Back
Andy White
Senior Vice President of Technology, Salesforce

Win More Sales: Salesforce Agentforce for Sales Productivity

🎥 Sep 01, 2025 📺 Salesforce ⏱ 36m
Learn how Salesforce supercharges sales productivity with Agentforce. We'll share our deployment playbook, ROI methods, and ...
Watch on YouTube

About Andy White

Andy White, Senior Vice President of Technology at Salesforce, has discussed the company's deployment of Agentforce, an AI tool for sales productivity, at Dreamforce 2025 and in subsequent interviews. He stated that the tool has unlocked over $100 million in value for Salesforce and described its use as a 24/7 assistant for sales development representatives, automatically sending personalized emails to low-scoring leads and scheduling meetings. White noted that the agent's emails "beat human performance on day one" and that the quality of follow-ups has improved, with the first deal closed through an agentic follow-up that re-engaged a lead the human team had moved on from. He emphasized that the differentiator between a mediocre and a great agent is the focus and effort invested by business partners. White also addressed challenges in implementing agentic AI, including discomfort around automating deal approvals due to fears of unintended consequences. He advocated for rapid prototyping, sharing work early, and using multiple large language models for experimentationebb. White encouraged teams to embrace AI openly rather than feeling guilty about using it, and he expressed excitement about the potential for agents to assist with key pain points for sellers and employees, as well as the ability to use large language models to augment human capabilities.

Source: AI-verified profile updated from Andy White's recent appearances. Browse all interviews →

Transcript (22 segments)
M
Marcy Ladderman0:02
It's great to see everybody today. Welcome to Dreamforce. I think this is probably your first session, so thank you for joining us bright and early on Tuesday morning. We have a great session for you today and I'm joined by a great team here that will introduce themselves in a moment, but we're really here to talk with you about how we are driving sales productivity with Agent Force and the journey that we've been on for the past year. I am Marcy Ladderman. I'm going to let Adam and Andy introduce themselves when they come up and talk, but we are your team today. As I mentioned, I've been here at Salesforce for 19 years. In my current role leading the global business strategy and operations team, one of the great privileges I have is helping to define and drive execution around how do we completely identify the sales cycle for Salesforce with Agent Force. So it's been a really exciting year for us as we've been on this journey and that's a lot of what we're going to share with you today. We do have great content. The first is I'm going to talk you through what that journey's been like for us, where we were a year ago, where we are now, and everything that we've learned across the way. I'll share with you also a really good playbook or framework for you to think about as you look to become an agentic enterprise. Adam's going to talk about how we are solving and Agent Force is helping to solve specific business problems in SMB. So we'll get very specific about the work that we've done there and the real results that we've seen. And Andy is going to talk through some enterprise implementation best practices and the lessons learned from his view and ours as a company primarily on the technology side.
All right, so let's get into it. Agentic AI is completely transforming the way we work and I think you're all probably seeing this really being led in the consumer world with products like ChatGPT. But the way that we used to work before is kind of more of an apps and tabs, manual processes, point solution, siloed data in many different places. But the world is completely being disrupted and the expectations of consumers and your employees is drastically changing as a result of it. They want to engage in conversational intelligence. They want and expect as they are engaging in these conversations that the things that they're engaging in have the real-time data information and are true and accurate and that they're talking to each other. And so the expectations are changing and it's really critically important that we meet our employees and our consumers and our customers where we are and meet them in that moment. And we have this pivot point. And so you can choose to kind of be what I would call a legacy enterprise, which we all were and many of us still are. But in a legacy enterprise, more and more we're going to see slower action, lower productivity, and slower growth from those companies who are continuing to operate that way. But you can choose to sort of seize the moment and look to become an agentic enterprise. And this really has an amazing amount of potential to drive productivity and innovation for your organization. Think about an agentic enterprise as a way to augment humans to drive increased and amazing capacity without having to hire more people, but to make the work that your employees are doing better and to fulfill your employees in a new and different way and that's going to result in more limitless growth for you as an organization.
And that's the journey that we've been on as customer zero. So you'll hear a lot of us talk this week about being customer zero. And for us that means that we are the first users, the first adopters, and the first testers of our own product. And this is really critical for us not only because we want to reap the benefits of the amazing innovation of our technology teams. But we need to stand here firm and say to you, these are the best products because we've used them, we've tested them, we've validated them. And that's really on principle. It's really important for us and that's the journey that we've been on over the past year. We've worked incredibly close with our product organization and our people like Andy and our technology and Adam in our business to become customers there and to be the best users of Agent Force possible.
