Perplexity: From search to source-rich research

- You are already using AI tools like ChatGPT, Gemini, and Copilot. But here's the truth, the real productivity breakthroughs don't come from finding the perfect AI tool. They actually come from combining the right tools together in the right way. In this course, you'll learn the exact mechanics of AI tool stacking. We'll build proven stacks for research, business analysis, sales communication, and more. You'll walk away with repeatable workflows you can implement immediately. So first off, when you need to research trends, competitor analysis, or market insights, searching Google and clicking through dozens of articles isn't just time consuming, but it's also inefficient. Now, two of my favorite tools that we can actually stack together here are that of Perplexity and NotebookLM. Now, we'll get onto NotebookLM in just a little bit, but for this video, let's use Perplexity as your research powerhouse. But before we dive in, I just wanted to mention that all the prompts we use in this course will be available to you on the course page. Now, that leads me to our first prompt. So, we're going to dive straight into Perplexity and paste in our first prompt here, which is, what are the real examples of companies using AI agents in 2025? We want to be focusing on specific company names and use cases, measurable results and outcomes. And ultimately, for the data for this prompt, we want to be just retrieving stuff from 2024 to 2025 only. Now, what I love about Perplexity here is you can actually turn on this deep research mode, and we're actually going to be using one of our three queries today. So it doesn't rely on you having a paid Perplexity account. And we're going to go ahead and kickstart this query. So first things first, what Perplexity is going to be doing is it, it's going to be doing this research task for you. So it'll take approximately three minutes. We can even turn on the notification bell here. So once the research is done, we can get notified. But this research task is up and running. Now, as we can see here, Perplexity is learning from a whole bunch of sources. We've got sources from Microsoft, LinkedIn, Deloitte, some really, really sort of key sources here, and it's actually forming this research answer that will then be presented to me as the user. Perplexity is really thinking deeply for this. We have the option of skipping the remaining steps if we just want to get right to the answer. But we have that notification bell enabled, so we're busy working. We can always get notified when that research report has been completed. Right, okay. So we can see that the answer has been produced to us. So we've got some great sources here in addition to those ones I mentioned earlier, the likes of Fortune and Salesforce have come in starting with the real world examples of companies using AI agents in 2025 all the way through to enterprise in implementations, JP Morgan Chase, bankings AI transformation, Microsoft Copilot, Klarna, some really brilliant examples here that we can really sink our teeth into. But if we go down to the bottom here, as we can see, it's still producing the answer. It's a very sizable piece of research that it is producing for us today. And what I love about this, it even provides a sort of looking ahead steps. So it goes, look, you know, the trajectory is clear, thinking about that sort of forward looking picture. But what if we want to be a little bit more pointed in our research and go that one step deeper? So let's paste in our second prompt here, which is based on those examples, those brilliant examples that Perplexity has just presented to us. What about the specific AI agent platforms or frameworks that these companies are using? And I've also given some examples here such as AutoGPT, Langchain Custom GPT Actions, and a few others. And we can go ahead and use our second deep research query available to us on that free account. And it will go ahead and be a little bit more pointed in that research answer. And as we can see, Perplexity is already starting to generate its answer here. And actually, it's highlighted really nicely some of the key platforms that these examples that we explored earlier, right at the top here that they're actually using. So if we dive in, for example, Salesforce Agent Force, it dives into the platform architecture, things like the pricing structure, and, also, some of the non-obvious things, such as hidden costs. So, us as busy professionals, you know, we want to be understanding that from a cost lens. So when we're bringing that to our manager, things like that, just being really, really able to be a discerning user of these tools. Okay, fantastic. And it even wraps up with some of the key trends in 2025, such as consumption based pricing dominance. And as we know with AI, you know, these LLMs run on APIs, costs can often skyrocket. So it's really important to keep that big picture on that cost. And if we even then look a little bit further forward, such as open source for core, managed for scale, in addition to industry specific solutions. So vertical specific platforms such as Thoughtful AI for healthcare. So really thinking out the box here when it comes to that sort of fundamental overview of what it is that we're wanting to explore. The final prompt that we can actually use here within Perplexity is more of a holistic one to wrap things up. So we've done an awful lot of summary from the examples to the actual frameworks that they're using. What if we actually want to now turn this into a structured summary of AI agent adoption within 2025? And so here, I've laid out some key sections, such as key trends, success stories, emerging platforms, common challenges, all the way down through to future predictions, which we did touch on in that first prompt as well. So let's go ahead and use our final free query for that and see what we get. Okay, brilliant. So we have our AI agent adoption in 2025 structured summary. So we've got all of those key points that we were looking for. So key trends through to the success stories, all the way down to emerging platforms as well as common challenges. And then finally, looking to those future predictions. And what I really love about this is Perplexity is brilliant at citing these sources. So if we click through here, we can see all the sources that each of these answers actually use to inform the result. And what I love about this, finally, is we can ask things such as rewriting it, but we are not going to do that. We're actually going to export this now. This will form the basis of our research that we'll then pull across to NotebookLM. But for now, let's just start and actually export this as a PDF, which is the perfect file type for NotebookLM Now that we've got that PDF, what we can do is we'll actually take this research into NotebookLM and actually transform it into these interactive insights that you can actually share with your team.