You’ve likely heard about ChatGPT, the free tool released by OpenAI.
Imagine being able to expand your thinking, create draft documentation with ease, and streamline your work as a business analyst.
This is all possible with ChatGPT, and despite the industry-wide fear of AI replacing business analysts, I think a better question is “how can we use these tools to increase our efficiency and effectiveness as a business analyst?”
In this video, I’m sharing how to use these tools to expand our thinking and make our job as business analysts way easier!
I share my experience creating a Use Case with ChatGPT, in the video, which includes many of the same sections that we have in our free Use Case Template, but not all.
You can download our Use Case Template for FREE which includes instructional text for what belongs in each section.
- Validating that the use case reflects true end user needs, so the solution is what the business expects.
- Describing system and user steps at the right level of detail, so that your use case contains exactly the information your technical team needs (even if you don’t understand how the system is built).
- Ensuring you aren’t making some very common mistakes that make your use case unclear, and lead to ambiguity or missed requirements.
Download our FREE Use Case template to gain the confidence you need to lead your next project.
You’ve likely heard about open AI and, specifically, the currently free tool Chat GPT. Imagine being able to expand your thinking, create draft documentation with ease, and streamline your work as a business analyst. This is all possible with this tool, Chat GPT, despite the industry wide fear of AI and artificial intelligence somehow replacing business analysts. I think a better question is, “How can we use these tools to increase our efficiency and our effectiveness as business analysts?” There’s immense potential for these tools to expand our thinking and make our jobs easier. If that’s sounds like something that you’d like to do, just stick with me and I’ll show you exactly how to do that.
Hi, I’m Laura Brandenburg with Bridging the Gap where we help you start, succeed, and excel in your business analyst career with weekly videos on business analysis tips and techniques.
About OpenAI and ChatGPT
Chat GPT was originally released back in June 2020, which feels like a lifetime ago at this point. It was an open source language model developed by Open AI. Open AI is a for-profit AI research laboratory. I think that’s just a really important piece of context to keep in mind. Sometimes when free tools are available, it can just feel like they’re from non-profits or sponsored by the government or whatever. But this is a for-profit company that’s released this tool for, most likely, a limited as a free resource to the community, and it’s in an exploration stage with how that tool is going to become part of its business model.
In November 2022, specifically, this is when the GPT Chat bot was launched as a prototype, and that is when it enabled anyone to register for a free account and start using the tool with natural language chats. As of this recording, in February 2023, Open AI has announced making a version of this tool available for $20 a month that would provide more guaranteed access. Sometimes when you go today, it’s like our systems are too busy because everybody’s using this tool to help be more efficient in their work. So that subscription, as I understand it, would enable you to have VIP or primary access to the tool.
It’s clear this is just the beginning. The current tool and AI generally, these are powerful resources for us as business analysts to use. From everything from drafting requirements models, to brainstorming questions to ask your stakeholders. I’m going to show you how to do that. So, let’s go ahead and see how this could be used to draft a use case.
Example: Using ChatGPT to Draft a Use Case
I am going to type in the question here, “Please draft a use case titled Combine Accounts.” And then we get to wait for a moment, and here you go.
One of the things, while it’s generating this response that I want to share is that one of the limitations of this tool is it’s using information from pre-2022, and it’s compiled that together, and as it’s doing this, my understanding is it’s creating unique content from the available information. It’s being written and regenerated in real time, always using a different perspective. I did this before, as a test, before I did this video, and this use case now is different than that one that was generated before.
So let’s just take a look at what’s going on here.
There’s a description. The combined and accounts use case allows users to merge multiple separate accounts into a single account. Makes perfect sense. This is useful for individuals who have created multiple accounts over time and would like to streamline their account information.
We’ve got the actors, the pre-conditions, some user system interactions that seem to make pretty good sense, prompts the user to log in credentials, verifies those credentials. That’s interesting.
I would often put “login” as a separate use case. That’s kind of typically considered a best practice, that a pre-condition would be the user being logged in versus how they have it here. A user logging in, you know, has their credentials. There are different nuances on how different people handle use cases, but typically a pre-condition is something that the system can actually verify. The system can’t verify that the user actually has their login credentials.
The spin on the use case here is maybe slightly different than a typical use case that I personally would write. Some of that is personal organizational preference, though. Just throwing that out there.
