This Ai Agent can research your new leads and update your CRM (no code)
In this tutorial Rory Ridgers demonstrates how to build a no‑code AI agent that automatically researches incoming leads and updates a CRM (Airtable) with enriched information. The solution combines Relevance AI’s Website Researcher, LinkedIn Researcher, and Update Airtable tools, a Make.com scenario, a Typeform lead capture form, and Airtable for storage. Leads submitted via Typeform (name, business name, website URL) trigger a Make.com webhook that passes the data to a Relevance AI agent. The agent first uses the Website Researcher to scrape and summarize the lead’s website, then the LinkedIn Researcher to locate the lead’s LinkedIn profile via a Google search, scrape profile and recent posts, and generate a LinkedIn summary. Finally, the Update Airtable tool sends all collected fields (including optional LinkedIn URL, LinkedIn summary, and company summary) to a Make.com webhook that creates a new record in Airtable via a JSON POST. The video walks through each tool’s configuration, shows a live example with Ben Francis of Gymshark, and highlights the time‑saving benefits of having enriched lead data ready for outreach. The creator also promotes his Commonus marketplace for sharing templates, mentions future use of the NAN platform, and invites viewers to request custom AI agent implementations.
Chapters
What's up guys? In this video I'm gonna be showing you how you can create an AI agent that can research all of your incoming leads for you and update your CRM with that new information. It's quite a simple agent, but it's gonna save you so much time, especially if you're getting in loads of different leads and you're spending loads of time researching them to then prepare you for your upcoming meeting with them or your outreach email to them, whatever it might be. I just wanna mention at the start of the video that all the templates that I made in this video will be available on my new platform, Commonus. If you didn't see my last video announcing that I've come out with this new platform, Basically what it is, is a platform made by me. Anyone could go and create and download templates made on Make.com, Relevance AI, NAN, any platform you like. You can go ahead, put it on there, and other people can download it. And you'll be able to promote your agency, your software, your product, and the right people will be able to see it. Also, if you haven't seen me, my name is Rory Ridgers and I help business owners to integrate AI agent systems into their businesses. So that's something you might be interested in, then I've left the link to my website in the description. You can you go ahead and book in a call or just fill out the contact form and I'll get back to you. Let's not waste any more time. Let's get into the video and I can show you how it all works. So in this video, we're going to be using Relevance AI, a few scenarios on mate.com, Airtable to actually store this lead data, and then Typeform to collect these leads. So first of all, on Relevance AI, we're going to have 3 different tools. We have the website researcher tool, the LinkedIn researcher tool, and then the update Airtable tool. And what each of these tools are going to do is quite self-explanatory. So the website researcher tool is going to research the lead's website and give us a summary and a breakdown of their site. The LinkedIn Researcher tool is going to find their LinkedIn account, it's going to look at their posts and their profile information, and it's going to create a summary on their LinkedIn. It's also going to output the LinkedIn URL for us so we can put that into our Airtable database. And then the Update Airtable tool is just going to put all of that data into the Airtable database. I'll show you the Airtable database just so you can see. So we have name, business name, website URL, LinkedIn URL, LinkedIn summary, and company summary. And name, business name, and website URL. That is what's going to get passed through from the Typeform. Because if we look at this Typeform, I can show you it quickly. It's just 3 different questions. So we have name, business name, and then website URL. So that's what's going to get past the AI agent, and that's all it's going to use to be able to create the summary, find the LinkedIn URL, and then create a summary on their LinkedIn. So I'm going to show you how it all works. First of all, we come over to make.com. We have this simple scenario that's just going to trigger the AI agent. So really, really simple. It's just a Type 4 node. We have hooked that up to that form that I just showed you. We're getting the full name, the business name, and the website URL. And then we're simply passing that over to our agent on Relevance AI. So the role is going to be user. The content is the full name, the business name, the website URL. Then we've chucked in our agent ID in there. And then the first tool this agent is going to use is going to be the website research tool, because it's probably the most simple. The only input is going to be the website URL that we got from that Typeform. That's going to get put into this tool. We're going to extract the website content. So we really simply just put in that variable, the website URL. The method is going to be text. You can do HTML if you like. I've just done text. And then we have an LLM step here that is literally just going to go through that scraped data and output a summary for us. And then we've given it the variable that was outputted by this step. So it's just going to be scraped content. It's just going to go through that, create a summary, make it look nice. Really, really simple tool, but it's just nice when you get a lead come through to be able to just look at a summary of their website. It's just something that's really, really handy. So we have that tool. Actually, I want to show you this tool in action just so you can see. We're going to do a random URL. Let's say Gymshark. So I've said www.gymshark.com. Hit run tool. It's a really, really simple tool. This step is literally just going to extract all of their website content and then it's