When Human Ambition Meets Turnkey Agents | SMB Keynote, Dreamforce 2026

Please welcome President, Global SMB and AI Labs Go-to-Market, Salesforce, Adam Alfano, and Senior Vice President, Sales, Salesforce, Whitney Hillier.

All right, friends.

Adam, we are back.

Oh, we're so back. It feels good. And look, Day 1. I mean, we've been parked as far out as the Thursday morning after the concert. That's a tough slot. We got a lively room. It's a big day after the keynote. How you feeling?

I'm feeling great. Every year this gets better and better. But what are you specifically excited about for today?

Well, look, we saw people teaming in the room. It's pouring with enthusiasm. We're going to hear from some of the best progressive companies growing at breakneck speed today. You're going to hear about off-the-shelf agents that are going to change your business. It's going to be incredible. I can't wait to do it. We'll have demos, we'll have customer speakers, but of course, we gotta do something before we do any of that, don't we?

We do. We need to start with a huge thank you. Thank you for putting our— your trust in us. Thank you for your partnership, and thank you for continuing to push us. The innovation that you all are driving is just incredible, and we are so grateful to be on the journey with you.

Yeah, I mean, I gotta give SMBs and startups credit. You're getting after it. It's unbelievable. Thank you for your breakneck pace of innovation. Thank you for your experimentation. 92% of you are adopting AI. That's a pretty breathtaking number. And of course you're doing that. You're resource constrained, right? You need to do more with less. But Whitney, there's all these peripheral AI apps that folks are trying that have big promises. They haven't necessarily delivered the value, have they?

No. And this is exactly what we're hearing from our customers every day. New AI apps are popping up constantly with huge promises, but most are just not delivering. They're delivering small incremental gains, and that's That's not what anyone in this room is here for. We're here for the kind of transformation that leads to 10x, 20x, 50x gains for our businesses. And we're constantly thinking about how do we get to this promise of the agentic future?

Absolutely. And look, the great news is if we can advance the slides, we're going to answer that question for all of you. And it really comes down to having a great platform underneath your agentic execution, right? You heard it today in the keynote. It starts with the data. You know that you need great data integrity and context to build good AI, and the agents need to live in the same place that you're leveraging your apps so that you have continuity in your execution. And when you can do it on any surface, whether it's Claude or for us, obviously Slack is a dominant work center, it becomes an incredible ability to drive innovation to the field. We see the 10x rep ahead of us, the 10x employee, and Whitney, we're so excited for everyone to do so as well. And it's really going to come in the form of making some critical digital hires, right? These off-the-shelf agents that you can bring in that will diametrically change the way your teams align and the outcomes of your business. Whitney, I know one near and dear to your heart, and that's, of course, Slackbot. So take us on a little tour.

Yes, well, thank you. I cannot wait to introduce you to our first digital teammate and my personal bestie, Slackbot. Slackbot is the teammate that always knows what's going on with your business. It knows your data, your systems, your teams, and your organizational structure, and it brings all of this context together and it turns it into action. It can help you prep for meetings, understand your sales forecast, help you build marketing campaigns, and organize your day so that you can focus on what matters most. And today I'm gonna take you through a live example about how this has changed the game for me personally as a sales leader. So to set the stage, I lead our growth business sales team here at Salesforce. I have about 700 AEs, and we cover over 16,000 customers. Now, those numbers can be really daunting, especially when you're the type of leader who prides themselves on staying close to the customers and close to the team. But with Slackbot on my team, I don't have to choose between scale and connectivity. Slackbot helps me cut through the noise. It helps me understand what really needs my attention, and then it gives me the context I need so I can jump in and make an impact. So let's look at what that really looks like in my day-to-day. I start my morning in our forecast command center. Now, this view pulls together data from a number of different systems. It pulls it from our CRM, from Slack, from Momentum, and it gives me a quick overview of what's going on in my business. I can see what's working and what, eh, maybe needs a little bit of work. And from here, Slackbot and I can start to dig in. I can ask Slackbot, hey Slackbot, what are the top deals that we're hoping to close at Dreamforce this week? And just like that, I have a ranked view of my top opportunities right there in Slack. And I know that we have a big meeting with Aim Social tomorrow, so I'm gonna go ahead and jump into the account channel and understand what the team has been working on and where I might be able to help. And since it's such an important meeting, I wanna make sure I know absolutely everything that is going on with this account. And that used to take me a long time. I used to have to look through emails and Slack messages, go back and review meeting notes and jump into cases and records, but now I can simply ask Slackbot to help me prepare for the meeting I have with AimSocial tomorrow. And that prompt just kicks off my Slack meeting skill. Think of skills as reusable workflows that you can, will always pull the right data for you at any given time. And just like that, I now have everything I need to feel confident in going into this meeting tomorrow. I know the attendees, I know the key points that my team wants me to hit on. I know competitive analysis and potential concerns from the customer, and it's all put together in this beautiful canvas exactly the way I want it. There's no more digging through documents or circling up the teams. I have everything I need right here, and that gives my team time back to focus on what they do best, which is selling and I get time back to focus on connecting with customers and my team. And that, my friends, is our first digital hire, Slackbot, driving productivity, scale, and connection like never before. And now, I am so pleased to introduce my good friend and our SVP of Solutions Engineering, Sam Davetti, to introduce you to your next hire.