Then there are some really good steps here. The system merges those selected accounts. But again, it’s kind of missing those details. As an individual writing this use case for your organization, you will need to figure out what are those details that need to be merged and what overwrites what? What’s the primary account? There are some business rules that are specific to your organization that you’ll need to fill in. And then finally the user logs out. Again, I would put that as a separate use case. But some of that stuff is preference.
Let’s look at what we are receiving here, because this is a really great draft. If you were sitting down and were like, “Oh my goodness, we have this feature to combine accounts. I have no idea what I should be doing.” This would be a great way to give yourself a start and to start thinking about it. But again, we can see your critical ability, or thinking ability as a business analyst is still really important in making sure that this reflects what is used for your business.
One of the things I wanted to note, too, is this use case does include most of the sections that are available through our free use case template download.
Example: Using ChatGPT to Identify Requirements Questions
Okay, so let’s take this a bit deeper. This was a great draft of the use case. I could work with this as a starting point. I’m going to say, well, you know, I’m going into a stakeholder meeting. Can you please give me 10 questions I could ask a customer service representative? A very specific role in the organization, about this feature. This is another challenge that business analysts face. I’m going into this meeting and I know I’m supposed to have an agenda. I know I’m supposed to be asking questions. What questions should I be asking?
We actually have a whole product for this called our requirements discovery checklist pack. But I think it’d be great to use Chat GPT for this as well.
These are great questions. How do I combine my accounts? Is there a limit to the number of accounts I can merge? What happens to the information from the accounts that I discard? Can I undo an account merge once it’s complete? These are great, great questions to getting through to the root cause of the features and the full business rules. Let’s just take this yet another step deeper.
Example: Breaking Down a Use Case Into User Stories
What we teach in The Business Analyst Blueprint® certification program is a strategy for extracting user stories from your use case. So your use case includes the big picture of that user system interaction and how the whole feature flows. The user stories might have specific parts of that feature that are going to get implemented in different development cycles.
Could you please, notice my politeness? I’m always like, “please.” I’m sure the machine does not really care, but it just still feels like the right thing to do, break this use case down into user stories? Here we’re getting those user stories. As a user I want to be able to do something. As a user with multiple accounts, I want to be able to combine my account so I can streamline my account information. Different user stories that are covering this whole feature of how do I combine accounts. The specific pieces that might be implemented piece by piece.
Limitations of ChatGPT – And Why Business Analysts Are Still Critical
That is an overview of Chat GPT and kind of how it can work to improve our efficiency as a business analyst. I think it’s a great starting point here creating user stories. There’s quite a few. That’s a great starting point for us, and there are a lot of limitations as well.
One of them that I already mentioned, is it’s only data from pre through 2022. If there’s a new feature or a new capability or something that is relatively recent with your functionality, it’s not going to have the information, it’s not going to be training to use that kind of information to generate a response that would be meaningful for you.
It’s also really unable to determine what’s important and what’s not from a…there are no ethics in humanity involved in this. It is trained on the data set, so it is as good or as weak as the data set that it’s been given. This is a really important piece to keep in mind as we think about AI tools in general is like where’s the human ethics aspect of it? In this use case of combining accounts, that might not be very relevant, but in general, when we’re asking it questions, it is not going to bring through our norms of how we want to handle diversity and inclusion or how we want to be a proponent of social justice or other values that may be important to us or to our organizations. That is not going to come out through a data set or a generation like this. We really need to bring that to it.
It’s really unable to determine as well, what is important to your business and your stakeholders. We’re not mining information specific to your company and it’s definitely not helping you get your stakeholders aligned on what this feature should look like. It’s giving you a starting point, a set of questions to ask, and it’s not going to be specific to your business at all, although you could definitely drill in and say, like, if I’m running an insurance company, what would this merge account feature look like? And it will give you another version of that that would be much more specific to your domain.
The sense that I want to give you here is that AI, it’s a powerful tool. It’s an efficiency creating tool, but it’s not an end all and a be all. It’s going to give you a draft, it’s going to give you a starting point, but not your final product, and it’s not going to help you, it’s not going to create alignment and clarity among your stakeholder group.
The key message I want you to receive is there is no substitute for the thinking you do about the requirements and how they impact your business as a business analyst.
