Thank you, guys. Nice to be here.

Thank you to Jorge. Thank you to the GBO team.

Introduction

So today is my technical. I'm going to do the technical talk. I'm trying to make it so much technical, so I'm going to introduce it a little bit.

And today's topic is going to be basically how to sell more. How do we bring this to AI? We bring it with different stages of implementing AI automated systems for B2B client acquisitions.

OK?

Why AI for Client Acquisition?

Why AI client acquisition?

This is like an AI talk. Obviously, AI is talked about a lot nowadays.

And my job here is to see how we bring it back and how we really generate that real business value for everyone here.

Speaker Background

So how did I come to work with AI acquisition systems? To give it a bit of background, let me just try to... To give a bit of background, I've been in this AI space for almost the last four years and I started taking my first steps into AI when I had my own company.

It had nothing to do with AI, it was actually like an e-commerce and in this e-commerce I took the operational role. In this role, I started digging into a bit of AI when the first GPT models came out, and I started automating my own operations processes. That being said, I kept going until I had almost the whole operations part of this e-commerce automated.

I ended up selling that e-commerce and I went into this beautiful field that is AI automation and I started working with a company in the healthcare system. I worked with them giving out services to automate healthcare processes.

This was really cool and I decided I had to come out of this company and start my own company because I love this. And I started my own AI automation and implementation company.

So being that said, when you...

From Hype to Business Value

try to sell automations like i mean ai is like we've got a lot of hype like everyone loves ais everywhere everyone's making pictures of their dogs surfing but that doesn't really bring real business value right so me being in the business of selling ai i need to for my clients to see the real value in ai if not they're not gonna buy from me right so that's where uh the main focus of companies and clients need to see the real return on investments for their AI projects. If not, they are not really going to invest.

I don't know what's happening with the TV. In order to another real return on investment, they also need to know what's happening.

That's why nowadays in the Spanish market, that's a bit of a greener market, the biggest services being sold right now are more like consultancies, because people don't really know what they can do. So that's a good part.

Maximizing ROI with AI

So that being said, companies need to know the ROI, the return on investment, for their AI projects.

And how do we maximize this ROI? There's two main tips I'm going to give you today to maximize the ROI, the return on investment on AI projects.

Two Types of AI Automations

One is going to be that there is two type of automations. or two type of AI automations when we try to bring AI into our companies.

There's the AI automation that That's a process we were already doing, so automates and relieves hours from our work. So there's something we were already doing, and now AI does it.

And there's the other type of AI implementation, where AI comes and does new things. So it amplifies our business volume. And that is the main point where we optimize for ROI. Because not only automates and makes us spend less hours, it also brings bigger and grows our business.

So that's one of the places we should focus.

Focus on Revenue-Proximate Processes

And then the other thing is there is a lot of business opportunities, business processes, business systems where we can bring AI. 1And to see real value, the closer the solution we bring is to the revenue generating part of the business, the better. What does this mean?

So normally in the business, you've got this scale where you've got the money you're bringing home, basically. And there's different parts of your business.

You've got sales. This is the part that brings the money right away. You sell one service, you bring X amount of dollars. There's like a relation of one to one. You can sell two services, X amount of dollars.

Then there's the... then there's a service. You sell a service, it's also really related to the money coming in because you sell one service, and then X amount of money comes in.

But then it comes like more admin and human relations kind of processes and automations that don't really have that kind of one-to-one relation to the money coming into the business. And as you all know, businesses invest in what brings money in. To maximize ROI, the key idea is to focus on these initial processes so people really see the value behind those systems.

Choosing the Right Sales Channel

That's why we focus on sales systems. That's why sales is the best place to implement AI for visible ROI.

But where in sales makes the most sense to implement ROI. There's a lot of areas.

There's three main client acquisition systems, ads, content, outreach. And if we think about it, when we started to offer these services, we were thinking like, where should we focus? Because there's a lot of... area, and we need to niche down and choose one of them.

Why Not Ads or Content

And if you think about it, ads is like a space where there's a lot of volume, but there's a lot of budget as well. And right now, we can't really trust AI with big budgets. It can do weird things, and we wouldn't like that. So we discarded ads.

Then there's content. Content is really good. AI is every day better at making content, video content, content, image content, text content. the problem with content is not that measurable. You do not really know if the sale is coming from this channel or TikTok or Instagram. So we also ruled it out.

Why Outreach Wins

So we ended up in outreach. Outreach, basically, reaching out to people.

We started it and it's like more or less cheap area of sales and also really measurable. You know how many people have you contacted, you know how many from those people, how many have gone to the meeting and how many of them you ended up closing the sale, right? So we decided to go with outreach.

Stages of AI in Outreach

So once you make that decision, there's different stages of AI adoption in the outreach systems. I made up four different stages of AI adoption in outreach sales systems.

And I'm going to go into the more technical part. Let's see if we fix this, because I'm going to show some things.

Version 0: Standard Automated Sequences

The first one is not AI at all. The typical standardized automated sequences where you establish a sequence and it sends a series of messages to each prospect, to each potential client, and it's always the same. This is version zero without AI.

Version 1: Personalized Messaging

Then we start bringing a bit more AI and we start bringing some message personalization to this standardized automated sequences. So this is still like an automated sequence, but the actual message, the actual text is personalized to each prospect.

This brings us more value because we have more conversion rates because people really respond more.

And I'm going to show you I'm going to show you a quick demo.

Well, first, here I have a mirror with what you could see is a cell system. This is like a funnel. I'm not going to really dig into it because it's a bit complex. But you could see this is like the, sorry it's in Spanish for everyone, but this is an internal thing we use, so I wasn't going to translate like

all of this and all of this. So this is where you can see how we join the automated messages. We have an email sequence. We have a chat WhatsApp.

