This is a chapter from #PRstack: AI tools for marketing, media, and public relations, the latest #FuturePRoof publication. You can buy a copy of the complete book from Amazon in Kindle and print formats. The Kindle version doesn’t include images due to format limitations.
Advanced Data Analysis plug-in by James Crawford | Practitioner
The OpenAI Advanced Data Analysis plug-in allows practitioners without coding skills to analyse data and gain insights by writing in simple English prompts. This feature can help automate tasks like press coverage analysis, audience research, measurement and evaluation, and report generation that were previously time-consuming or required data expertise.
For the past 20 years the public relations industry has seen countless talented account handlers glaze over when faced with talk about purchase order numbers, accounts or data. Sadly, the biggest barrier to making public relations agencies and teams more data driven is data literacy.
The OpenAI Advanced Data Analysis plug-in has changed all that. It has recently been renamed from Code Interpreter. It is still a beta product.
The Advanced Data Analysis plug-in allows users to execute Python code in a sandbox environment. This means that users can write and run Python code without having to install any additional software or libraries.
Most people probably got into public relations for creativity, not coding, but, here’s why python code is relevant to practice and can help bring creativity to the fore.
Python code is a set of instructions that tells a computer what to do. It is written in a language that is similar to English, but with stricter rules and syntax. Python code is used for a wide variety of tasks, including data analysis, machine learning, web development, and automation.
The Advanced Data Analysis plug-in allows you to write Python without really knowing the programming language. Not many agencies have access to a data analyst or a programmer who can programme in Python, but now, thanks to the Advanced Data Analysis plug-in they do.
The case uses for the Advanced Data Analysis plug-in are endless, but let’s look at it in layman's terms. Imagine an account executive has been asked to look at the brand visibility of a client and has downloaded some search data that shows the branded search volumes per month. The account executive wants to do some simple analysis to look at trends, year-on-year and against seasonality too.
The account executive needs to simplify the data into a CSV and into simple columns with no arbitrary text apart from the titles (the Advanced Data Analysis plug-in currently doesn’t like formatting – it just won’t work). Then while uploading it the account executive should describe using words what the spreadsheet says when uploading.
Once, ChatGPT says it understands the document the account executive can simply ask questions, using very simple language, like ChatGPT is a data analyst.
I first like to start really broad with a question like:
What can you tell me about any trends over time? What can you tell me about seasonality? It first comes back with a description of the data and asks permission to look at it in more detail, to which tell it to proceed.
This is what came back.
Time-Series Analysis for "Client A"
Observations:
• Trend: The trend does indeed show a general decline over time, with a slight uptick in the most recent quarter. This suggests that the branded search traffic for "Client A" has been decreasing but may be showing signs of recovery or stabilisation.
• Seasonality: Clear seasonal patterns are observable. Peaks often occur around March and troughs around December.
• Residuals: The residuals are the deviations from the trend and seasonal components. These seem random and don't indicate any additional pattern.
Would you like to delve further into any specific aspect?
I then can ask it to look at the last three months in greater detail as this just so happens to correlate with our own public relations activities. We are particularly interested in the year-on-year comparison.
What it comes back with is significantly quicker and better written than most non analysts could do and significantly quicker than even a well paid analyst.
Year-on-Year Comparison for July, August, September
July, August, September 2022:
• Mean branded search traffic: 263.3263.3
July, August, September 2023:
• Mean branded search traffic: 450.0450.0
Analysis:
• Year-on-Year Percentage Increase: There is a 70.8970.89 increase in the mean branded search traffic for July, August, and September 2023 compared to the same months in 2022.
Interpretation:
• The branded search traffic for these specific months has shown a significant year-on-year increase, which aligns well with the broader year-on-year comparison.
Impressive. I can very quickly see the year-on-year increase and describe it to clients professionally.
Other applications in public relations workflow
The example above was deliberately simple but any public relations practitioner with an imagination can now put this tool to good use. Here’s a few topline suggestions.
• Press coverage analysis
The OpenAI Advanced Data Analysis plug-in can be used to analyse press coverage data from Google Data
Sheets that are saved as CSVs. This can help practitioners to identify trends in coverage, identify key influencers, and measure the impact of their campaigns.
• Audience insights
It can be used to gain insights into target audiences by analysing data from GWI Index, Google Analytics,
keyword data, and SEMrush. This information can be used to develop more targeted and effective public relations campaigns.
• Measurement and evaluation
The Advanced Data Analysis plug-in can be used to measure and evaluate the results of public relations
campaigns by analysing data from Google Analytics, keyword data, and other sources. This information can be used to identify what is working well and what needs to be improved.
• Research summaries and insights
It can be used to generate summaries and insights from public relations research data. This can help practitioners to save time and produce more effective reports.
How to use the OpenAI Advanced Data Analysis plug-in
Getting started is simple. To use the Advanced Data Analysis plug-in you will need to:
1. Create a ChatGPT account.
2. Upgrade to ChatGPT Plus.
3. Enable the Advanced Data Analysis plug-in feature in your settings.
4. Create a new chat and select the Advanced Data Analysis plug-in .
5. Upload your data file (e.g., CSV, Excel, text file).
6. Enter a prompt to instruct ChatGPT on what you want it to do with your data.
7. Generate the output.
The OpenAI Advanced Data Analysis plug-in is a powerful tool that can be used by public relations practitioners to automate a variety of tasks and gain valuable insights. It is not perfect and it does get things wrong so some work does need to be scrutinised but it is, for me, an essential tool for professionalising evaluation and measurement across account teams so data can exist and be used outside of the data and analytics teams.
James Crawford, managing director, PR Agency One
https://www.linkedin.com/in/jameswdcrawford/
James founded PR Agency One, which since its launch in 2012 has put creativity and data at the heart of what it does. PR Agency One’s proprietary evaluation system OneEval predates the AMEC framework and has since been developed into three distinct products that focus on three areas Commercial, Reputation and Brand.