All right, thanks, Whitney. All right, good afternoon, everybody. How's Dreamforce been so far?

Woo!

That's right, and we're just getting started, so pace yourself. All right, Whitney just walked us through how Slackbot helps drive 10x productivity for your organization, but the one thing I hear from customers all the time, what they really care about, is how can they generate more pipeline for their business? So to answer that question, let me introduce you to the second member of your digital team, Piper. Piper is your inbound automation There we go. Piper is your inbound pipeline agent that engages with your buyers all across your website and your email inbox. From the moment a buyer arrives, Piper starts a personalized conversation with them via text, voice, and video, answering questions, qualifying intent, and even booking meetings directly on your sales team's calendars. If the buyer leaves without converting, guess what? Piper reaches out with a personalized message to keep the conversation going. So instead of going on your website, filling up a form, and waiting, every buyer is now part of an instant, always-on engagement designed to convert interest into pipeline. Now, Piper is having a big impact on small businesses, and one such business that I would love to highlight today is SaaStr AI. So please join me in welcoming on stage Amelia LaRouche. Amelia.

Hey, good to see you.

How's it going?

Good.

Thanks for having us.

How's your Dreamforce been so far?

It's good. You know, we have the whole company here, all 3 of us.

Well, we'll get to that. Actually, that's my first question. So we're going to ask Amelia a couple of questions. Amelia, we've worked together for a few years now, and SaaStr.ai is a force in the events world in tech. You guys have a conference that has over 100,000 people that have attended. Half a million people read your subscription. That's a lot. Most companies, that's an army that staffs that. Tell me more about the staffing behind that at SaaStr AI.

Yeah, so SaaStr AI, we're 3 humans, but we do have 21 agents in production helping us behind the scenes. So talking to like the theme and what you just mentioned today about Piper and 10x using AI agents, our Piper specifically, our qualified, which we call Amelia AI, has been up for over a year now, and it's already had over 3 million interactions with SaaStr community members, either about the podcast, media, our events, Sponsorships, anything, you name it. Which is crazy to think because if it was just the 3 of us humans, that'd be about a million conversations each. I'm not sure we could really scale to that level, just the 3 of us. And so it's nice to see our go-to-market agents all throughout the funnel.

That's amazing.

I'm still amazed how much you guys accomplish with the way you're structured. All right, one more question. When AI first came out, and you and I have talked about this before, you were using AI in a certain way, but now you use it completely differently. Can you walk us through that?

Yeah, I think 2 big takeaways here. One is what the individual agents can do themselves, and then 2, what the agents can do with each other now. I think so much progress has been made in the last year. So we have Agentforce, we have Qualified, we've got Slackbot, we have Momentum, we have Salesforce Classic Sales, we have now Marketing Cloud. And I think before all that, you know, think about 13 months ago, probably like many of you, we had a contact us form. And so leads would come to the website, and as you mentioned, they would have to wait for us to respond. You know, we're a small business, There's only so many team members and hours in the day. And so now with everything being identified, Piper and Amelia AI can respond to people not only in real time, but in a way to answer their questions, get context from them, book the meeting so that when the humans now have that meeting with your potential customers, you have a much better meeting because you already know all the information that they're looking for. It's given them answers. You have all the context that you want going into the meeting so that you don't go in cold.

Awesome. All right, final question. Your business is all about growing and activating the community around you. How has Piper helped you on that journey?

Yeah, Piper, along with the other agents, I would say altogether— so we use them full funnel now, right? So you could say, we could kick off an outbound email in Agentforce. And then if they come to the website, they're hitting Piper as the inbound agent. And then once we have a call with the humans, they're hitting Momentum on the side. So we use Salesforce Headless to control and monitor all these agents, see what they're doing. And actually, SaaStr has doubled in the last year in terms of revenue because we use all these agents and they're empowering all of our go-to-market. But maybe more importantly, everybody in our community can have a touchpoint with us now in a way we couldn't scale with just the 3 humans without the agents.

