As analysts, typically we’ll spend a lot of time examining and critiquing the processes of other departments. We might be looking for ways to make an operational process slicker, quicker and ‘better’, or we might be looking to solve particular problems. There are a whole set of elicitation, problem analysis and process analysis techniques in our toolkit that help us do this. Yet how often do we turn these tools in on ourselves to examine our own practices to see if there are ways that we can improve? As practitioners of change, surely we should be looking to continually improve ourselves too?
I suspect we all intuitively do this, at least to a degree, but I wonder if it’s an area where there’s room for improvement. I’ve included some examples to whet your appetite below:
- Deciding How To Decide: Have you ever got to an approval gateway in a predictive (waterfall-type) project and it’s unclear who needs to make the approval? Or the approver shirks responsibility? This can manifest differently on adaptive (agile/evolutionary) initiatives, for example with wrangling over the ‘definition of done’ with different stakeholders having different (and unresolved) perspectives. These things can fester if unresolved, perhaps stakeholders concede in the short term and sign on the dotted line, but resentment builds and bubbles up, only to explode out later at the most inconvenient time.
This creates friction and drama that takes time to resolve, and when it comes to organizational change, time is money. It’s therefore beneficial to spend a little bit of time up front agreeing how these types of decisions will be made. This sounds so obvious doesn’t it? Yet, so often in the blind rush to “just get going” it isn’t discussed… and it is only as the decision emerges is the omission spotted. Practical tools such as the RACI matrix can be extremely useful here.
- Planning the Requirements Architecture: When BABOK® v3 was released back in 2015, one of many significant additions was a more explicit recognition of requirements architecture. If you have ever found yourself mid-project with a whole set of useful, but disparate requirements artifacts (“These process models are great. But how on earth do they relate to those user journeys, these scenarios and those wireframes?!) you’ll know what I mean.
Taking time up front to quickly and roughly sketch out how it’s anticipated that the requirements will fit together helps avoid this. Of course, things change, as requirements emerge, new types of artifacts suddenly become relevant (“Ah, so statuses are important… I think I’ll include a state transition model…”), however having a starting point to deviate from when appropriate is better than fumbling around in the fog.
On a legislative project, you might write a quick problem statement to act as a high-level goal/outcome, and define some critical success factors/key performance indicators which will act as desired outcomes. These will all link to the organization being compliant with a piece of legislation, that’s another (external) artifact, but one which some rules can be derived from. Those rules will include internal decisions about how the legislation is interpreted; so those probably need to be captured too. Those rules might be automated or orchestrated via processes, and there might be steps in a process which will be automated via some detailed functional solution requirements. You can see how these different concepts relate to each other here; and of course there will be other types of artifacts too. The point is that creating a quick sketch showing how the concepts map together before creating them will prevent artifacts from being created that don’t clearly relate to each other.
- Planning How To Store/Manage Requirements Artefacts: If you’re lucky, you’ll have some form of requirements (or story) management tool. If that’s the case, does everyone actually know how to use it? Is there a common agreement around how things like priorities/statuses and other metadata will be used? If not, will this create noise and friction as the initiative progresses? If so, a short discussion up-front is likely to yield significant benefit.
If you are using documents and a shared document repository (e.g. word processing tools, drawing tools, spreadsheets etc.), decide things like naming conventions and version control conventions up front. It can be very confusing when someone in the team is using a file naming convention that uses “v0.1”, “v0.2” then “v1.0” when it’s signed off, when another person is using “FILENAME v1.0 (Updated) (Second Updates) (Final) (Really Final This Time) (Updated Again).docx”
- Plan For The Analyst And Stakeholder Of Tomorrow: Stuff you’re creating today will be useful for the analysts and stakeholders of tomorrow. That process model you’ve created? If it’s detailed enough, I bet the training team would love to use it, and the operations team might too. It might be the catalyst to the creation of a single, unified process repository (if that doesn’t already exist in your organization).
Those performance non-functional requirements for your customer-facing website? You know, the ones that were like pulling teeth to elicit, that required benchmarking and speaking to the technical folk? They might be a useful baseline and starting point for others producing customer-facing web applications within your organization.
As we create artifacts, we ought to be thinking not just about how they can be used today, but how they might be used tomorrow and how we can ensure they will be found. This is a much bigger, organizational, question—however it’s one of the many areas where BAs can nudge the agenda. By creating common pools of knowledge, and by encouraging information sharing we open up channels for these types of artifacts to flow. This has the advantage that information flows in both directions and also that there will be a wider range of documents available at the start of projects too.
Of course, these are only four suggestions, there will be many more. The key is that we shouldn’t rest on our laurels; as BAs we should be looking to hone our craft and improve the efficiency and effectiveness as much as we can. This involves not just “speeding to get the job done”, but also thinking about the stakeholders of tomorrow!