I am trying to create a prompt that will research companies in the finance industry that have used data, in particular data engineering to leveraging data engineering to streamline due diligence and investment management processes.
For example I have created the following prompt:
“You are to act as a research analyst. You are to list small to medium size companies in the Finance Industry, (with a focus on Asset Management Firms), that have used the services of the Data Engineer to build Data Pipelines to extract and analyse data from various sources, and a Data Analyst to make Data Driven decisions Please provide a short description on how they have used Data Engineering and the benefits”
The outcome of the above prompt is as follows:
Bessemer Trust Location: New York, USA Overview: Bessemer Trust is a multi-family office that offers wealth management and investment services. Use of Data Engineering: Bessemer Trust employs data engineers to build data pipelines that integrate information from various financial systems, customer portfolios, and market databases into a single platform. These pipelines ensure that all client portfolios are continuously updated with the latest market data. Benefits: Having a unified and up-to-date data view allows Bessemer’s analysts to provide better client service, as they can offer real-time investment recommendations based on comprehensive, accurate data. This enhances their ability to manage client portfolios more effectively.
However, when I manually research the firm, it has never used Data Engineering in it’s organization, however it may have used Data in some form to make data driven decisions.
Therefore, can someone help with a prompt which will help research how firms have exactly used data in their organization to make data driven decisions?
Because I don’t have ‘DATA’ to test it, I cannot see the exact output.
You may try prompt below. You can add what kind of ‘Sources or references’ you want to see in section 4:
You are to act as a detailed research analyst. Your task is to identify and list small to medium-sized companies within the Finance Industry, with a focus on Asset Management Firms, that have publicly disclosed the use of Data Engineering and Data Analysis to improve their due diligence, investment management processes, or overall decision-making. The focus should be on real-world examples where data engineers have built data pipelines to extract, process, and analyze data, and where data analysts have used this data to drive investment strategies or business decisions. Please provide:
1. The name of the company.
2. The location of the company.
3. A brief description of how the company has specifically used data engineering and data-driven decisions in their investment management or due diligence processes.
4. Sources or references (such as articles, reports, or case studies) that support these findings.
Sometimes one has to think how one would search for the information and then adapt that for use with a prompt. The information you seek might not be incorporated in the learning material used to train ChatGPT, however ChatGPT might know of databases that contain such companies.
As such I modified the prompt to
List online public databases of small to medium size companies in the Finance Industry, (with a focus on Asset Management Firms), that have used the services of the Data Engineer to build Data Pipelines to extract and analyse data from various sources, and a Data Analyst to make Data Driven decisions
and this was the result
I found several resources that highlight companies and databases relevant to your interests in finance, particularly focusing on asset management firms that utilize data engineers and data analysts for data-driven decision-making:
1. **Enigma** offers corporate intelligence including firmographic and financial well-being data for small to medium-sized businesses. This data is particularly useful for risk monitoring and internal operations in the financial sector【6†source】.
2. **Flowcast** utilizes AI to help businesses and financial institutions make data-driven credit decisions, enhancing predictive modeling without the need for in-house coding【6†source】.
3. **Coresignal** provides a vast dataset including financial data of companies across multiple countries, which can be used for data-driven decision-making【7†source】.
4. **HitHorizons** offers an API that provides aggregated financial data on companies across Europe, which can be integrated into various software for comprehensive financial analysis【7†source】.
5. **Bright Data** has a Historical Financial Dataset that covers a wide range of company-specific financial metrics across numerous countries, aiding in performance analysis over time【7†source】.
6. **Keboola** and **Domo** are platforms that offer robust data integration and management solutions tailored for businesses, including those in the finance sector, to facilitate data-driven strategies and decisions【11†source】.
These platforms and companies represent a blend of data aggregation, analysis, and financial insights, which are critical for asset management firms looking to leverage data for strategic decision-making. Each offers unique tools and datasets that can enhance data pipeline construction and data analysis capabilities.
As always, check the reply for hallucinations and accuracy.
Another useful bit of knowledge when prompting is to notice the keywords returned that are substitutes for word(s) in the prompt, in this case the word resources was used. As such when modifying the prompt again, I would use the word resources instead of online public databases.
