understand why they are testing, it shifts their timeline to start sooner. Embracing this type of methodology breaks down walls between your software teams, giving you a competitive edge over competitors who are still stuck in the past.

CA offers comprehensive solutions that automates the most difficult testing activities – from requirements engineering through test design automation and optimization. Start your Free Trial of CA Agile Requirements Designer today and fast track your Testing at the speed of agile.