going to get passed over to the LLM, which is going to summarize that website content. So we can have a look here in action. Okay, so this is what it's given us. So brand focus, their promotions, some of their categories, community engagement, motivational messaging, content and resources, brand heritage. So really, really nice, super simple tool, but it's really handy, as I said, to when you get a new lead in, just to be able to see a summary of their website. It's just so handy. Now we're going to look at the LinkedIn Researcher tool. This one's a tiny bit more complex, but still nothing crazy, so don't worry. The inputs here are going to be name and company name. What we need to do first is we need to actually find their LinkedIn URL. So we do a Google search for the name's LinkedIn, and then in brackets we put company. So again, let's use Gymshark as an example. The name might be Ben Francis, is the CEO of Gymshark. So name might be Ben Francis, and the company name will be Gymshark. So the Google search will look like Ben Francis's LinkedIn, and then in brackets Gymshark, just so we know which Baron Francis we're talking about. But obviously that works for any lead. We put in their leads LinkedIn and then in brackets their company name, just so we know we've actually got the right LinkedIn URL. An LLM step is just going to literally go through those Google search results and output just the URL for us, because it's going to output a bunch of search results. We just want the URL that is most likely the URL for our leads LinkedIn. So that LLM step is going to go through all the Google data and output just the URL for us. And then we do a get LinkedIn profile slash company and post, and we put in the answer from this LLM, which hopefully should be our LinkedIn URL. And then we put the method to get user posts, and then we do the exact same thing but get user profile. So we're just going to scrape their posts and then scrape their profile. And then we have another LLM step that is just going to basically look through both of these pieces of data, which is going to be their posts and their profile, and it's just going to create a summary for us. So So we wanted to look through the profile data, create a summary on that, and then look for the post data and give us a summary on their 5 most recent posts. And then we give it the LinkedIn data and the LinkedIn post. So again, we're going to test this out. Let's just say— let's just use the example I used before, Ben Francis and Gymshark. So run tool, and then we should see it find— okay, so this is what it gave us, the Google search step. Gave us a bunch of LinkedIn URLs. And you can see we got a LinkedIn URL that the LLM spat out. And then the get user posts. Let's have a look at that. If we go to data, you can see these are all their most recent posts. And then the profile data. And you can see this is what the last LLM step outputted. So we've got a full name, current position. So Ben Francis, CEO and founder of Gymshark. Their LinkedIn profile, connection count, follower count, education, professional experience, uh, and then recent post summary. It's given us a summary of his 5 most recent posts. And having all of this just in your CRM, or in this case in an Airtable database, when a new lead comes in— for all of you that get quite a few leads in, you'll know how valuable this is, not having to spend all that time researching these different leads to be able to prepare for calls coming up with them or outreach emails, like I said. It's just such a lifesaver. It saves so much time. So quite a simple tool, but really, really useful for a lot of you guys. Then the last tool is just going to be Update Airtable. If we have a look at that, you can see the inputs are just going to be name, business name, website URL, LinkedIn URL, LinkedIn summary, and website summary. So obviously all of this data, the first 3 ones, gets passed through from the Typeform, but these other ones are just going to be the outputs from the tools that I just showed you. And you can see we have set this from required to optional. So these 3 are required because obviously we get them through the Typeform, so they're always going to be there. But these ones, we might not find them. There might be— if it's, say, a lead that comes in that doesn't have a LinkedIn, because not everyone has LinkedIn. Most people have LinkedIn, but not everyone has LinkedIn. So sometimes it won't be able to find them. So we set these to optional because if they're set to required and the tools before, for instance, can't find the LinkedIn URL then it's just not going to be able to run this tool and it's not going to be able to update all these other data points. So we set these ones to optional and just tell the AI agent, obviously, if it does output something, then put it in here, but if it doesn't, then just ignore it. And then we just have an API request step here, and we're just connecting this to a webhook on Make to be able to actually update the Airtable database. So I'll show you that. So this is the Make.com scenario to update the Airtable database. Really, really simple. We literally just have a webhook. If you've never used webhooks before, we basically get this URL here and then we put that into the URL section on our API request and reference— they are set the method to POST because we want to send off data. In this case, the headers are going to be Content-Type: application/json because we're sending over JSON. The reason we send over JSON is because we have these different variables here and we want them to be inputted into make.com as separate variables rather than just one variable. So we send it over as a JSON object, and then in this case you can see the variable name in Make.com will be name, and then we put in the name variable. So that will then save the name variable, whatever the name variable is, to name when we use it in Make.com. So you'll see what I mean in a second. But yeah, if you ever wondered the reason why sometimes we pass it over in JSON, it's because we want these as separate variables. It's just much easier to use in Make.com. So anyway, we send