OpenAI Secretly Launches a Sales Agent. Here's the details

Foreign have just launched their very first B2B agent. It's a sales agent. They did it all the way over in Tokyo at a closed event and demoed the product. But we got the exclusive. We're going to take you through the demo and give you a good overview of what it's going to do and what it means for all of us doing knowledge work. Let's get into today's episode of Marketing against the green. Okay, so OpenAI had a big event in Tokyo. They're all at high, hanging out there in Japan, having a good time. They had a really big event and they showed some future product use cases in this event and one of them caught my eye because it was a sales agent. Now why is it so interesting? Because OpenAI for the most part has focused on consumer products and they do have a large enterprise sales team and a large enterprise team service in B2B companies. But it's mostly how they use the foundational model. And this is really the first example of OpenAI building a vertical agent. Right? They're building an agent that actually helps knowledge workers do their work. And they're building a use case that really does work. It's in sales. And so I want to go through the demo, show you exactly what it does and kind of give you some of my thoughts as we go along in this two minute demo. All right, so here's the demo. It's a pretty lowly lit kind of place, but again, it's not meant to be streamed live. But there's so much interest in this stuff. Someone recorded it. And so let's get into the demo. This is just the sales agent. It's very short two minutes, but I think it gives a really good indication of where OpenAI going with these vertical agents.

This is a small demonstration just to hopefully inspire you and give you something you can bring back as you look ahead to the rest of this year. Let's start with something we know, sales. We have a sales contact form where someone is trying to reach out to OpenAI to learn about ChatGPT.

The first thing we should really note here, because I posted about this and there was a lot of hysteria where people were saying, oh my God, like OpenAI are going to kill the human part of sales, right? They're going to kill the human to human connection we need during sales. So first of all, I want to make sure you all understand. And as part of that, actually what they were saying was, hey, I don't want to be, I don't want opening, I don't want OpenAI having a sales bot that reaches out to me, sends me cold emails. So the reason I'm making that point is because what got lost, I think on people is this is an in band lead, which means that it is someone who has signed up to a form and they want to actually contact your sales team. So intent really matters. The way you feel about your AI experience will really be dependent upon the intent and how well AI makes that experience better for you. I think for the most part, humans will not care whether something is AI or is not AI if they have the intent to do something and AI helps them do that thing.

Now we know how this usually works. Someone submits this form, it goes into a system and then a team of people, account associates, are reviewing the leads to reach out and figure out how best to respond.

And this can take like I think in B2B, we've had them working in automation for a long time. We've used machine learning and data science for a long time to be able to qualify and prioritize leads. But somewhat, I think not all companies do that.

What could it look like if you had agents inside of your workforce working for you right now?

So the other really important thing is I think the way autonomous agents are going to work best within the workforce is they are going to work within your workflow and they are going to be connected to your tasks. And actually your tasks will be the starting point for the agent starting to do its work. What do I mean by that? Someone filling out the form here is the trigger. And that trigger triggers an agent to go do work. And that means you are in. That agent is part of the workflow it does. It's not separate or tangential to workflow. Right? You have these triggers that actually go tell agents to go do work. And that work is part of the consistent knowledge workers daily tasks, right? It's part of how they do work. It's not separate to how they do work. It's just like integrated to their daily workflows. It's how they actually do work. And it's all triggered based upon what the customer and buyers doing across your customer journey.

So first we'll submit the lead here and we'll come over to our system here. You can see what a virtual teammate, a virtual sales associate could be doing.

So this is what I meant. This is a good indication of what I meant, which is like if you're a sales rep, this is what you do anyway, it's in your workflow. But now you have these small autonomous Agents enriching the data for you, enriching that contact for you. And I think he's going to get into this. What I think is a great use of their technology is their deep research that they've just launched that has reasoning connected to the web is able to enrich the contact records for them.

If we open the task here, we'll see that the agent is picking up work for you already to try to help this lead. The account email came in and already it's doing what a sales associate might do on their own and researching that lead, figuring out the industry, the revenue title and other information. Think about the deep research that you just saw and imagine the ways that this could plug.

Exactly. So this is like standard firmographic data, but deep research. With deep research you're going to get like really tailored information per company and per contacts in that company. So what he's shown here is like again when you look at the use case, don't look at these use cases and think to yourself ah, that's kind of underwhelming. I already do that. Too many people do that. Look at this use case and then think about the technology that exists within OpenAI like a deep research which is like a reasoning agent fine tuned on research connected to the web and the create how you can get really creative using that data in terms of how you enrich this contact record to provide reps with interesting data to be able to reach out to that person to book a meeting.

Now in this case the agent was able to see that this is a really good prospect. OpenAI wants to buy OpenAI so we think they're a good customer and so we're going to get the calendar availability.

One thing that will happen very soon is the agent will be able to pattern match against previous if this is connected into your CRM or something else, it can pattern match the it enriches the contact and then it pattern matches that contact against other contacts you've closed in the past. And that's how we'll know whether it's a good contact to reach out to or not. And that will be a pretty simple addition for OpenAI to make and figure.

Out when we might hop on the phone with them. Then it's going to run, it's done there.