Amazing. Hopefully everybody caught that, doubled their revenue. So amazing. All right, why don't we see a demo on how SaaStr AI is 10x-ing their pipeline with Piper? So to kick off the demo, We're going to start— remember this contact page? So this is the contact us page on the SaaStr AI website. You'd go in here, fill up the form, hit submit, and then most likely somebody from the SaaStr AI team will reach out to you over time. But that's not what we have today. That experience has completely changed with Amelia AI right in the front, in the center of that page. Instead of filling up forms, now I simply ask questions.

Yeah.

Hi, Amelia. I'm interested in SaaStr Annual and would love to learn more about sponsoring the event. When I ask that question, Amelia AI has all the context she needs to not only answer the question but actually feed the data back to the Saastr AI team so they can continue qualifying me, as if she's just another member of the team. Now, I go ahead and enter my email address, and because she wants to personalize the conversation, she's going to go ahead and ask me some follow-up questions like more information about my company, what stage we're at. And when I select that right within this conversation we're having, Amelia AI instantly is going to decide what the next best course of action is based on the conversation we've been having. So that's going to come up in a second here. And what you'll see is when that selection is made, the AI— so there we go. So we select what stage we're at. And now the AI decides, you know what, based on all the information we got from this buyer, potential buyer, maybe the next best thing to do is set up a meeting directly with Amelia the human. So we go ahead right within this conversation. I'm able to pick a meeting and set up time with Amelia. How amazing was that? In under a minute, Amelia AI was able to convert an anonymous lead from Disaster AI website into a qualified event on Amelia's calendar.

Pretty good.

And it doesn't stop there, as you know, because you built it.

Yes.

Right, right after this, when Amelia is ready to have the conversation with that buyer, All of that information from that conversation is available to her. AI is letting her know, hey, here's the stuff you should think about when you speak with this buyer, things to focus on, things to stay away from. So Amelia now has the context she needs to help accelerate this deal. And because all of this data is within Salesforce's ecosystem, Slackbot has access to this. So when the Saastr AI team behind the scenes is working in Slack, they can continue helping this customer go through the journey. So what we just saw is Saastr AI 10xing their pipeline with Piper. Amelia just shared 2x revenue within a year, that's incredible. Thank you so much for your partnership and everything you're doing with us.

Yeah, thanks for having us.

Great job.

Great story. All right, so that's Piper. Now, we just saw how you can 10x your productivity with Slackbot, you can 10x your pipeline with Piper, Let me now introduce you to the 3rd member of your digital team, Momentum. Momentum is your coach. It's your strategist. It's the data foundation that makes all your agents and your go-to-market teams smarter. It automatically captures customer conversations, texts, emails, calendar events, even in-person meetings natively within Salesforce. So that rich context gives your agents everything they need to accelerate deals in the funnel. Now, Perplexity is using Momentum to automatically update their CRM records, move call recordings into Slack, and power their voice of the customer dashboard. So to talk to you a little bit more about Perplexity and how they're supercharging their new AEs with the context they need in this hypergrowth environment, please join me in welcoming Nate Folan, Head of Enterprise Ops and Systems at Perplexity.

All right.

Thanks, Sam.

Come on up. All right. Nice to see you again.

Great to see you too.

Nate and I had some good coffee a couple of weeks ago at his office in Austin. It was fun.

Yeah. Got you a hat.

Give me some merch. I should have worn it actually. Yeah. All right, Nate, couple of questions for you. So you guys are hypergrowth mode. Perplexity is, you know, you're in the enterprise ops and systems space. And can you walk us through what hypergrowth actually looks like?

Yeah. Well, first of all, thanks for having me. Perplexity is an accuracy company, and we offer agent platform. And it's great to be in this room with small and growing businesses, are some of our best customers. They're very AI-first. With our agent platform, they're able to connect to tools like Momentum, Salesforce, and get the most out of the latest AI models. My role is revenue operations, so systems and operations for our enterprise team. And yeah, it's been like basically building a plane while flying it this whole time, as well as adding 10 new passengers every week. So I just came from an onboarding call about an hour ago. So yeah, it's been a lot of fun, and we've been getting a lot from Momentum and the ability to pull these transcripts.

Awesome. And you love Momentum. We know that because you brought Momentum in a company you worked in before, and when you joined Perplexity, you brought Momentum to Perplexity. Can you tell us more about your background with Momentum and what are you doing with it today?

Yes, that's a rare thing, by the way. I'm very tool agnostic, but Momentum is one of the few tools I brought on over at Ramp and I brought at Perplexity. So in hypergrowth, I think it's super important for all of our reps to have the best tools to allow them to do their best work at Perplexity. Momentum is one of those tools. So bringing it on, I really trusted the team at Momentum, and I know the product is very AI-led and AI-first, and they deliver on it. I was able to implement it in under 1 week at Perplexity and about that same timeframe at Ramp.