that off to Make, and then we have just a simple create record Airtable step. We've hooked up our base and our table, and then we literally just input the different data points. You can see here is what I mean. We have different variables because we passed over in JSON, so we can just drag them in. So name, website URL— okay, I'm doing this twice now, but you can see really, really simple. We just drag in the different variables here into the correct column on Airtable. So really, really simple. I'm going to show you now how to put all of this into an agent. I've already made an agent called Lead Researcher. So I've given it a name, a description. This part doesn't really matter because we're not doing multi-agent systems, so you can put whatever you like in there. The tools are obviously going to be the tools I just showed you: the LinkedIn Researcher, Website Researcher, and Update Airtable tool. And core instructions, nothing too crazy here. All we've done is the role: your job is to receive new leads and research them and update the Airtable database. And task: your task is to receive incoming leads from a Make.com scenario. Obviously that's coming from Typeform, but as far as Relevance AI knows, it's just coming from an 8.com scenario. You'll receive a name, business name, website URL. You will use the website researcher tool to research their website and then the LinkedIn researcher tool to find and research their LinkedIn. I don't want to pass all of that information, including the outputs from the two tools, the name, the business name, and the website URL that you received, to the update Airtable tool that will update the Airtable database. So really, really simple. The agent isn't really doing too much here. It's just receiving an input using the two different tools and then passing all of it over to the third tool, the Update Airtable Database tool. So quite simple for the AI agents, not really having to do too much here. We should save changes and we're going to start this from start to finish so you can see how it all works. Okay, so we're on the Typeform now. Let's use the exact same example. We're going to say Ben Francis, just so you can see this all working. So Ben Francis, Jim Shark. And then the website URL www.gymshark.com. So the website URL www.gymshark.com. Let's hit submit, and then we're going to manually run this Make.com scenario. Obviously usually it'd be automatic, but just for this example we're going to manually run it so you can see it working. So let's hit run once, and you can see we have answers. If we go to one, so Ben Francis Gymshark, uh, www.gymshark.com, which is perfect. So triggered our AI agent. So if we have a look, yeah, so it's used the website researcher tool already. So you can see this is what we've got, kind of similar to the example I showed you earlier. Now it's using the LinkedIn researcher tool, and then we should see this update our Airtable database. Okay, so it's used both the tools, and now just to show you, the Airtable database is blank. We hit refresh. Yeah, you can see it's still not in there. Just so, just so you know, I haven't put it in there already. It is blank, so we should see it use the update Airtable database tool. You can see just used it straight away. Didn't even say running, it just said it's used it. So let's have a look. You can see Ben Francis, Gymshark, gymshark.com, LinkedIn URL. Let's have a look, make sure that's right. Looks right to me. Perfect. So come out of that. We got a LinkedIn summary and then we also have a company summary. So absolutely perfect. A really, really simple tool, but just having this in your CRM is so much more helpful than just having a name, a business name, a website URL. Obviously in this case, Ben Francis from Gymshark, might know who he is already, but 99% of your leads, you're gonna have no idea who they are. So just having a name, a business name, and a website URL, and having to manually look up and create a summary for yourself and find their LinkedIn, create a summary on their LinkedIn, everything like that, it's just so annoying. And when you've got loads of leads coming through, it just takes so much time, which means you're gonna be wasting money, wasting time. You don't want to be doing that. So just having their LinkedIn URL, a LinkedIn summary, and a summary of their company already there in your CRM as the lead comes in is so helpful. And it was pretty simple to build. And obviously templates are available for completely free on Commoners. So if you want to get started and use this straight away for free, you can go ahead and download that on Commoners. So that's it for this video. I hope you did enjoy it. Just to let you know, by the way, moving forward after this video, I'm going to be using NAN for majority of my AI agent builds. So if any of you guys use NAN or you're thinking of using NAN and there's some video suggestions you'd like to see, I really would appreciate it if you just drop them in the comments, send me over an email or some stuff that you'd like to see with NAN, and I'll try my best to get around to doing them because I'm still learning it. So it'd be really great for me to sort of learn it with you guys and try out some stuff that you want to see on there. Also, if you didn't see my last video and you haven't used Commoners before, please do check it out and tell me if there's anything you'd like to be improved. I have already had some suggestions for some improvement. So hopefully you'll see them very, very soon on Commonus. As I said, if you did want to work with me so we can implement some AI agent solutions into your business, then just check out the link in the description. You can go ahead and book in a call or just fill out the contact form and I'll get back to you. But that's it for this video. I hope you enjoyed it and I hope you found this AI agent system value, because personally I think it's amazing and it's really, really simple to build and just super, super helpful. So that's it for this video. I'll see you in the next one.
More from Rory
Dynamic Workflows are crazy. This is how to actually use them (beginner's guide)