And so this journey that we've been on, it started a year ago essentially at Dreamforce, right? We were sort of with all of you as customers sort of embracing, digesting, trying to learn about the technology. We sort of shot out of the gate really quickly and we're like we're going to launch a bunch of agents and we launched about 100 agents and we learned really quickly that that wasn't a recipe for success. And so we've learned a lot over the last year and have a bit of this what I would call a framework of how you can become an agentic enterprise. I want to walk through each of these eight critical steps for you because it's a bit of a way how you can take it and apply it to your organization to become an agentic enterprise. The first is setting a vision. Set a goal for what you want to achieve and make sure everybody is aware of it. Have intention of what you're trying to do. The second is prepare and inspire your people for the future. This transformation is not just a technology change. It's a complete workforce innovation and workforce transformation. But you need to bring your employees along. You need to help them understand why it's helpful for them and beneficial for them and do not leave them behind. How you can reskill them, how you can make them more productive. Number three, begin with the right use case. So I mentioned that we started with 100. Not great, right? And what we were realizing is we had a hundred or so agents out there, but about 80% of them were not driving any business value for us as a company. And so we sort of rolled back that strategy and got much more intentional around the areas where we thought we could drive the most amount of business value and be very focused on driving the most within those areas. And the fourth, create business processes that think. You should never think about identifying a bad process. There is foundational work that you have to do as a company to streamline to simplify your processes deeply. Understand your processes so that you can bring agents and AI into the flow of work for your employees. If you don't understand your process and you don't simplify the process, the output is not going to be what it could potentially be. Number five, ground agents in trusted reliable data. You've all heard a bit of the expression bad data in, bad data out and that is incredibly true in this situation. Agents can look across data from all different sources, external and internal, and you as a company get to decide which data you want your agents to be pulling from and connecting to. But if the data isn't structured or clean, you are not going to deliver the best outcomes for your employees. And what they're going to get in that agent interaction is bad data and bad information. And think about the scenario where a customer or an employee is pulling bad information from an agent and then they go and do something that is not actually the right thing to do and you have some big negative impacts.
Number six, think about elevating humans and agents together. I think when we got out of the gate early with not just us but from an AI perspective there was a bit of a agents are going to replace humans and one of the things that we've come to find is it's not about an or, it's an and. When you think about where agents can drive the most amount of impact for your organization it's augmenting your humans. It's strategically being placed across the processes where you can automate the work, kind of the more mundane repeatable work that your humans are doing and free up their time to go on more high-value work. So think about this really helps to bring your employees along the journey where you can think about how the agent is going to augment them, make them more productive and free up their time. And it's not about a replacement, it's an and. You need humans and agents strategically working together.
Number seven, now I think this is probably the most important thing and something that has been a huge learning for us. Test, tune, and iterate. When you launch an agent, it's similar to basically hiring a new employee. When you hire a new employee, you don't give them a badge and say good luck, right? And walk away and just say you're on your own. You onboard them, you train them, you performance manage them. And that thought process needs to be applied to agents in the same way. You need to train the agent. You need to direct the agent. And you need to do this continuously. You need to understand the output of the agent. Is it giving good accurate answers? Is it solving the business problems? If it's not, you need to continue to test and tune it to make sure it's delivering the best business value. And this is a process that never stops. It's going to be iterative and it's going to be continuous. And finally, measure the value and impact of it. And this really connects to making sure you set out the right vision for your company and what you want to achieve and you are tied to the right use case. You set a goal for what you're trying to achieve and measure it. And it's not just about measuring the business impact, but it's about measuring the efficacy and the effectiveness of the agents.
And so foundational to this strategy is the new way of working. And so when you become an agentic enterprise, you as a company are going to actually find that barriers that existed between your teams no longer exist before. And I think having just the three of us up here talking to you about it is a great example. Whether you are a business owner who's defining the vision for what you want your transformation to look like or you're part of the technology team that is responsible for building and deploying an agent or you're an end-user who's testing it, validating it, giving the feedback or the teams that are focused on measuring it. You all have to work together very close-knit and in a continuous process to make the agents better and to deliver more scale.