Yeah. Now you guys are obviously experiencing hypergrowth on the sales side. How does Momentum help you ramp your new sales hires that you've been hiring?

Yeah, so given that we're hiring so many new reps, we really expect a lot of them as well. We expect a new deal closed in their first 30 days, a contracted customer. To do that, we really can't hire and give people at-bats to get there. So Momentum, their first sales call is not their first sales call. They've seen different parts of many, many sales calls along the way. And they're prepared on that very first call. It's also like having their own data scientist. If you use Momentum with AI tools like Perplexity, you're able to ask some very specific questions about the breadth of all the transcripts and get really detailed answers.

Amazing. Well, thanks for answering the questions. Why don't we see a demo? Let's, let's look at the demo for Perplexity and how they're 10x-ing their deal wins with Momentum. But this time we're actually going to start with Nate. He's going to walk us through his voice of the customer dashboard that we talked about earlier. Take it away.

Yeah. So this was actually, believe it or not, basically one-shotted by Perplexity Computer with Momentum transcripts and Salesforce data, as well as external data like LinkedIn. So this is our voice of customer dash. You can see it's pulled over 80 calls. It's found some really unique quotes from over 592 specific quotes. If we go along the left-hand side, there's different things that we care about as a company, like customer love. Like, where— what is our message that's resonating with our customers, like SMBs? How can we showcase that to new reps as well as to marketing and product to see what's working? From there, not only the quotes, but we can see some summaries of those quotes. And my favorite part is that you can actually take action. So we've linked out to run these things in Perplexity. What should we do about the information we're learning from Momentum in these transcripts? And take advantage of AI learning what to do better based on just not reading the news, but executing on what you should do.

Amazing. Well, thanks for walking us through that. Now, this is a great view, but how does it all start? How does Momentum capture those signals in the first place and bring them into CRM? Let's talk about that. So this is what an opportunity page used to look like at Perplexity. Fields were not filled, data was empty. The reason is reps had to do this manually, so they never got it done. Who here is seeing the same challenge? Data is not filled in their CRM. A lot of you, right? Well, Momentum takes care of that. Momentum automatically captures those key signals directly from the calls that the reps are having and brings it into CRM. For Perplexity, that means they can now get a consistent view to their opportunity that follows all the structure they need. So for Nate and his team, they want to get their med pick information that is in here. They want to look at all the risks with the deals. They want to look at the key stakeholders and the different roles. Momentum is capturing this data automatically, putting it in here in a structured way, and it maintains it moving forward. That is incredible. Now, Momentum doesn't just update records. It actually brings those key signals to where people work. And in the case of Perplexity, it's in Slack in the deal signals channel. Right within this channel, Momentum is putting in key signals around deal expansions, key call quotes. It's even taking those calls that the reps are having and clipping them into bite-sized 30-second pieces that they can focus on, right? That's the most important part of a call. So it really saves them a lot of time. Now, here's an example of a rep that's going through the data from Momentum where they can see that Momentum has actually highlighted an expansion opportunity for 60 seats for one of their customers. Within seconds, they get some valuable information that's sent to them. They can even look at AI-generated next steps and take action. So this is how reps get faster in the work they do, and Perplexity is guaranteed to get trusted, structured data in their CRM so they have a consistent view of all of their deals at all times. So what we just saw— and because all of this data is in Salesforce, which means Slack can access it to better train reps, Piper can access this data to continue to sharpen her prospecting skills, and as we just heard, Perplexity is getting a 10x return on the deal wins with momentum. Nate, thank you so much for being here. Thanks, Sam.

Appreciate it.

Awesome.

All right.

We are now going to move to the next section to listen to our next team member. I want to welcome Giovanni Crispino, EVP Sales at Salesforce.

Mamma mia, Sam!

Mamma mia!

Mamma mia! Oh, but did you see this? I have to say, customer love really got my attention. So we should all have customer love sections across all our digital teammates. So I'm here to introduce you the last one— sorry, the 4th one, which is Agentforce Coworker. Agentforce Coworker is the planning analyst. It's actually your autonomous AI teammates that work already with your existing employees to make sure that when those employees are in Salesforce, in Claude, in Slackbot, ChatGPT, so forth, so on, they have somebody that helps them to to find the right information, to answer those questions, to get work done without any technical setup. So that's very important. Do you know if you are a Salesforce customer, that is already available for you in your Salesforce environment? It's a true representation of what we have seen today, AI force. But, you know, the beauty of it, you have seen it with Saastr, with Perplexity, is to have here a customer that's going to come and tell us about it. So, It's a pleasure for me to introduce Jeremy Frankel, the CEO of Fundamental. Jeremy, come on, come on, give us a hug. Give us a hug as a true European, right? You know, I just learned Jeremy is from Belgium, you know, a keen soccer fan as well. But Jeremy, tell us about the power to predict. Tell us about Fundamental, what you do.