I have refined the prompt, see below. Can someone show me how to formulate the prompt into more manageable outputs:"
As a research analyst, your task is to identify and compile a list of small to medium-sized companies within the Finance Industry, specifically focusing on Asset Management Firms. These companies should have utilized Data Engineering or Data Analytics in some form to create data pipelines that facilitate the extraction and analysis of data from various sources, as well as a Data Analytics to make data-driven decisions. For each company listed, please verify the firm has used data engineering and/or data analytics within its organization and please include a brief description that elaborates on their use of Data Engineering, detailing the specific ways they have implemented data pipelines and how these practices have positively impacted their operations or decision-making capabilities. The output should include a short description of the firms primary business focus and be structured in a list format, with each entry containing the company name, a concise description of their data engineering efforts, and an explanation of the benefits they have achieved as a result.
Let’s try again, I don’t think some of our key points are standing out enough.
See if ChatGPT has at the very minimum a list of a few companies that you seek.
Start with a simple prompt to list a few companies, then modify that to get just companies within the Finance Industry and keep trying more prompts to see if you can get a list of a few companies you seek or realize that ChatGPT can not make such a list and then you made need to ask for resources that might have such instead.
If you can get ChatGPT to generate such a list of companies, even if it is only two or more then use those company names with a few-shot prompt to get a larger list.
Once you have a prompt returning a list of desired companies, then modify the list to return the needed details.
If you know how to write SQL queries this is similar. Start with something simple then build up the query to achieve the desired result.
Also, please show us what you would expect as a result and the prompt and result that ChatGPT is returning. At present we can only guess if you are making progress and guessing is not a wise use of time.
As a research analyst, your task is to compile a list of small to medium-sized companies in the Finance Industry, with a specific focus on Asset Management Firms. These companies should have publicly disclosed their use of Data Engineering or Data Analytics to create data pipelines that facilitate the extraction, processing, and analysis of data from various sources. Additionally, these firms should have used this data for making data-driven decisions.
For each company, please ensure the following details are included:
1. Company Name: The official name of the company.
2. Primary Business Focus: A brief description of the company’s main operations within the finance or asset management sector.
3. Use of Data Engineering/Analytics: How the company has utilized data engineering, particularly data pipelines, and data analytics within their organization. Focus on specific examples where possible.
4. Business Impact: A concise explanation of how the company’s use of data engineering or analytics has positively impacted its operations or decision-making capabilities. Examples might include improved efficiency, enhanced decision-making, or increased client satisfaction.
5. Sources: Please provide verifiable sources (e.g., news articles, reports, case studies) that confirm the company’s use of data engineering or data analytics.
The output should be presented in a list format with each company’s information clearly separated.
Thank you for the prompt. The output from the prompt you printed is perfect. However, when I enter your prompt I don’t get the same output, I get the following:
Thanks for getting in touch. I’m expecting the output that polepole is getting from his prompt. When I enter the exact same prompt I get the output shown in my reponse below.
I’m using chatGPT 4.0
Side question that might help those of us helping you.
Do you have ChatGPT Plus and are you using it? I am thinking that ChatGPT Plus has some features such as searching the internet and that is pulling in the companies in that list.
I used GPT-4o on my mobile phone, and I did not ask real company because of privacy of the real companies. I just asked 5 hypothetical companies to show how the output looks.
What is the diffrences on your output? What result want to see, can you clarify please?
Hi polepole, sorry if I’m not being clear.
I just would like to understand why I’m not getting the exact same output you’re getting - even though I’m using your prompt?
Based on AI’s training data, when it creates hypothetical scenarios, it sometimes uses real company names to make the examples seem more realistic. However, the descriptions about how these companies use data engineering and analytics are made up.
The purpose of this approach is to provide a scenario that feels real, while the specific details, like efficiency improvements or the use of certain technology, are invented. This helps show what similar companies could do, even if the actions described aren’t actually true for those specific companies.
In the output, the companies mentioned, showing whether they are real or fictional:
Arbor Wealth Management - Real (but the detailed scenario is fictional).
Granite Ridge Capital - Fictional (a similar name exists, but not as described).
Evergreen Asset Partners - Fictional (a real company exists, but with a different focus).
SummitBridge Capital Management - Fictional (no real company by this exact name).
Riverstone Investment Group - Fictional (similar to the real “Riverstone Holdings”).
Hi everyone, this was an interesting back-and-forth that I came across.
Why aren’t we thinking of taking help from other AI models?
For example, for prompting an Assistant or GPT-4o completions, I heavily use Claude and Perplexity to break down the tasks.
Manual first approach
Also, @carlton, I wanted to check with you whether the info you’re looking for is available in public knowledge with high reliability. So, have you been able to use perplexity, Bing and Google to find out such 5-10 examples and verify them? My hypothesis is this: if I can’t get something reliably manually, I wouldn’t count much on the GPT’s help.
to the community!
