Download as pdf or txt
Download as pdf or txt
You are on page 1of 15

ChatGPT

FOR AGILE DEVELOPMENT

Business Breakthrough 2023


Elevating Agile with ChatGPT
Imagine a world where generating
ideas, writing user stories and
validating them with data and
customer feedback, all take
seconds, not hours. A world where
Agile teams can work more
efficiently and effectively,
delivering better products and
services to their customers.

Actually, there's no need to


imagine, as it is already possible.

ChatGPT, developed by OpenAI, is


a powerful language model that
has been trained on a massive
dataset of text, allowing it to By using ChatGPT to generate
understand and generate human ideas, write user stories in a
1
language like never before. consistent format, and validate
them with data and customer
feedback, Agile teams can
improve the efficiency and
effectiveness of their user story
writing process.

In this e-book, we’ll explore the


ways in which ChatGPT can be
used to enhance user story writing
in Agile development, and provide
tips for teams looking to
implement ChatGPT in their Agile
processes.

1
Business Breakthrough 2023
The Rise of Language Models
Language models, also known as
natural language processing (NLP) Language models have been used
models, are a type of artificial in a variety of applications, such
intelligence (AI) that are designed as text generation, question
to understand and generate answering, text classification,
2 machine translation, and more. 6
human language.

They have been on the rise in GPT-3, for example, is being used
recent years, driven by advances to generate text for chatbots, help
in machine learning and the with data analysis, and even write
availability of large amounts of code. 7
3
data.
Another example is BERT,
developed by Google, which is
being used for language
understanding tasks such as
sentiment analysis and named
entity recognition. 8

One of the earliest and most well- It is clear that language models are
known language models is the becoming increasingly
ELIZA, developed in the 1960s, sophisticated and versatile, with
which was able to simulate the ability to handle more
4
human-like conversation. complex tasks and understand
human language better.
Recent advancements in machine
learning have led to the The technology is expected to
development of much more continue to advance and be used
advanced language models such in a growing number of
as GPT-1, GPT-2 and GPT-3, applications, from improving
developed by OpenAI, which have customer service to automating
9
the ability to generate human-like various business processes.
text on a wide range of topics with
5
high coherence.

2
Business Breakthrough 2023
ChatGPT
ChatGPT is the most recent and
most powerful language model ChatGPT as a tool can be used for
developed by OpenAI. It has been a variety of natural language
trained on a massive dataset of processing (NLP) tasks such as
text, which allows it to generate text generation, question
human-like text on a wide range of answering, text classification,
topics.10 machine translation, and more.11

The possibilities are endless and


while still in early days we are yet
to explore the capabilities and all
the use cases ChatGPT and future
language models have to offer us.

ChatGPT is a continuation of the


GPT family. It uses the GPT3
model that was trained on a
dataset much larger than its
predecessor GPT-2, and it has the
ability to generate text with high
coherence and context awareness. 11

3
Business Breakthrough 2023
User Story Writing
User stories are short, simple descriptions of a feature or
functionality that a user wants or needs. They are used in
Agile development to help teams understand the
requirements of a project from the customer’s perspective.

User stories are important in Agile


development because they provide a clear
and concise way to communicate customer
needs to the development team, and they
help to ensure that the final product or
service meets those needs. 12

For example:

As a [user], I want [goal/desire] so that


[benefit].

This template is based on the widely used


"As a [user], I want [goal/desire], so that
[benefit]" format and it follows a simple and
easy-to-understand structure.
This format helps to ensure that the user
story is written in a way that clearly
communicates the user's needs and the
value that will be delivered to them. It also
makes it easy for the development team to
understand the requirements of the user
story and how it fits into the overall
project.
However, even the most experienced Agile
teams can struggle with writing clear and
effective user stories.

This is where ChatGPT comes in.

4
Business Breakthrough 2023
ChatGPT for User Story Writing
By using ChatGPT, Agile teams can streamline the user story
writing process and improve the quality of their user stories

Let us explore how...

1. Generating ideas for user stories

ChatGPT can be used to generate ideas for user


stories by analyzing market trends, customer
feedback, and industry data. This can help teams
identify opportunities for innovation and validate
assumptions about customer needs.

2. Writing user stories in a consistent format

ChatGPT can be used to write user stories in a


consistent format, which helps to ensure that all
user stories are clear and easy to understand.
This can improve communication and
collaboration within the team.

3. Validating user stories with data and customer


feedback

ChatGPT can be used to validate user stories by


analyzing data and customer feedback. This can
help teams ensure that their user stories are
aligned with customer needs and that they will
deliver value to the customer.

5
Business Breakthrough 2023
Scenario
An Agile team is working on a new
project to develop a mobile app for a
retail company. Among the key
features of the app is the ability for
customers to scan barcodes and view
product information.

The team is struggling to come up with a


clear and concise user story that describes
this feature. They have several ideas but
they are not able to put them in a clear and
specific format that aligns with the Agile
process.