I love this slide. This is our agentic enterprise. And so I talked with all of you earlier about the hundred agents we have. We deprecated a lot of the agents because they weren't driving business value and we approached our agentic transformation in a very different way. We thought about it in sort of four main pillars. And the first is really around improving, and when I say four pillars, these are four sort of goals and visions. So tying back to that kind of the playbook slide, these are the vision that we set out for the different areas we wanted to help. The first is improving the employee experience. And so think about there's a lot of things that every single employee at your company does on repeat. And we wanted to bring a better experience to help them take away red tape from their work and to deliver a better experience. I'll give a couple examples of some of these. The IT support is a great one. You know, you guys, employees contact it a lot for questions. For me, for example, I got a new iPhone a couple weeks ago, and we have this technology on our phones called MDM software. And before when I would used to get a new iPhone, I would have had to wait till Monday and pick up the phone and call somebody in IT and ask them to deregister my old phone from the MDM, reregister my new one. I was able to do that in five minutes over a Saturday using our IT support agents. The second is we are focused on enabling our sellers. We're focused on delivering agents into the flow of work to our employees to make them more knowledgeable and more equipped. And we're going to talk a little bit more about that. The third is focused on streamlining our external engagement. And so we have a lot of times where we engage with you as customers, you and partners, and we want to make that the most delightful, beneficial experience for you. And we think about how we deploy agents into that experience as well to make your engagement with us even better. Hopefully, some of you are going to use the Dreamforce agent and the events in the events app, and you can go ahead and ask it. So test it out and see how it works. It's a great way to really test the technology. And finally, and this is something that I think most people know about, but agents for us in becoming agentic enterprise is about delivering more operational efficiencies. How do we drive more productivity? How do we drive more cost savings that really help and benefit the bottom line for you as an organization?
So today we're here to focus primarily on sales and the agentic opportunity for us as a company is massive within our sales organization. We have 15,000 sellers here at Salesforce. In fact, 30% of our employees are either in sales or in a role supporting our sales organization, but our analysis has shown that 76% of their time is focused on non-customer-facing, non-selling activities. And that is ripe for an agentic transformation. And it gives us the ability to help reclaim selling time to give our sellers time back with customers, which is where we want them, not doing admin work behind the scenes. The other great opportunity for us is really focused on driving revenue growth. We know that our top sellers have 4x deal velocity of all of their peers and we need to learn and deeply understand what do our best sellers do and how do we deliver that at scale for all of our sellers globally. So the opportunity for us is both about reclaiming time, driving productivity for our sellers, but actually driving true revenue growth as well. And that's what we've done. We have begun to completely identify our sales cycle. Our goals are to take what our best sellers are doing and bring it to every one of our sellers around the world. And this now all of our sellers throughout their sales process have an agent working hand-in-hand with them, guiding them, nudging them, providing with the best information, enabling them to have the most meaningful conversations with their customers. And I think now is the perfect time for me to hand it over to Adam to actually give us some more specific examples of what we're actually doing within his organization.
A
Adam Johnson14:18
Awesome. Thank you so much, Marcy. Good morning, everybody. My name is Adam Johnson. I'm the SVP of sales for the small business industries team here at Salesforce in North America. In addition to that, I also help lead our customer zero initiatives to use our tools and technologies out in the field. I'm really excited about this because it's a topic I'm very passionate about and I also think it's one of the most exciting times ever to be a business leader because of the opportunity that we all have to reimagine the way that our teams work and the way we empower them with tools and technology. So over the last year our SMB group has grown headcount tremendously all while managing a really large customer base, significant lead volume and a huge number of transactions. And as you can imagine that creates a lot of challenges for us like how do we create personalized engagement at scale? Which processes do we choose to automate and augment with this amazing new technology that's become available? And we've done this while also onboarding and upskilling a huge set of salespeople out in the field and a business that's moving really quickly. And what's most exciting about this is that we've been able to partner together not just to come up with ways that we might be able to address these issues, but we've actually improved rep productivity over the last year while increasing headcount significantly by deploying our own internal agents. If you take the SDR agent as an example, we've now gotten to the point where we have zero leads left behind. Whereas previously, we would have had thousands of leads that were going untouched. We now have a meaningful two-way interaction between an agent and a prospect. And that agent is engaging one of our human team members at the right time when it's ready for us to schedule a call with them. This is really amazing innovation that's driving fantastic results for us. With our sales agent, we've consolidated over 10 tools into one, making it much easier for our sales teams to find the information they need to get ready for a customer conversation. An example of that is the skill that we added recently, which we call the account POV. So, this is aggregating a bunch of information about our customers in just moments, providing that information to the seller so that they're ready to go have a great conversation with their customer. Today we're going to actually take a look at a few agents in action and the way that our teams are using it and how this is helping us. The first is the SDR agent. The second is the sales agent and the final one is our sales coach. So first let's take a look at how the SDR agent is helping us drive personalized engagement at scale.