Thank you for having me here. So over the last several years, we've all seen how AI has transform the way we work with text, with audio, with video, with code, with pretty much every data type. But when you look at businesses, that's actually not the data they really run on. The data they run on comes in tables, and that's, you know, production data, sensor data, maintenance records, CRMs, spreadsheets. And essentially, that's the operational data that companies already generate every day. And so And businesses use that data to make decisions. But remarkably, that's also the data that has been left out by the AI revolution. And so what we do is we build an AI model called Nexus to enable companies to make much better predictions on their structural data. And so we've raised close to $300 million to accomplish this. And today we work with Fortune 500 companies across a bunch of industries, from sports to media to energy and shipping on their most important prediction problems. And for businesses, predictions is everything. Like it's, for example, being able to predict which customers are going to churn, what equipment is going to fail, which assets should you invest in across thousands of different assets you have in front of you. And so there is a historical opportunity to take all of that data and to use our model to make better predictions on what happens next.

Very good.

Very good.

So, and talking about before Salesforce, right? Your team has different applications, some of them custom built. What brought you and your team to choose then Salesforce as your go-to-market infrastructure to scale it?

So there is a lot of talk on X about people vibe coding their own CRMs, and it's not a focus of ours. It's not a priority. And, you know, we're building Nexus, right? Our— it's a superhuman prediction engine, and every minute we have is spent on it. And so if I have the choice to ask my team, my engineers, to build a system of record or, you know, to pay for Salesforce and be able to focus on building a billion-dollar revenue-generating company, where do I sign? And that's essentially the entire point here is that like Salesforce is our single scalable source of truth, and we very much understand that because Nexus, our model, uses enterprise data to make better predictions. But all of it starts with having a source of truth.

Okay, very good. So, and selling to Fortune 100 companies, you have the source of truth, but those companies are complex, as big. You have a lean team. So how important is for you and for your team to go to take those faster decisions, right, when engaging them?

It's critical because, you know, every day we talk to Fortune 100 companies and we ask them what prediction, if you could make it materially more accurate or faster to deploy, would change your business? And so our sales team's entire focus is to really go and understand what better predictions can lead to clear action that drive meaningful, measurable results quickly for enterprises. And our goal is really to compress that time to value. And so that's why, you know, now given that Nexus, our model, can pretty much work on any prediction task, the goal is to know how quickly can we get value for our customers, because the sooner we can get value, the faster we can expand within an organization.

Very good.

Very good. So let's look live together how a day in Fundamental looks like, right? So if you move to, you know, we remember before the information used to be available, but then a team of analysts had to pick them up and stitching together. So now in a conversational way, what we are very focused, well, especially with somebody like Jeremy is, okay, how is my team training? How is the results for this quarter going? And it's great, right? We can see $28 million closed. There is still $9 million in pipe, $1 million got— by the way, $28 million. Congratulations. Very, very big number.

$100 billion.

And so what we want to know, this is the first part, right? But now let's see what can I do? What can I ask in order to close the gap? And Agentforce Copilot helped me already to prioritize and give me a snapshot of what are my top deals. And those top deals is not in amount of value, but in terms of what priorities, what actions should I take with them in order to solve, in order to get the best team focus on the fastest way to resolve the problem of the deal and to move it on. So what the Agentforce coworker is doing, so it gave me the information, it helped me on the insights, and now it's giving me as well the plan of action. It can be a compliance training, it can be an FDE, it can be a customer call, it can be yourself being involved in this call. And that's the beauty of it. So what we're doing is we're seeing in one command center without leaving Salesforce, insights that we are taking from Momentum, from Piper, even from Slackbot, collect together, and therefore straight away getting those decisions on the plan. Jeremy, thank you so much for being here. We're going to contact Jeremy for next World Cup prediction. If you have not seen the live video where he had predicted before, right?

Correct.

You said Spain was going to win.

Yeah, we predicted with our model that Spain was going to beat Argentina in the final before the World Cup started.

I mean, imagine how many things you can do in the business as well when you can predict the right outcome. Jeremy, thank you so much. Thank you. Thanks for being here. A pleasure now to introduce— it's a real pleasure to introduce co-founder and CSO of FINN, our latest addition to our family in Salesforce, Des Traynor. Des, come on board.

Thank you.

Congratulations.

Thank you, Alberto.

Thank you.