We kind of skipped over it. So basically it connects into a tool. So again OpenAI I think the, the kind of this is showing they're going to be the orchestration layer and then that orchestration layer will have this agent will be able to call different tools And I suspect companies will be able to build into their agent these different tools that work with different software. So the first tool he showed us is this agent calls a tool to enrich the contact record. And he showed us a pretty basic version of that like phone number, email, whatever company name. But where this will go is you'll be able to call Deep Research and that will be a phenomenally powerful thing to call a lot of perplexity. One of the great use cases for B2B and perplexity is Perplexity is actually really great for external research of companies and contacts. OpenAI are obviously going to usurp that and allow you to use Deep Research where there's a great reason engine focused on reasoning agent focused on research connected to the web. And you'll be able to call that tool as part of this agent's, you know, list of tools that it can, it can actually call.

Hey everyone, remember that episode we had with Dharmesh? Well, Dharmesh has exclusively gifted us access to agent, AIs, company research agent, just for all of you. So if you're looking to spy on your competitors so you can get a deeper, deeper dive into their strategy, this is the agent for you. You'll get access to their key demographics, their web traffic trends and, and you'll stay ahead of them with all of the competitor analysis tools. So if you want it, you can grab it at the link in the description below. Now let's get back to today's show.

The second tool it calls is it actually has access to the calendar. So you can go and look at the reps calendar and say, well, you have free time available in these slots. And then, then it calls a different tool to craft the email and get back to them. So it's added the time slots and has the email.

In this case, this agent hasn't been told to write any specific language. It's really smart. It realizes that the prospect wrote in Japanese, so it's going to write back.

That is huge. Right? So like the way companies have to go international is they have to wait until they have international capabilities or language capabilities when each country they're trying to go into with agents, you actually don't need a lot of that because the agent itself knows a prospect can enter a form within a certain IP location and actually can adjust its language to match that IP location.

And sure enough, if we come over to our inbox, we can see that email right here waiting for you. Again, this is a small demonstration that we wanted to pull together to help you See what it could look like. You can imagine the applications for this are endless. If you start to look around your organization, think about all the small tasks that people on your team.

Small tasks. I really want to make this point. The first wave of agents, the ones that really will stick are going to be small autonomous agents that do small tasks and are triggered off by the person doing work. So as the person does work, they create triggers. Triggers get sent to small agents that do single tasks. Single tasks get integrated back into that person's workflow so they're never outside of the workflow. It's just like part of what they're already doing. You have all of these micro agents doing small tasks for you and it's going to seem, it's going to be great because the first wave is going to take all of the mundane tasks off you, right? It's really all of the mundane, boring things you probably don't want to do. These small agents are going to be able to do for you. And that's, that's really what I want. He, like, he really mentioned it here. He said small tasks and I think there's a reason he said small tasks because these micro agents doing single tasks, taking mundane off work people is going to be the first wave of agents that really stick.

And the ways those add up. This doesn't have to be an all or nothing.

They add up. That's really important as well. You have like 50 micro agents doing 50 tasks for you. You get back a lot of time.

You can really chip away at this and bring the latest technology into your workforce now and help save your team time and focus on the next steps after that. I hope this is inspiring and I can't wait to hear what you do with this deep research and everything else.

Awesome. So I think this is a great example of it's hard to know where these large companies are beginning in. So OpenAI is now getting into the vertical agents. There's a whole industry of vertical agents built on top of AI OpenAI. And so that's like one takeaway is like it's just hard to understand where the, the kind of these companies begin and end. What use cases will they build? What you case use cases won't they build? I think in this case they're going to build vertical agents but in some cases will they get the depth of single companies focused on that specific problem? If you're a startup and all you do is try to focus on sales and solving that problem, will OpenAI ever be as good? Because it's One of hundreds of things that they're trying to solve for. And that's a debatable point. But the second thing is it's another great example of the future that we are heading towards where you will have a single knowledge worker, like pseudo AI knowledge worker who can complete a bunch of tasks for you. But really what they're doing is it's an agent that's calling tools and complete on a variety of tasks. And those agents, these, these kind of micro agents exist across your workflow. So as you do work, you create triggers and these triggers trigger agents to go do things for you. And they all come back and integrated, they're all come back and integrate it into your workflow. Because adoption and usage is two of the biggest problems with AI. People see it, use it to begin with and they're like, oh, this is cool. And then they just go back to their normal workflows. And the reason they go back to the normal workflows is because the AI stuff lives outside of what they do day to day. They have to remember, oh, I have to go use the AI assistant, I have to go use the AI agent over here. It's going to make my life better. And I use it the first time, the second time, but then I just kind of default back into what I was doing anyway. The user behavior is the hardest thing to change. But you can imagine what they're showing here is this is all just part of how sales reps works, right? The agent takes the contact record, enriches it. The agent actually looks at your calendar and books to meet in. The agent actually books a meeting for the sales rep and then crafts the email and books a meeting for the sales rep. And the sales reps, just what happens to the sales rep, their, their work is now just a little further up the stack. So now they just have meetings appearing in their calendar with good fit leads. And what sales rep wouldn't want to live in that future? Most people don't want to do the mundane work of the qualification. They're sending out the email to see if you want to actually book a meeting, going back and forth with calendars. But I do think there's like really important things here and I think one of the most important things is first wave of agents likely going to be these micro tasks that get built into your workflow and get triggered off the things that you do, the work that you do. And those tasks will add up to lots of time saved. So OpenAI sales agent, I think it's a great example, great demo of where this is going and some examples like where I think agents will start to become really prevalent and where OpenAI will be able to continue to build into all of the work that we do. Hope this was useful. We will see you again on the next episode. This data is wrong. Every freaking time. Have you heard of HubSpot? HubSpot is a CRM platform where everything is fully integrated. Whoa. I can see the client's whole history. Calls, support tickets, emails. And here's a task from 3 days ago I totally missed HubSpot grow better.