N
Narrator16:58
To make sure our sales development reps don't miss a single opportunity, we introduced Agent Force for sales with lead nurturing. In the past, our US sales development reps were only able to follow up on one in every four leads. The other 75% scored too low to justify human outreach. And we knew that was a problem. It meant we were missing opportunities and letting revenue slip away. To solve this, Agent Force for Lead Nurturing now automatically nurtures low-scoring inbound leads around the clock with personalized email outreach and meeting scheduling so that no leads are left behind. The journey wasn't perfect at first. We started with our out-of-the-box email prompts, but we quickly realized the agents weren't generating specific enough responses. A lead would ask about a certain product, and the agent would just give a generic reply. To fix this, we customized our prompts to be more detailed, grounded in data from our own Sales Cloud and Data Cloud. This helped us ensure Agent Force acted as a true trusted extension of our sales team. Today, Agent Force is a 24/7 assistant for our SDRs. When a low-scoring lead comes in, Agent Force automatically sends a personalized email based on their product interest. It also follows our rules of engagement and opt-out preferences, acting as a true member of the team. Once a lead shows enough interest, Agent Force sends a link to schedule a meeting and seamlessly passes them to a human SDR. The result, we've transformed our problem of too many leads into a great opportunity. Our reps can now focus on the leads that are truly ready to buy. And Agent Force ensures every other lead gets the attention it deserves. Learn more at salesforce.com/customer-zero.
A
Adam Johnson18:52
So this is really amazing innovation and this isn't just a marketing video. This is a tool that is out there deployed by us in the field working 24/7 driving real business results. And we're continuing to refine it and improve it so that it can take on more and more work for us over time. And by doing so, we've actually solved a problem that many of our customers face, which is how do you maintain speed to lead with a high-quality two-way interaction between a prospect and your business without relying solely on humans and headcount as the solve to do that. This has been really fantastic and I'm so excited about the promise that this has for our future. Now, we've had the SDR agent book a meeting for us. I'd like to welcome all of you to take a look at how the sales agent is going to get us ready to have that conversation with the customer.
N
Narrator19:42
Giving our sellers actionable data in a presentable format can be a time-consuming process. At least it was until we started using Agent Force to better equip our sellers. Our sellers have started using Sales Agent which can be launched via Slack, CRM, or on mobile. Here they can generate an account point of view or summary, get competitive insights, and get contextual customer stories. With Sales Agent, we've turned questions that can take hours to answer into ones that are resolved in minutes, giving valuable time back to our sellers. Instead of our sellers digging for information, now they can ask Sales Agent to deliver that to them right when and where they need it. For more tenured reps, time is often wasted trying to locate the latest information on competitors and product pricing. With Sales Agent, competitive intelligence and product details are available right at their fingertips by simply submitting a question to the agent. To do this, we prioritized the agent's topics and skills, which are specific tasks the automated agent can handle based on their impact on seller productivity and alignment with our revenue goals. Sales Agent also helps speed up customer and internal meeting prep. With a simple generate an account POV prompt, we're able to reduce the amount of time to prepare an executive brief or a quarterly business review deck by hours. We've learned a lot from our sellers, adapting the agent to work more intuitively and better assist them. Now that you've seen how Agent Force empowers our sales teams, imagine the impact it can have on your own organization. Learn more at salesforce.com/customer-zero.
A
Adam Johnson21:22
So, we're using Data Cloud to create a unified customer profile so that all information from multiple sources like our CRM, other systems, and information that's available on the public web to create a consolidated point of view for our sellers. As we were working to build this, we asked our sellers how long it would take for them to compile all of this information. And the average response was 14 and a half hours. Unfortunately, what this means is that most of the time, our salespeople are probably going into customer conversations unprepared. With the skill inside of Sales Agent, we're able to deliver all of that information to our sellers in just moments. And this has been used on average 4,000 times a month since we launched it by our sales team. So now the SDR agent has booked us a meeting. Sales Agent has given us the information that we need to go have a meaningful conversation with our customer. Let's take a look at how Sales Coach is going to help us practice so that we knock that call out of the park. I'm going to play the role of an AE and I'm preparing for a meeting with my account Omega who is really interested in learning more about Agent Force Sales Coach. Now I'm ready to start refining my pitch for this opportunity and I'd love feedback on this critical deal but my leader is away at a conference this week. Fortunately, I have Agent Force Sales Coach that can help and I can launch it directly from the opportunity record in Salesforce to practice my pitch.