We could also have an SMS sequence. And the traditional way would be to each of these messages would be a templated message. If we bring AI, we can start personalizing all these messages and all these touch points we have with the client. We personalize it exactly for the client.

So I'm going to show you a really small example of how we could do this. OK. So I don't know if you guys know Make. It's like N8n, like this really popular note code platforms.

And this is an example of how we could personalize some messages. For example, we have, this is like, list of different prospects. And we've only got their name, their LinkedIn profile, and a few more data like job title, company website, et cetera.

So we can process this through the other automation I probably showed. And for example, this is really dynamic. So you could build it as you want. But for this example, this is an example we've also actually implemented in clients.

We can get their profile info. This has been their LinkedIn profile info, like their whole description and everything. We can also get their profile posts, like all the posts and interactions they've done through LinkedIn. This is just joining them together.

And then we can also get the company posts, like all the activity of the company through LinkedIn. And then after that, we join them again, and then we can crawl their website. We use FireCrawl, a really popular crawling system, and we can crawl the whole information through their website. We can also use Perplexity to do an AI research on the website, on the person, to see if this person just had maybe an upgrade on their contract, or maybe the

the company just came public or something like that so we can reference that and we have like this huge context of information before going going like talking with the person right then we obviously pass it through our lovely ai we write the linkedin copy we write the email copy and then we uh put it back on our database. So here you see the yellow titles is the input data.

And then the ones in green is all the information we've got through the header, the about, the location of the LinkedIn profiles. And then if you go to the right, you've got the LinkedIn personalization and the email personalization. And you can see, if you don't know Spanish, here it says, I've read your post about the difference between. So it makes direct.

direct reference to actual things happening in the lives of the people. So there are much more probability that they respond and they start to talk with you, right? So this is like version one. We were talking about the different versions.

This is version one. Where were we? This is version one, standardized sequences, plus messages, personalization. And then we have version two.

Version 2: Personalized Video at Scale

This is the same, but we bring something really, really cool. even more possible nowadays, that's not only message personalisation being text, but also video personalisation. There's tools where we can record the video and in the part where we say the prospect's name, with the tools like voice cloning and dubbing tools, we can replace that exact part with any name we want.

So, for example, I'm going to give you an example of a project we did with a client where one part of the funnel was after they book the call we send them via whatsapp a selfie like selfie style video of saying hey name I love to talk to you on Thursday something like that so with AI we actually the standardized video we actually like cut it up and personalized personalized it for every prospect that was really cool and also skyrocketed the conversions and the interactions

Version 3: Full Outreach Agent

Then we have a full outreach agent. This is really cool, and this is something we thought internally in our company that we needed to go a step further. Because the problem with the first ones is once the prospect responds to you, it ends there. Because you cannot keep a constant conversation with them.

And we actually, right now, we are working on an MVP of of a SaaS, I'm going to show you guys. Have in mind, the UI is not definitely an MVP of something we're working on.

So we can take the prospect from the initial list to the final sales call without any human interaction. OK, so I'm just going to show you.

We actually did a LinkedIn launch. This is my business partners. And we had really good traction.

We had 123 comments of people interested, wanted to get into. There was a wait list, so they wanted to get into the wait list. And this is kind of it.

Agent Network Architecture

So basically, what we're building is a network of different agents, and each agent is specialized in some part of this outreach.

Campaigns, ICPs, and Services

So it's structured with campaigns. Also, you have your ICPs and your services, because each campaign has to match one ICP and one service. So the AI knows what is selling and who is it selling to.

Context-Rich Conversations

And also, we take all that information we talked previously, like all the LinkedIn posts, the company's website, the company's LinkedIn post. We take a lot of that into the context, and we start talking.

This is the ICPs. This is different services you see here.

And we start talking with our leads. And we have different leads here. And we have different leads here.

For example, this is me. I've been talking with the AI this morning.

And basically, what this does is it thinks the next steps based on the interaction.

Multi-Channel Signals and Next Actions

It not only measures the interaction, the text interaction, it also has access to different signals. What does this mean? For example, if the prospect enters our LinkedIn account, the AI can see that, can see the signals.

When it interacts with, it hits like. It comments that I can see this and can edit the next actions based on this.

Right now, you only have access to two main channels. It's LinkedIn and email. But in the future, we are open to bringing more and more channels.

And this is basically what we think is what AI is capable of doing right now.

What’s Next: The Future of AI in Sales

But also, I wanted to mention what I think, super quick, like one minute, sorry, what AI I think is capable in the future. What do I think?

From Single-Agent to Multi-Agent Systems

Basically, what I see is right now we have one, we were talking about different channels before. And right now we have one agent focused on outreach. And it does the outreach channel from start to finish the whole way through.

But as this continues, I think we're going to see these different agents in each part, in each different channel.

Channel-Specific Agents and Orchestration

For example, right now, inside our company, we're also developing another ads agent using the latest NanoBanana. And I'm sure you've seen NanoBanana VO3. We're also doing some Facebook ads and all of that.

But I think where this is going, we will see different agents in each part of this. And then we'll have an orchestrator more like a CRO, like chief revenue officer agent, where it plans, it communicates with the rest of the agents.

So then we have like a whole solution for your whole marketing and sales area of your company. This is where I think this is more or less going.

Conclusion and Contact

And I'm just going to finish here.

And you have here my LinkedIn and my email. If you want to have any questions about our software, Thali is the name of the software or anything, feel free to reach out through LinkedIn or email. And I'll gladly hear any questions.