Team members also have different understandings and


interpretations of the feature which makes it hard to agree on a
single-user story. They spend hours discussing and trying to
come up with the perfect user story but they are unable to
reach a consensus.

As a result, the team is falling behind schedule and struggling to


move forward with the development process.

A few specific examples of how ChatGPT could be used to solve


the issue the Agile team is facing

6
Business Breakthrough 2023
1. Generating a clear and concise user story

ChatGPT can be trained on a dataset of previous user stories,


and then used to generate a clear and concise user story for the
feature of scanning barcodes and viewing product information.

The output at the end could be something like this:

“As a customer, I want to be able to scan barcodes on


products using my mobile phone’s camera so that I can view
detailed information about the product, including reviews,
price, and availability”

2. Facilitating team discussions

ChatGPT can be used to facilitate team discussions by


generating questions and prompts that help team
members to clarify their understanding of the feature
and reach a consensus on the user story.

To assist, ChatGPT could generate questions such as:

“What are the main benefits of this feature for the


customer?”

“What are the key requirements for this feature to be


successful?”

“What are some potential challenges or limitations of this


feature?”
7
Business Breakthrough 2023
3. Identifying and resolving conflicts

ChatGPT can be used to analyze team members’ perspectives and


identify conflicts or discrepancies in their understanding of the feature.
ChatGPT can then generate suggestions for resolving these conflicts,
such as conducting additional research, seeking input from customers,
or reworking the user story to better align with the team’s understanding
of the feature.

Specific examples of how ChatGPT can be used to analyze team


members’ perspectives and identify conflicts:

Identifying conflicts in team Generating suggestions for


members’ understanding of resolving conflicts
the feature
ChatGPT can generate
ChatGPT can analyze team suggestions for resolving conflicts
members’ discussions and input based on data and best practices.
related to a feature, and identify For example, ChatGPT could
any conflicts or discrepancies in suggest conducting additional
their understanding of the research on industry standards or
feature. For example, ChatGPT customer feedback, seeking input
could identify that team members from subject matter experts, or
have different ideas about the reworking the user story to better
functionality of a feature, or align with the team’s
different interpretations of understanding of the feature.
customer requirements.
Prompt:
Prompt:
“What are some suggestions for
“Analyze the different resolving the conflicts regarding
perspectives of the team feature X?”
members regarding feature X.”

8
Business Breakthrough 2023
Assessing the impact of 5. Estimating effort and identifying
conflicts risks

ChatGPT can assess the impact of ChatGPT can be used to assist with
conflicts on the project’s sprint planning, by helping the team
timelines, budget, and overall to estimate the effort required for
outcome, and suggest each task and identify risks.
adjustments. For example,
ChatGPT could suggest extending Given that ChatGPT is trained on
the timelines, increasing the data related to the specific sprint
budget, or modifying the scope of and has access to the information
the project to mitigate the impact about the identified dependencies
of the conflict and potential risks, it will be capable
of answering prompts related to
Prompt: such issues.

“What would be the impact of the


conflicts on the project and how Specific examples of how ChatGPT
to address them?” can be used to assist with sprint
planning:

Analyzing task dependencies

ChatGPT can analyze the tasks


included in the sprint and identify
any dependencies between them.
For example, ChatGPT could identify
4. Prioritizing and tracking user that a task to implement a new
stories feature is dependent on the
completion of a task to update the
ChatGPT can be used to assist underlying software architecture.
with backlog management, by
helping the team to prioritize user Potential prompts:
stories based on their business
value and dependencies. ChatGPT “What tasks are dependent on the
can also be used to track completion of task X?”
progress, identify bottlenecks and
generate reports and dashboards. “Which tasks will be impacted if task
Y is delayed?”
9
Business Breakthrough 2023
Identifying potential risks

ChatGPT can analyze the tasks and


features included in the sprint and
identify potential risks, such as lack
of resources, unrealistic timelines,
or technical challenges. For
example, ChatGPT could identify
that a task to implement a new
feature has a high risk of technical
challenges due to the complexity
of the required software.

Potential prompts:

“What are the potential risks Suggesting adjustments


associated with task X?”
ChatGPT can suggest adjustments
“What are the chances of task Y to the sprint plan in order to
not being completed on time?” mitigate the identified risks and
ensure that the sprint is
completed on time and within
budget. For example, ChatGPT
could suggest that the team
allocate additional resources to a
task or extend the timeline for a
task to account for potential risks.

Potential prompts:

“What adjustments can be made


to the sprint plan to mitigate the
risks associated with task X?”

“How can task Y be completed on


time despite the identified risks?”