N
Narrator22:52
Hi, since our last conversation, I've been thinking a lot about how we can help you with your surge of outbound sales calls that are customizable to fit your sales culture and methodologies. Would you want to see a demo or learn more? Now, Agent Force Sales Coach is processing my performance to provide individual personalized and actionable feedback using details from the account and opportunity records in our CRM. Now here I see my feedback and includes a deal summary, feedback on my skills, my knowledge of the product, and even additional recommendations and next steps. All of this will help me feel more prepared and confident so I can ace my upcoming meeting with Omega. It also tailored the feedback on things most important to our company's goals. In this instance, Agent Force. Now, all of this is available out of the box and can also be customized to add in more specifics. At Salesforce, we initially rolled it out to 500 sellers and after only two weeks on a pilot, we've rolled it out to our entire sales team. With Agent Force, our sellers can practice their pitch or roleplay to sharpen negotiation skills with personalized unbiased coaching on demand.
A
Adam Johnson24:05
So, as a seller, it's not always easy to get time with your leader to practice. In most sales organizations I've been in, frontline managers are some of the busiest people in the company. And what that means again is that reps are probably going into customer conversations without practicing and making sure that they're on message and making sure that they're nailing it. This is a really important skill, especially as you're hiring and bringing new people into your business, to make sure that they've got your messaging down and ready to go and that they're using the right sales skills in those conversations. And this isn't just something that works at an individual level. We're using this tool at scale. We actually certified our entire go-to-market organization on new product announcements and messaging using the Sales Coach. And I can tell you from personal experience, the feedback is really good. It's very actionable and very precise. In fact, it probably closely mirrors the feedback I would have given myself had I watched the pitch that I put into the system. So, this is really amazing stuff. This is going to help improve win rates. This is going to help shorten cycle times and make sure that your team is all ready to go. We've seen a few agents in action. And now I'm going to hand it over to Andy who's going to talk to us about how we did this and the lessons we learned along the way.
A
Andy White25:18
Thanks, Adam. All right. Hey, my name is Andy White. I've been at Salesforce for 16 years. I am very proud to lead the Salesforce on Salesforce technology team, which means my team's whole job is to make sure that the way Salesforce is using our own technology is the best and that we're an example to all of you that you can follow. We also want to help prototype the path so that you can learn from our mistakes and avoid them. And so, a lot of what I'm going to talk about here is maybe doing a click-down from what you've heard from my team members here today and maybe a little bit of why these became our lessons. So we started seeing these themes show up consistently in all of our agents that we were building. And if you think about them, this is not in a linear order. So when I talk about start small and go fast, testing, and repeat, you've heard some of these ideas earlier. This is something that you're doing throughout the process. Okay, does that make sense? Speed is really important and getting to a point where you can have some velocity so that you can see insights and understand what's happening. We were actually really hesitant to move forward multiple times. Marcy and I had lots of conversations about should we do this? Like the idea of you only get one chance to make a first impression. We had a lot of these same conversations about help.salesforce.com if you've experienced our external facing agent for customer support. And the better answer every time for us, which I would encourage you to do, is to rip the band-aid and experiment. And kind of the idea of learn by doing shows up really big. And Marcy talked about, well, what happens if it makes a mistake? You do the same thing with an agent that you would with a human. You give feedback, you course correct. And I think that's the thing that's really novel here. We have these two long-standing ideas of human performance and technology tooling coming together. And the answer is frequently how you would manage a team member is exactly what you would do with an agent.
