Introduction

In today's business world, companies are faced with the challenge of analyzing huge amounts of data and extracting valuable insights from it. Especially in ERP systems, which store all business data centrally, it can be tedious to extract relevant information quickly and efficiently. What if you could just ask a question and your system gives you the answer in natural language? This is where the integration of ChatGPT/azure GPT comes in — a transformative solution to make your ERP system more intelligent and user-friendly. In this blog post, we show how sales analysis in an ERP system can be significantly simplified by integrating ChatGPT/azure GPT.

The scenario

Imagine a sales manager wants to know:

• “What was the turnover last month? ”
• “Which products account for the largest share of sales? ”
• “Show me the sales trends of the last three months. ”

Normally, such requests would have to be answered either via complex reports or SQL queries — time-consuming and often not intuitive. With the integration of ChatGPT/azure GPT, the user can ask these questions in natural language and receive directly usable results.

How does integration work?

1. Connect to the ERP system's API

The first step is to integrate the ChatGPT/AzureGPT API into your ERP system. This involves setting up a service that acts as an intermediary between ChatGPT/azure GPT and the ERP system. This service accesses the ERP system's database or API to extract sales data.

Example: A method in C# that retrieves sales figures for a specific period of time:

public async Task <List <SalesData> > getMonthlySalesData (DateTime month)
{
Return await _dbcontext.SalesData
.Where (s => s.saleDate.month == month.Month && s.saleDate.year == month.Year)
 .toListAsync ();
}

2. Natural language processing through ChatGPT/azure GPT

The user's request—for example, “What was the revenue in September? “— is transmitted to ChatGPT/azure GPT. The model recognizes the intent behind the request, extracts relevant parameters (such as the month “September”) and forwards them to the sales analysis service.

3. Returning and processing the results

After the relevant data has been retrieved from the ERP system, ChatGPT/azure GPT prepares the answer and returns it in an understandable and user-friendly form.

Example of output:

• “Sales in September 2024 amounted to €150,000. The products with the highest turnover were: Product A (50,000€), Product B (40,000€) and Product C (30,000€) . ”

4. Contextual follow-up questions

A major advantage of this integration is the ability to retain the context of the conversation. For example, the user could ask, “How was the revenue compared to the previous year? “ChatGPT/azure GPT recognizes the reference to the previous question and provides the comparison directly.

Why ChatGPT/azure GPT for your ERP system?

1. Intuitive user interface

With ChatGPT/azure GPT, users do not have to complete training in using the ERP system or create complex reports. A simple question is enough.

2. Save time

The automated analysis and provision of data saves valuable time — for both managers and operational teams.

3. Contextual intelligence

Unlike traditional searches, ChatGPT can retain the context of a conversation, making follow-up queries easier.

4. Flexibility

Integration isn't limited to revenue analytics. ChatGPT can easily be extended to cover inventory levels, customer analytics, financial reports, or other areas of the ERP system.

Where is my data?

Data processing when using the ChatGPT API is carried out via the OpenAI or Microsoft servers. The following should be considered:

1. Data transfer

• Encryption: All data is encrypted during transmission (TLS/HTTPS).
• Minimal principle: Only necessary data (e.g. aggregated information instead of raw data) should be sent to the API.

2. Server locations

• OpenAI states that the server locations are operated within trustworthy regions (such as the USA or EU). For companies in the EU, additional measures may be required to ensure GDPR compliance.

3. Data processing and storage

• By default, OpenAI does not store data long-term. However, it is possible to activate data processing without saving it in the settings.
• Companies can check whether an on-premise solution (local installation of AI on their own servers) or a secure virtual private cloud (VPC) operation is an option to keep data completely under control. 
  

How can authorizations be mapped?

A key requirement is that users only have access to the data and functions that correspond to their role and authorization in the ERP system. This can be implemented as follows:

Integration with existing authorization systems

• The ERP system controls which data and functions a user can view or execute. These existing authorizations are adopted when integrated with ChatGPT/azure GPT.
• The API interface between ChatGPT/azure GPT and the ERP system checks the user authorizations before submitting the request.

Example of eligibility checks

• A user with the “Sales” role can only access sales figures for their region. A request such as “What was the company's total revenue? “is rejected or adjusted to provide only the relevant data.

C# code sample

public async task <string>getSalesData (User user, DateTime period)
{
If (! user.hasPermission (“viewAllSales”))
{
 return “You don't have permission to view this data. “;
}
 return await _salesService.getDataForPeriod (period);
}

Logging and monitoring

All API requests and responses generated by ChatGPT/azure GPT can be logged to ensure that usage remains comprehensible.

Data security recommendations

1. Avoiding sensitive data transmission

If possible, do not send raw data such as customer names, addresses, or complete transaction details to the ChatGPT API. Instead, use aggregated or anonymized data.

2. Use of encryption

Make sure that all data is encrypted during transmission and processing.

3. Check GDPR compliance

Review legal requirements, particularly when personal data is being processed.

4. Consider an on-premise solution

For maximum control and security, an AI instance could be run locally on your servers or in a private cloud.

Conclusion

Connecting ChatGPT/azure GTP to your ERP system revolutionizes the way companies access their data and make decisions. Instead of complicated queries and reports, you can simply ask — and your system provides the answers. If you are looking for a way to make your ERP system more intuitive and powerful, integrating ChatGPT/azure GPT is one of the most effective solutions. Try it out and see how easy sales analysis can be!

‍