Limitations of ChatGPT – In Succeeding as a Business Analyst
Now, let’s just talk about, also, some of the limitations to you when it comes to succeeding as a business analyst. Because as you saw, I started to kind of pick apart exactly how that use case was structured. And so you need to know what a use case is, how to evaluate the results that were generated, what are best practices in terms of how your organization organizes use cases and functional requirements. And if you’re not already familiar with that or you need a refresher, I did a full video on what a use case is and how to write one, and you can watch that video below.
I’ve also found, often, that this tool will put me into a little bit of a spin. So I have used it to flush out outlines for videos on social media posts, and it gives me a lot of great ideas. Often, it gives me almost too many ideas. And so I need to do the work to filter out what’s really important from my audience and the point that I want to get across. And so just taking the outline as it is and copying and pasting it, or even allowing this tool to write an entire video script, it’s totally possible. But if I did that, it wouldn’t come across as me. It wouldn’t be the kind of content you’re used to getting from Bridging the Gap.
The same thing is true as you’re using it to draft requirements. It’s your starting point, but you need to really bring that flavor. And if it’s putting you in a spin and a bit of overwhelm, a lot of business analysts really struggle with imposter syndrome. And so seeing a tool create a draft of a document in seconds when it takes you hours the first time through, or hopefully within time you’re creating something like that within 20 to 30 minutes as a rough draft. It can be a real mind bender to see a tool just spit it out just like that.
It can also have a sense of authority of like the tool knows better than me. This tool, they put “login” in there. Why am I not doing that in my use case? Or, “Oh, this tool has it done this way and left out these sorts of details. Why am I not doing that?” You are still in control. You are still the authority. The tool is a resource to help you. It’s a really important mindset to keep in mind as you’re using a tool like this.
Just being aware that it could send you off on tangents that your business rationale and goals do not support. You want to be really careful that your use of Chat GPT doesn’t give you scope creep that your stakeholders aren’t even asking for. We already have that problem within our project. We don’t need more.
This is just a side note. This is where access to coaching and mentoring and training really remain critical for you and your mental game. In our programs, our instructors are still doing real live reviews of your actual work and documentation because we know how important that is in cultivating your competence. Having that kind of review cycle and that training cycle is still incredibly important.
The Future of AI Tools and How They Impact Business Analysts
With all of this said, I think it’s also important to remember these tools are still relatively new and we’re just starting to see the shimmer of their potential and how disruptive already this technology has been just in the last few months.
They are going to continue to expand and evolve. They’re going to continue to improve. They’re going to get better. We could see a future where these sorts of tools identify and merge duplicate requirements, show requirements conflicts for us automatically, and map the information resources of your organization to be used to identify requirements and project impacts. These tools can be a great thinking and analysis ability. Even that is just probably the beginning.
Given that these tools will continue to get smarter and will continue to increase in capability, how can you position yourself?
There’s no point, really, in being afraid of the tool or just trying to push it down or put your head in the sand, which honestly is what I was doing for probably the last few years. Instead, this is the time to dive in, embrace and learn how to leverage the tool set to increase your efficiency. Bring these ideas to your team and into your projects. Free up your time right here and now for the capabilities that are not replicated by machines. Those capabilities are your ability to build relationships, your ability to bring people together, creating alignment and clarity, gaining buy-in, ensuring the requirements you write actually meet the business objectives, and that your team can deliver them and that they achieve real business value.
How do You Use ChatGPT as a Business Analyst?
With anything, these tools are here to support us with the more limited sort of menial work of a business analyst so we can shift our energy and attention to the more meaningful aspects of our work. We are just beginning to see the power of these types of tools in terms of supporting our workflow.
If you have used Chat GPT in your workflow as a business analyst, I would love for you to drop me a comment below and let us know, how are you using this tool? How has it helped your effectiveness as a business analyst? What are your wins? What are your questions? What are your fears around it? Where are the places that you get stuck? Where has it helped you and where has it left you wanting more.
Remember, the tool has a lot of limitations, but there is also so much more to a use case than that AI generated version. Just like AI is not going to replace business analysts, its output won’t replace human insight, creativity, customized problem solving to the needs of your specific business. And you need to know why each section of that use case is important and how to think through it so you can validate that information with your stakeholders. That’s why our use case template is so valuable and will help you do just that.