Over the past month, I’ve tested Deep Research extensively to evaluate its capabilities, its limitations, and how it fits into my world – helping smaller businesses position themselves for growth or acquisition. Here’s what I found.
What is Deep Research?
Deep Research is an AI-powered tool that can, in the words of OpenAI, “find, analyze, and synthesize hundreds of online sources to create a comprehensive report at the level of a research analyst.”
For those involved in M&A, business strategy, or senior-level recruitment, this type of research has traditionally required dedicated analysts. As an independent, I don’t have a team of junior associates to do the grunt work of building acquisition lists or sourcing market data. Deep Research looked promising as a solution.
How I Used Deep Research for M&A Insights
Working in the M&A space, one of your go-to-tools is a contact list of likely suitors for your customers. Could Deep Research help me out here? I asked it to identify acquisitive companies in the B2B e-learning space in the U.S. and return me a detailed report, alongside a summary table:
| Acquiring Company | # of Acquisitions (Last ~5 Years) | Key Acquisitions (2019–2024) |
| Cornerstone OnDemand | 4 (2019–2022) | Saba (2020), Clustree (2020), EdCast (2022), SumTotal (2022) |
| Skillsoft | 3 (2021–2022) | Global Knowledge (2021), Pluma (2021), Codecademy (2022) |
| Pluralsight | 3 (2019–2021) | GitPrime (2019), DevelopIntelligence (2020), A Cloud Guru (2021) |
| Udemy | 1 (2021) | CorpU (2021) |
| Docebo | 2 (2020–2022) | forMetris (2020), Skillslive (2022) |
| Degreed | 2 (2019–2022) | Adepto (2019), Learn In (2022) |
| Axonify | 2 (2021–2022) | MLevel (2021), Nudge (2022) |
| ELB Learning | 6 (2020–2022) | Trivantis, Edulence (2020), The Game Agency (2021), Rehearsal (2022), Origin Learning (2022), CoreAxis (2022) |
Beyond just providing a table of results, Deep Research generated a 10-page, 4,000-word report detailing the methodology behind its findings. This process, which traditionally could take a full day or require hiring an analyst, was completed in under 10 minutes.
Is Deep Research effective enough to replace other methods?
No AI tool is perfect, and Deep Research is no exception. However, my key question is: Is it more effective than traditional research methods?
Pros:
- It significantly reduces research time.
- Reports are comprehensive and include citations.
- Results are directionally accurate – great for a starting point.
Cons:
- Citations can be inaccurate. Some sources were misattributed, requiring manual verification.
- Time sequencing issues. The tool sometimes misinterprets when events happened, making it seem outdated. And, as the table above shows, its missing some of the more recent activities that have taken place (e.g. Cornerstone buys Talespin in 2024).
- Source selection is imperfect. Despite prompts to use high-quality sources, it sometimes pulls from biased or user-generated content (e.g., Reddit, G2 reviews).
Would a human analyst make mistakes? Absolutely. But in this case, Deep Research saves time while delivering comparable accuracy – as long as you fact-check its outputs. So yes, I think it is more effective, as long as you don’t think it’s once and done. It’s enabling me to scale my work in a way that would have been hard to do before.
The Rise of AI Optimization
A notable trend emerging from tools like Deep Research is what I call AI Optimization (AIO). Companies are now actively working to ensure their content gets indexed by AI-driven research tools – just like SEO for search engines.
For example, I’ve already seen businesses create subreddits with no user engagement purely to get their company or product included in Deep Research results. While AI tools will likely improve their ability to filter sources, this is an early indicator of how businesses will try to manipulate AI-driven research.
Comparing Deep Research to Other AI Research Tools
Deep Research isn’t the only game in town. Competitors like Gemini (Google), Perplexity, and Grok offer similar functions. Here’s how they stack up in my experience:
| Tool | Speed | Reliability | Detail of Response | Accuracy of Citations | Insightfulness |
| ChatGPT Deep Research | 4th | 4th | 1st | 3rd | 1st |
| Grok (X / OpenAI) | 1st | 1st | 4th | 1st | 4th |
| Gemini (Google) | 2nd | 3rd | 2nd | 2nd | 2nd |
| Perplexity | 3rd | 2nd | 3rd | 2nd | 3rd |
Deep Research excels in detail and insight but struggles with citation accuracy and speed. Grok is fast but lacks depth. Perplexity plays it safe, often refusing to provide an answer if it lacks confidence. Gemini is stable but lacks nuance.
For my work, ChatGPT’s Deep Research is the best fit – it produces nuanced insights that feel like working with a real analyst. But you might need to work on your prompts, include some guardrails and be prepared to ask a few times. Which can be a pain.
The Role of AI in Business Acquisitions
AI tools like Deep Research are incredibly useful for identifying potential buyers and market trends, but they don’t replace expertise in M&A strategy. Finding the right acquirer isn’t just about data – it’s about positioning, outreach, and negotiation.
Many small business owners struggle to navigate the acquisition process. That’s where my work comes in. With this sort of firepower available, I think you’ll see a rise in people like me, combining AI-driven insights with hands-on experience to help businesses get acquired by the right buyer at the right price.
If you’re considering an exit strategy, let’s talk.