The Tyranny Of “Default”: Be Careful When Choosing Analysis Techniques/Artifacts!
At its essence, a lot of what we do as business analysts involves co-creating a shared understanding. Whether that’s a shared understanding of a problematic situation, or of a set of requirements or features, a common theme is that we work to understand different perspectives and ensure that people are ‘on the same page’.
This can cause some challenges when we want to communicate complex ideas. Text is a challenging media to convey complexity, and the English language is just brimming with words that have multiple meanings. For that reason, it’s very usual to use a combination of text, models and conversations to ensure that key stakeholders are aligned.
When we create a model, we are essentially codifying some information in a way that can be later referred to. Codifying typically involves using some form of notation (whether that’s a business process modeling notation, or the symbols of an alphabet when writing text), and there will usually be particular rules on what elements of the notation mean and how they should be used. For example, particular shapes on a process model represent particular things, particular words in a textual paragraph have a particular meaning.
Yet when we encode our analysis in this way, how often do we reflect and think “will my stakeholders understand this?”. Will the consumers of the information both know how to read it, and, ideally like referring to it as it’s a format that works for them? Perhaps we don’t think about this quite as often as we could…
The Tyranny Of “Default”
I have a confession to make. I love Business Process Model & Notation (BPMN). I could quite easily ‘geek out’ and create process models to an executable level, using just about every BPMN symbol and construct that is available to me. But, as those of you who are familiar with BPMN will attest, if I did that the resulting diagram wouldn’t be very accessible to the average stakeholder. It would be somewhat akin to me sending an email in French to someone who only speaks English. It might be possible to pick out some elements of the communication and work out what is going on… but it’d be very difficult to draw conclusions.
This absolutely isn’t an argument against BPMN by the way—it’s stil one ofl my preferred notations—but it highlights the importance of thinking of the artifact’s audience and choosing an appropriate level. With BPMN, I might create a ‘view’ of a process using a cut-down palette of symbols, and model at a much higher level, so it is more intuitive. Equally, I might choose to use ‘boxes, arrows, diamonds and lines’ instead where it is appropriate to do so…
The key point here is flexibility. It is very easy to get stuck in a rut and choose a technique because it’s the default. You probably have a favorite approach or technique, which might mean you tend to default to a particular type of artifact. Or perhaps your organization has templates or tools that ‘nudge’ you towards a particular style. That can be useful, it can embed consistency… but it can also lead to things being articulated in a particular way “because that’s the way we do it here, and that’s the way that we’ve always done it”. Neither of which, on their own, are particularly compelling reasons.
Light The Fuse: Ask “Why Are We Using User Stories”
If you like a bit of controversy, next time you are working on an agile project, ask “why do we use user stories here?”. Usually, the response will be a blank face, followed by either “because we’re agile” or “well… the tool is set up for user stories… so….that’s what we do”.
But are these really valid responses? There’s certainly nothing in the agile manifesto that decrees user stories are mandatory. The term ‘user story’ isn’t in the Scrum guide at all. So why default to user stories?
Of course, user stories do have merit, particularly in agile. Yet the choice to use them should be a conscious one. In fact, user stories alone are probably not enough… they will probably need to be supplemented with regular collaboration, acceptance criteria and you might even decide to document some journeys or prototypes. It’s likely that most successful teams already do this, but it is worth highlighting. Some of this might emerge as you discover more about the requirements… but having at least an idea of what the requirements architecture will look like up front will save pain later (e.g. “We might use prototypes, so let’s agree that one prototype will span multiple user stories… we’ll capture the links here…).
So, this article certainly isn’t criticizing user stories. However, as with any profession, we need to regularly challenge ourselves and challenge our ‘defaults’. It is worth remembering that we have a wide toolkit at our disposal. If a plumber approached your sink with a hammer and started smashing it (rather than figuring out the right tool for the context), you’d be somewhat annoyed. An explanation of “ah, I always use a hammer” is unlikely to help. The same is true in our profession too… and it is worth remembering how broad our toolkit is. Which provides us with many possibilities!