Thank you. It's a pleasure to be here. If you haven't heard of Fin, you might know us from our previous name, Intercom. Intercom, we started in 2011. We worked entirely in the area of customer service. We were responsible for pioneering that chat bubble that sits at the bottom of every website you've ever used. And we long invested in AI. We had a long interest and investment in AI and automation and what it could do in the customer service space. When AI really happened, November 29th, 2022, we all remember where we were. We realized there was a challenge on the horizon, and we could either die or survive and thrive in AI. So we went all in on building the product that would ultimately redefine and reignite and refound and ultimately rename the company to be Fin. Today, Fin is the number one customer agent. Fin resolves 76% of the conversations that are thrown at it without any human involvement whatsoever. That is our average. We have customers in the 80s and in the 90s as well. We have 30,000 customers. We resolve 2 million conversations on a given week. And our goal when we partner with a customer is nothing less than a total AI revolution of exactly how they handle customer conversations. And what does that look like? Well, for us, revolution means you should be seeing extreme savings like Rocket Money. You should be delighting your customers like we do with Vanta. You should be resolving really well, like we do across the board with all of our customers, and you should be saving a phenomenal amount of time. If there's any one customer who you think would've built this themselves, it might be Anthropic. When they rolled out Fin, they saved 1,700 hours of team time in their first month. So how do we do this? We focus and obsess about 3 things. Firstly, We always want to lead the industry in AI for CX. We have gone further than anyone else here. We've post-trained our own models. We scientifically rigorously A/B test every single change we make. Fin is just one engine, 30,000 customers, et cetera, but it is one core engine. When we improve it, it improves for everybody. If somebody signs up tomorrow, they sit on all the improvements from the previous 3 years as well. Secondly, we obsess about Our deploy time and our time to impact. We deploy in days and we deliver extremely positive impact immediately. And lastly, we integrate really well with your help desk. So obviously we work with the Intercom help desk, but we work with Service Cloud and HubSpot and all the rest as well. To just give you one customer example, Kalshi, our predictions market company, they contacted us shortly before the Super Bowl. And obviously Super Bowl would be a very busy time for such a business. We were able to get them live in 10 days, and I think when we went live, we were doing 60% straight away. Shortly afterwards, we were processing 80,000 resolutions a month with a resolution rate of 80%. And when Kashi look ahead, they're forecasting maybe 2 million conversations happening over next year, and Fin remains fundamental to how they'll actually deal with all of this. So enough talk, let's do a quick demo. I'm gonna hand over to Gavin, and Gavin's gonna show you an example of Fin working for a banking company where a customer contacts them with a complicated issue. Over to you, Gavin.

Okay, so I am going to call Finn now.

Hi, you're through to Quenmark Bank customer support. How can I help you today?

Hey, I lost my bank card.

I'm sorry to hear that. I'll quickly check the steps for a lost card.

Okay, so Finn's going to look up the lost card procedure, and so we can follow along with the process.

Before I make any changes to your account, I just need to verify your identity. I have sent you a 6-digit code by text. Please read it back to me when it arrives.

It is 532381. Because this is an issue involving sensitive data, FINN will verify my identity before moving on to the next step.

Thanks, you're now verified. I can certainly help you cancel your card and order a new one. I'll just pull up your account details first. You have 2 cards, a Visa credit card ending in 2152 and a Mastercard debit card ending in 5892. Do you know which one you've lost?

Okay, so let's not make this too easy for Fin. Uh, sorry, Fin, I don't know.

I understand. That's no problem. I can help you work out which card it was. Do you remember buying anything with it recently?

Oh yeah, I got a cup of coffee this morning. You can see this transaction here.

I found a purchase for $8 at Sightglass Coffee today on your Visa credit card. Is that the card you've lost?

Yes, I guess so.

I recommend that we cancel your Visa credit card now and have a replacement mailed out to you. Would you like me to go ahead and do that?

Yes, please do that. So, I mean, this is real magic, right? Fin is using the backend systems to autonomously cancel my card and issue a new one.

The card ending in 2152 is now canceled, and a replacement is on its way. It should arrive within 3 working days. Is there anything else I can help you with?

No, that's everything. Thanks for your help, Fin. Bye.

That was cool, right? So I can't stress this enough. There is no smoke and no mirrors. This is real live Fin. To demonstrate this, we'll just walk through some of the backend of Fin here. The way you use Fin is you train it by feeding it articles, private articles, to sync it with your help center. You give it all the content you'd give a new Expert employee. Like, as if you imagine if you hired Einstein onto your support team, this is the way you sort of teach Fin. Done that, you can then— so you can guide Fin by giving it sort of instructions as to how to behave, what tone of voice to use, how to handle certain things. Also tell Fin specifically how to escalate. How to escalate. Sorry. How to escalate. Can I get someone else? Oh, here. Okay.

