Will Generative AI replace data analysts?

Generative AI are artificial intelligence systems that can create new content that did not exist before.

In this article I demonstrate a step-by-step scenario that shows how Generative AI systems can be integrated into the work of a data analyst and might even replace it in the future. This scenario might be fictional but it also might be realistic.

First step – Generative AI will write queries by itself

Currently (The start of 2023) one can describe a table to the Chat GPT system and ask it to write SQL queries on that table. 

This capability is still not useful in organization and mainly used by SQL students because Generative AI systems are not connected to the organizational data tables but there is no reason why in the near future data warehouses companies such as Snowflake or Azure will implement a built in Generative AI service that is connected to the organization data tables and can generate SQL queries by request.

If the DWH companies will add such capabilities, data analysts may have to acquire the skill of writing the correct Prompt (the input of the Generative AI system) that will generate the query they need for the analysis task.

Second step – Generative AI systems will create automatic dashboards

Every data analyst knows that answering business questions is just one part of the role. Another important role is to generate dashboards that will help managers understand the state of the company or product.

Will Generative AI know how to create dashboards on its own?

Dashboards contain graphs and KPI’s that suit the company needs, but although the dashboards are very specific to each company, dashboards  in general are quite similar. The main difference between companies in their dashboard is that they rely on other data sources. Most dashboards have KPI’s of revenue, conversion rate, Churt rate, customer value, LTV and so on and charts to describe these metrics.

If the Generative AI systems will be connected to the organizational data tables, there is no reason why they should not calculate the common KPIs that companies use and create dashboards based on the company data.

Even today there are Generative AI systems that know how to create automatic dashboards and there is no reason why this ability will not improve in the future.

Third step – Generative AI systems will answer business questions

“Hey Chat GPT, can you tell me why customers are churn”

If Generative AI systems would know how to write queries based on a company’s data tables without a data analyst in the near future, would they also know how to answer business questions?

If currently Chat GPT and similar systems know how to answer general questions very well  in natural language, imagine what would happen if such systems will have access to the organizational data tables. Will they be able to answer business questions instead of a data analyst? 

If Generative AI systems reach such capabilities, will there still be a need for data analyst analyses?

These are all open questions at the moment. I don’t know the answer to them.

Fourth Step  – The Generative AI system will produce automatic insights

Here is an example of a sentence that might be produced to managers at the beginning of a working day:

“Good morning, in the last two weeks there has been a decrease in the registration of new customers in Europe. I tried to find out what causes the decline and it turns out that it happened due to a campaign called “Sales for the European Market” which failed to target the right users. There is also a possibility that the war in Ukraine has a slight effect on the decrease in the number of new users.”

Would it be possible to imagine a future in which Generative AI systems will know how to create business insights on their own such as the one described above?

Based on all the previous steps we’ve seen, the answer is probably yes.

If Generative AI systems will connect to organizational data tables, write queries and answer business questions, there is no reason why they could not also produce interesting business questions by themselves and provide answers to these questions. 

In fact, even today there are systems that know how to find anomalies in the data and derive automatic insights from them. For example, such systems exist in the tools Google Analytics, and Microsoft Power BI.

Let’s dwell for a moment on Microsoft in that future:

Microsoft is a business partner of the Open AI company that created the Chat GPT tool. Microsoft also owns the tool Power BI in which one can analyze data, create dashboards and generate automatic insights. The business relations between Microsoft and Open AI can yield an interesting  connection between Chat GPT and Power BI. Such a link can lead to the vision of generating automatic insights from data in the near future, much faster than we might think.

Conclusion – Will there be data analysts in the future?  

In this article I have described a scenario in several steps that shows how the role of a data analyst may change or might even disappear. Is the scenario imaginary or realistic? I don’t know, but I believe that the emerging field of Generative AI will strongly affect the role of a data analyst as it will affect many roles in many industries.

This article was written by Yuval Marnin.
If you have a need to hire a freelancer data analyst you may contact me at: [email protected]

You may also hire me through upwork platform on that link:
https://www.upwork.com/freelancers/~018940225ce48244f0\

More articles on the s blog

The advantage of hiring a freelance data analyst.
What does a data analyst is doing and how it can help your company.