10
Business Breakthrough 2023
Monitoring and tracking

ChatGPT can monitor the


progress of the sprint, track the
identified dependencies and
potential risks, and notify the
team of any changes or new risks
that arise. This can help the team
to make adjustments as needed to
ensure that the sprint is
completed on time and within
budget. Identifying and mitigating
bottlenecks
Potential prompts:
ChatGPT can assist the team in
“What is the current status of the identifying bottlenecks in the
identified dependencies and process and suggest ways to
potential risks?” mitigate them. For example,
ChatGPT could identify that a
“Are there any new risks or certain task is taking longer than
changes that have arisen since expected due to a lack of
the last sprint review?” knowledge or resources, and
suggest providing training or
allocating additional resources to
resolve the bottleneck.

Potential prompts:

“What are the current bottlenecks


in the process that are impacting
the sprint’s progress?”

“What can be done to remove or


mitigate the identified
bottlenecks?”

11
Business Breakthrough 2023
Data Required
In order to provide answers to
prompts related to analyzing team
members’ perspectives and 3. Data on team members’
identifying conflicts, ChatGPT expertise and roles
would need access to data.
This could include information
What types of data would be about each team member’s skills,
needed? experience, and role within the
organization, which would help
1. Team members’ discussions ChatGPT to better understand the
and input related to the context of the team members’
feature perspectives and input.

This could include meeting notes,


emails, chat transcripts, and other
forms of communication where
team members discuss and
provide input on the feature.

2. Industry standards and best


practices for resolving conflicts 4. Data on the project’s timelines,
budget, and overall outcome
This could include information
about common techniques for This data would be necessary for
resolving conflicts within Agile ChatGPT to assess the impact of
development teams, such as conflicts on the project’s
conducting additional research, timelines, budget, and overall
seeking input from subject matter outcome, and suggest
experts, or reworking user stories. adjustments.

12
Business Breakthrough 2023
5. Customer feedback or external
data By leveraging the power of
ChatGPT, teams can work more
ChatGPT could use customer efficiently and effectively, and
feedback or external data to deliver better products and
validate the assumptions and services to their customers.
understanding of the feature, this
will help to ensure that the user With its natural language
story aligns with customer needs processing, data analysis, and
and that they will deliver value to decision-making capabilities,
the customer. ChatGPT can be a valuable tool for
Agile teams looking to improve
their user story-writing process.
The future of Agile development is
here and ChatGPT is a game
changer.

Conclusion
Hopefully, you have found this
User story writing is a crucial ebook informative and helpful in
aspect of Agile development that understanding how ChatGPT can
can often present challenges for be used to streamline the user
teams. However, with the help of story writing process in Agile
ChatGPT, Agile teams can development.
streamline the user story-writing
process and improve the quality Thank you for taking the time to
of their user stories. read this ebook and for your
interest in using ChatGPT to
ChatGPT can be used to generate improve your team's workflow.
ideas, write user stories in a
consistent format, and validate
them with data and customer
13
feedback.
Business Breakthrough 2023
References
1.Roose, K. (2022). The Brilliance and Weirdness of ChatGPT. Retrieved from NYTimes:
https://www.nytimes.com/2022/12/05/technology/chatgpt-ai-twitter.html

2.IBM. (2023). What is natural language processing (NLP)?Retrieved from IBM:


https://www.ibm.com/topics/natural-language-processing

3.Dean, J. (2022). Google Research: Themes from 2021 and Beyond. Retrieved from
Googleblog: https://ai.googleblog.com/2022/01/google-research-themes-from-2021-
and.html

4.DeepLearning.AI. (2023). A complete guide to natural language processing.


Извлечено от deeplearning.ai: https://www.deeplearning.ai/resources/natural-
language-processing/

5.OpenAI. (2022). Aligning Language Models to Follow Instructions. Retrieved from


OpenAI: https://openai.com/blog/instruction-following/

6.Wang, H., Lib, J., Wua, H., Hovyc, E., & Suna, Y. (2022). Pre-Trained Language Models
and Their Applications. Engineering, https://doi.org/10.1016/j.eng.2022.04.024.

7.Davenport, T. H., & Mittal, N. (2022). How Generative AI Is Changing Creative Work.
Retrieved from hbr: https://hbr.org/2022/11/how-generative-ai-is-changing-creative-
work

8.Google. (2023). Getting started with the built-in BERT algorithm. Retrieved from
google: https://cloud.google.com/ai-platform/training/docs/algorithms/bert-start

9.Yoon, S. (2022). 17 ways technology could change the world by 2027. Retrieved from
Weforum: https://www.weforum.org/agenda/2022/05/17-ways-technology-could-
change-the-world-by-2027/

10.OpenAI. (2023). Commonly asked questions about ChatGPT. Retrieved from openai:
https://help.openai.com/en/articles/6783457-chatgpt-faq
11.Marr, B. (2022). ChatGPT: Everything You Really Need To Know (In Simple Terms).
Retrieved from Forbes:
https://www.forbes.com/sites/bernardmarr/2022/12/21/chatgpt-everything-you-
really-need-to-know-in-simple-terms/?sh=35471e8ecbca

12. Cohn, M. (2004). User Stories Applied: For Agile Software Development. Addison-
Wesley Professional.

14
Business Breakthrough 2023

You might also like