Keep going.

Okay, cool. How to escalate. So if you really want to make sure certain conversations do or don't go through to support, you can teach FinDot. What we just saw was a complicated one. So for this, we use this thing called procedures. Procedures are where you can express in natural language the way you would talk to a human. You express exactly how to resolve an issue. You tell it where to look for data, how to parse, how to pull data in from backends, how to update certain pieces of backend information, cancel a card, etc. And that's basically how you teach Fin how to handle certain more complicated scenarios. We have a lot of cool features here. I'll just briefly cover deploy. So you can deploy— it's one Fin over chat, email, voice, as we just saw. And within chat, we are everywhere— Facebook, WhatsApp, Instagram. If your customers use it, if your customers' kids use it, we use it too. We are everywhere you would expect us to be. And there's a lot of other features that will teach you exactly what's going on across your entire customer experience. But I'll leave you with one called Operator. So Operator to me is like prototypical of where the future of software is going. We could teach you how to use all of our beautiful UI, or we could just say, tell Operator what you want to do and it'll do it for you. So in this case, if we can just ask Operator, hey, I'd love to know why are people contacting us? What are the top issues and trends? Now again, this is live. So while that's cooking, I'll show you a different example. Here is a pretty complicated flow that we designed in, say, Figma to explain exactly how to handle a user wanting to cancel a pending transaction. Now, normally, the next step from here would be a big workshop with whiteboards and all sorts of stuff. In the future, we're just going to hand things like this to Operator and say, you do it. And Operator reads, parses the image, understands everything that's going on, and actually drafts a full, complete procedure for you. It's basically saying, here is what I think we need to do. Does this make sense? You can tweak it if you want and put it live and boom, that's yet another complicated issue handled immediately without ever really having to learn anything. Just express your intent and the software works it out. Now maybe our previous one has finished cooking. Let's have a look. We asked it, why are people contacting us? Again, it's generated all this bespoke UI to explain exactly how people contact us, why they contact us, what are the main issues, what's trending, what's not. And you can have a long back and forth conversation here if you want. You can drill into certain topics and it'll keep answering with UI, with actions, with provocations to keep the conversation going. Our goal here is to make it so we have the best AI and also the easiest interface so that we can literally go live for you in days and deliver extremely high impact. So that's all we wanted to show from the software. I'll just leave you with this final piece, which is if you're interested in Fin itself, you'll find we have a talk starting just downstairs from here in about 10 minutes' time. Paul Adams, our Chief Product Officer, is leading that. We have another talk tomorrow. If you want a demo, please swing by our booth. As proof of the fact that we're quick to deploy, we've generated custom demos for basically every company who's attending Dreamforce. So we're already ready to go for you if you want. Just come by and say hi. And with that, I'll leave you and hand you back to Adam. Thanks so much.

All right. Incredible. Des, great job. Welcome to your first Dreamforce. I'm so thrilled to invite up Zeynep, CMO of Lagora. Talk about a fast growth company. Lagora is scaling like crazy. It is great to have you here. Thanks for coming.

Great to have you here, Justice.

All right, well, you've had quite a career. You've been the CMO at Atlassian, Palo Alto Networks, huge companies. What attracted you to Lagora? Obviously, it's moving at breakneck speed, but what did you see that got you excited?

Well, thanks for having me here. There are a lot of reasons I said yes to Lagora. First of all, I think it came down to 2 things really, the moment that we're in and the company itself. The AI, the legal industry is going through a profound transformation and it was quite irresistible not to be at the heart of that. So that's one of the reasons. But Legora is also one of the fastest growing AI companies in the world right now. It has such a deep command of both law and AI. I think we have almost as many lawyers working for the company as engineers. And with that comes this customer obsession and willingness to build with customers that is very refreshing and unique.

How big is the team now? And where was it when you started and where are we today?

Yeah, the team grew 9x year over year. So I joined recently, but went from 100 to 900 in one year. And the ARR grew, back to your theme, 10x. So that's where we are right now.

So you're getting the 10x growth, and as luck would have it, you've made a few of these digital hires, right? Which is incredible. So you're using Fin, you're using some Slackbot. I think soon you're going to use some Qualified as well, I think is on the horizon. Tell us a little bit about how these tools are impacting life at Legora.

So to your point, the growth is hypergrowth. It's very high. And we decided at some point to stop viewing, treating hypergrowth as a hiring problem. And then we started treating it as an operating model problem.

Yeah.

