You've probably seen the headlines: OpenAI just dropped "ChatGPT Agent," a tool that can autonomously handle multi-step tasks like analyzing competitors, creating slide decks, and even booking your travel.
But, this is just another agent, right? Sort of.
This, along with Claude’s “Computer Use” model, are agentic, but this represents another leap forward at making agents that can use your computer to do tasks far more accessible. The new tool from OpenAI is a combination of their previous “Operator” tool and the “Deep Research” tool.
This isn’t groundbreaking on its own, but these tools are becoming more common, more powerful, and more accurate.
Why These New Tools Matter to Analytics Professionals
Instead of viewing this as just another chatbot or a Zapier-like workflow tool, think of it as an analyst’s assistant that can actually operate a computer. It can see the screen, use input devices, navigate websites, run code, and synthesize information.
Data Scientists/Analysts:
This technology can potentially eliminate the most tedious parts of our workflow. While I’m not saying these tools are 100% perfect, they are making big leaps.
Here’s a few ways I can envision their use:
- Automating the Grunt Work: Imagine telling an agent, "Run my saved ‘Daily Sales Analysis’ query, export the data to a spreadsheet for me and schedule a meeting with Dave to discuss it. While you’re at it, attach the file to the invite and create a pivot table with (fields)." The agent could do that, and you could schedule it to run this workflow every Monday. This frees you up, allowing you to return to high-impact interpretation and strategy.
- Rapid Prototyping & Analysis: Need to quickly build a competitive analysis deck for a meeting in an hour? The agent could create an editable first draft, complete with charts and summaries. Your job then becomes refining and adding the strategic insights that only a human can.
- Lowering the Technical Barrier: These agents can handle complex tasks by prompting with natural language. This means you could spend less time wrestling with a specific API or coding a complex data-cleaning script and more time solving the business problem.
Analytics Leaders:
This can be a massive force multiplier for your team.
- Unlocking Team Capacity: Your analysts will no longer be bogged down with repetitive, low-value tasks. They can focus on strategic projects that drive real business value. You get more output without increasing headcount.
- Shifting Skill Requirements: The most valuable skill is no longer just the ability to do the analysis, but the ability to define the problem and critically evaluate the AI's output. This also allows your non-technical stakeholders to self-serve in some capacities.
- Accelerating Insights: The speed at which your team can conduct research, analyze data, and generate reports will increase dramatically. This creates a significant competitive advantage.
How You Can Use This Today (and Tomorrow)
While these tools are still getting better and more accurate, this points to where the industry is heading.
Here’s some things to consider to be a step ahead:
- Become a Master Question-Asker: The quality of the agent's output is directly tied to the quality of your prompt. Utilize the CREATE framework or other prompt engineering strategies.
- Double-Down on Subject Matter Expertise: An AI tool can give you a summary, but they still aren’t giving you the final insights, nor do you want to blindly trust them. Your strategic, industry-specific knowledge is more valuable than ever.
- Be the "Human in the Loop": These tools can make mistakes. The best analysts will be those who can effectively guide, validate, and correct the work of their AI agents. Still, even with having to correct, the output of your work will be significantly higher.
In our recent webinar, we talked about some of the challenges with these tools and what’s coming next. Watch the recording if you missed it.
.jpg)
