Join us for our next session!

Insights

BAs are uniquely positioned to embrace Generative AI

AI and reusable workflows – content creation

BAs are uniquely positioned to embrace Generative AI

2 December 2025 | Elizabeth Fullerton

When I was in my 20s, I took a break between school and uni. For most people, that’s a year – for me… closer to a decade. During that time, I lived in a number of different cities and countries, figuring it out as I went. That first leap into the unknown was terrifying and a little addictive. And I learned resilience, self-reliance, and the thrill of the new.

Then came Uni, and I ended up in a profession where every engagement is different – business analysis consulting. Every new role brings new people, new technologies, and new problems to solve. For many BAs, the ability to adapt to new circumstances is our bread and butter, and for me this was well-supported by my earlier wanderings.

Yet nothing has stirred that same excitement in exploration that I felt back then – until Generative AI. I’d dabbled with it over the past few years – like most people – for writing, summarising, brainstorming. Then I landed at a client with a quarantined instance of Gen AI available for all staff. Suddenly, I had the opportunity to really think about how I could use it to support my work, and how my BA skills could make the most of it.

In some ways, its introduction reminds me of the early days of MS Office. Organisations installed it and said “go forth and be productive” – but provided minimal training. People figured out Word was for writing, Excel was for numbers, and PowerPoint for pictures. And while these apps are widely used today, I think most people use only a small fraction of the functionality they provide.

Gen AI doesn’t have similarly clear boundaries. The possibilities aren’t obvious – you have to search for them, and it’s difficult to break out of old patterns. During experimentation, I kept reverting to thinking in the old ways – one prompt for written procedures, another for process flows – until I realised a single prompt could do both and be refined on the fly as needed. Suddenly, the possibilities bloomed from low-level assistance with basic tasks to rethinking how we work, end-to-end.

The power of Gen AI is that it can take on the lower-level, mechanical, time-consuming jobs, leaving us with more brainpower for the higher-level thinking we’re employed for but often don’t get to focus on. But it’s not set-and-forget. Ask it the same question twice and you’ll two get slightly different answers – not necessarily incorrect, but different. BAs have a lot of experience in distinguishing the useful output from the unhelpful. We can guide the AI, interpret its responses, and use it where it provides the most value. I think of it as a very smart grad who knows a lot but doesn’t yet understand nuance – and that’s how I work with it.

Of course, there’s the temptation to explore revolutionary AI solutions, and I’ve wandered down that path. What I found was transformative, but would require overhauling the BA methodology entirely, and we’re not ready for that, yet. Right now, we should be looking for incremental assistance that provides genuine value. The revolution will come when we better understand AI and how to best work with it within our profession.

Fortunately, one of the things BAs are very good at is looking at how things happen, identifying improvement opportunities, and integrating tools seamlessly into existing processes. We investigate, we explore, we talk to stakeholders, we map processes, and we dig into the detail. We have the skills to take Gen AI, understand its potential, experiment with it, and fold it into the way we already work. And as BAs, we’ll measure whether the changes are actually improvements or just a different way of doing things.

In terms of where we can take it – well, we’re limited only by our imaginations … and a single imagination is very limited, when you think about it. We must work together to plumb the depths of what AI can offer us, to seek uses beyond the obvious.

There’s a responsibility here for us as well. As AI takes on more of the low-level work, how do we make sure up-and-coming BAs gain the experience they need to develop the intuition and judgement needed to support future projects? We senior BAs need to consider this carefully and ensure that the next generation has a mechanism to build the foundation of critical thinking we were able to earn by pushing through that same low-level work in our own early years.

Generative AI isn’t just another tool – it’s a new and exciting way forward, one that BAs are uniquely positioned to make the most of. Our skills are exactly those that can make Gen AI a valuable addition to our toolkits, rather than just a curiosity.

I've rediscovered the thrill of stepping into the unknown, and it has a new name … Generative AI.

Prompting, environment, and ethics

11 November 2025 | Steve Gerlach

What an incredible November gathering for the AI Practitioner Network! We had our biggest turnout yet, bringing together a wonderfully diverse group spanning consulting firms, banking professionals, and sole traders – with skill levels ranging from those just starting their AI journey to seasoned practitioners.

The conversation was rich and thought-provoking as we explored three critical areas: the fundamentals of proper prompt form and formatting, the often-overlooked environmental impacts of AI systems, and the important question of how we frame and refer to AI-generated results in our professional work. The variety of perspectives in the room made for truly engaging discussions, with everyone contributing insights from their unique contexts.

A huge thank you to Aurora Marketing for providing the perfect meeting space, and to Lonsdale Solutions for keeping us fuelled with pizza throughout the evening!

We're already looking forward to December's final get-together for the year. If you're interested in joining us as we continue exploring the practical applications and implications of AI in our work, reach out to learn more about the Network.

AI and reusable workflows – content creation

14 October 2025 | Elizabeth Fullerton

Our October use case explored generative AI in proposal development, ensuring it supports human expertise rather than replacing it. The AI creates a draft based on curated foundational material, then humans do the bulk of the writing. The AI is used again under different personas to review the content from multiple perspectives, with humans providing the final polish.

This was an excellent opportunity to consider how this workflow can be applied elsewhere – anywhere content needs to be created, in fact. For most people, it's easier to edit something than to conjure words from an empty page. Even if the AI produces something less than perfect, we still save a lot of time and can preserve our brainpower to use where it's most effective.

The session naturally led into a discussion about ethics and accountability. Some organisations issuing RFPs now ask for transparency about how much of a proposal was AI-assisted and require that a named human takes responsibility for the content. This sparked a robust conversation about whether (and when) using AI counts as cheating. Ultimately, it was felt that as long as humans take responsibility for the output, AI can be an accelerator.

The content creation and review workflow we looked at can be adapted for almost any kind of content creation, with humans retaining control and accountability for the final output.

One prompt, two AIs

9 September 2025 | Elizabeth Fullerton

In September's session, we looked at how different AIs interpreted the same prompt using a deliberately messy recipe as an example. ChatGPT ignored our basic formatting instruction but intelligently reordered the steps to handle prerequisite steps properly. Claude formatted perfectly but followed the poorly-written step order without questioning the logic.

This reinforced an important lesson – we can't assume all AI tools work the same way, especially for structured outputs. What works in one might need tweaking in another.

We also explored how the AI output could be further refined into a format suitable for uploading into tools like Visio – creating a first-pass swimlane diagram ready for discussion and development with business stakeholders. There is clear value here – AI doing the grunt work in a fraction of a time it would take a human, and enabling us to spend that time on tasks that require more in-depth thinking.

Page updated

Google Sites

Report abuse