And when we did that, we asked the question, where is the seller's time going, right? And it turned out that it was a little bit of selling, but it was mostly research before the sales calls and all the admin work you need to do after. So we said, we're going to accelerate all of that with agents. And that's when we had like a 3-layer operating model, essentially with Salesforce as the foundation, Slack as our workspace, and an agentic layer on top as the execution layer. So this is Salesforce as the foundation was really critical for us because it actually not only had our customer data and source of truth, but all our shared definitions. What's the definition of stages, exit criteria, et cetera? So we were able to unleash agents onto all of that context. And then Slack is our office basically. Our entire company runs on Slack. And we were able to, I mean, we have a very unique fun fact about us, we actually encourage everybody to have, to be in public channels so that we can create this layer of institutional knowledge that agents can tap into going forward. So we have, for each account, we have a Slack channel, and that's where all the fun stuff happens. So every morning when our sales reps get up, their meeting briefings are ready in the channel. They kind of like look through that. They don't have to do any of the work based on account history. And then they, when their meetings are finished, all the transcripts, action items, and decisions are captured both in Salesforce and published to the Slack channel. Every Friday, we get a report card seeing how the deals moved, what's at risk, not always the most flattering reports. But so that's kind of how it comes down to, it comes to life. But just want to note, Finn, we were actually users of Finn's operator product agents very early on when it was in beta. And that kind of like, that's how we run our customer support operations, gives us a really good picture of volume, where the load is concentrated, and all the patterns in our customer tickets.

Very cool. Well, we appreciate the partnership. And you were sort of headless before we were even saying headless, right? I mean, very similar to the way we're using it. Slack is the work center for— I have some employees in the room. That's where we congregate. That's where we execute. We're rarely actually in our old Salesforce instance. We often do use Coworker. But now you're using Claude a lot on top as well. So tell us a little bit about the headless experience. Slack's on one side, and then you're building some bespoke experiences with Claude.

Cloud, of course. Yes, it's very powerful for us to be able to sort of tap into that foundation from wherever we can, and cloud being one of the top use cases. I'll give you a couple examples. In my world in marketing, we actually have a lot of conversations with our data. We like having open-ended questions and iterations with our data. And so we'll ask about dormant opportunities, inbound leads, et cetera, and keep the conversation going. And being able to converse with our data and not be confined to a dashboard is actually very liberating for marketers. I'm sure everybody will resonate in the room. Another really good example is, you know, one of our sales enablement team members actually built something using our Salesforce Foundation, a few Claude skills, and Claude Design. It's a dashboard that helps us track our new sellers and how they're ramping. And that's become a very useful artifact overnight.

And that would've never been possible without the headless It's incredible to see the ideas move into action and production. And often these things have really unique styles and aesthetics that become really pleasing to use. We saw some of that today with Patrick Stokes' demo. So that's very, very cool. Now, there's a lot of partnership ahead for us and Lagora. Talk a little bit about some of what lies ahead for this partnership and how you see us working more closely. We're very—

first of all, very happy with the partnership. I want to call out to Allison, Jim, and Jeff, who are our account team, who are just amazing supporters. support for us. The partnership is moving forward in very exciting ways, actually. We have a few announcements tomorrow, but I'm happy to tease one of them today. It's the integration between Legora and Salesforce. Going forward, joint Legora and Salesforce customers will be able to access customer and deal context from within Legora during contract drafting. And this is a big deal because most of contract drafting actually starts with facts. in Salesforce, customer details, pricing, and deal terms, et cetera. And so all of that will be available to the Legora agent going forward. So your contracts will be drafted with a lot more context.

Incredible value add. It's going to make people way more productive. Thanks for all you do.

Thank you so much.

Thanks for the partnership.

Incredible.

All right, Des, Whitney, you want to come back up?

Wow.

We covered a lot of ground. We hit 5 key hires I know you're all going to make if If you haven't made them yet, we do have a ton of sessions left. This isn't the end of the road. We got 2 days of this. We got a concert. Scan the QR code for more sessions. We also want your feedback, of course. So dig in, and there'll be coffee on us. And if you can go in and give us some feedback on the keynote, I hope you enjoyed it today. Des, how about a little feedback from you? Your first Dreamforce, you're taking it in. Any feedback for us?

It's daunting. It's incredible to see the ambition and the optimism from all the people here. Yeah, we're really excited to be here. I'm really excited to get to know all of you a lot better over the coming years.

Well, we're thrilled to have you. Whitney, any final words?

I just want to end where we began with a huge thank you. Thank you guys for coming out today and for being on this journey together. Please provide us feedback. That's how we grow and get better, and we appreciate all of you. Have a great rest of your Dreamforce and a great rest of the day.

All right.

Enjoy the rest of Dreamforce, folks. Thank you. Hey, great job.