Marcy showed that great slide that showed our top agents in the company and how we reduced from a hundred to significantly less. And these are the agents where we've spent the vast majority of our time. And the differentiator between a mediocre or non-value-add agent and a great agent is focus. It all boils down to how much time, effort, and energy are you and your business partners putting into the agent. Right? Makes sense, right? Very straightforward. But apparently it took us six months to get that lesson. So I'm hoping that you will avoid it. We call these our hero agents. And these are the agents where we are putting our most time, effort, and energy. And by making these agents great, it actually improves all of the other agents and raises all the boats. And so part of what we've had to figure out is how are we sharing information across teams. Marcy mentioned that agents actually tend to push the boundaries of your organizational design, Agent Force and Data Cloud really, and it's super interesting like sales and marketing is a very interesting experiment for us where if you think about what the jobs are to be done across sales and marketing they're very fluid but we have silos that have been established so that we can work well and we keep running into issues across salesforce.com and the lead nurturing agent because they want to flow together. And so it's like it's forcing us to work differently internally.
We talked about testing and tuning and why this is so important is that people will use the agent differently than you expected. It's happened every single time with every one of our agents. And so until you get to that launch point when I was talking about like go fast, it's really important because then you start understanding what it's actually doing and how people are using it. And this is a view that we have in Tableau where we're looking at the responses and the questions to the agent. Now, keep in mind from a permissions perspective. Some of these are very controlled and restricted because we have employee success or HR use cases. So, not everyone has access to that data. But we come in here and we look and we understand, oh wow, they're all asking for this thing and we didn't build it. And so, we use that as part of our feedback. And then we feed that information into Data Cloud or we build or refine our instructions. The other thing that I want to give you as a warning is if you have an externally facing agent, you need to be spending time thinking about how it could be abused or manipulated. In full transparency, when we launched help.salesforce.com, we spent very little time thinking about how could someone purposely abuse this agent and try to get it to do things that we don't want it to do. Being honest with you, people love messing with generative AI and we've now had to spend time on all of our external facing agents thinking about how it could be abused or taken advantage of. And then we also look for that in our prompt responses. And so I talked before about how we then use those responses and those utterances to see where we have misses in our data. And this has been a continual process of how do we pull in more of the right data, but it's also cleaning up the data that we have.
Another learning that we had is that we had a whole bunch of password reset instructions in our IT agent. So if you think about it like today, if you go into a knowledge portal and you ask how do I reset my password, there may be an article for your Mac, there may be an article for your phone, there may be an article for Windows, we may have an article where they're all combined together. And so the problem was the agent wasn't hallucinating. It was pulling parts from multiple documents because it wasn't written for a large language model to understand the different document types. So a big part of our focus has been we got to clean up the data we have because we have too much or we don't have enough and this has been critical. And the other thing that's highly beneficial for us and a huge unlock here is Data 360 or what used to be called Data Cloud. So the whole idea, the account POV that was mentioned by Adam where it used to take sellers 18 hours. The reason for that is because it's across a whole bunch of different data repositories and what we're able to do with Data Cloud is pull those into one place make it very easy for our sellers to find through an agent. But we're able to leverage our existing investment. So we're pulling things out of Snowflake. We're pulling things off of the public internet. We're pulling things from our unstructured data within our collaboration tools and that's been super helpful. The other thing that we're doing here is the teams at Salesforce, our best information about what's happening with our products is stored in slides. Okay, that's what our product management team uses. That's really difficult for sellers to know where to find the right information at the right time. But we can ingest those slides where they are in the existing team's flow of work into Data Cloud and then present them and refresh them regularly and then present them to sellers. So sellers always have the right latest information and our product marketing team doesn't have to change their workflow of what they're doing. It's really really been powerful for us.
So the right data is gold. The next one measure value and impact. Marcy talked about having a strategy, having a plan. And I like to say begin with the end in mind, right? Like start out with knowing what you're doing. But you can't measure value and impact if you don't start that way from the beginning. You have to think about if this thing is successful, how are we going to know it? And it's really surprising how late that typically happened for us in the process when we were first working on agents. And so we've really been trying to be more disciplined at the beginning of saying how are we going to measure it? This also starts with thinking about how do you measure what good looks like for your humans today. And again, frequently we weren't measuring it. I'll give a great example about the SDR agent or the lead nurturing agent. One of the things that I love about that, we're using Einstein Activity Capture. So when our human sellers send emails and engage with customers, we have all that information. It's all already going into Sales Cloud. We've always had it, but we weren't looking at it. We did no analysis or forensic information around what do the best sellers do, what do their emails look like, are they using special techniques, and what drives results with customers. We did none of it. And so one of the first things we did when we got serious about the lead nurturing agent is we started looking what do the best humans do and how do we baseline that? And we're like, oh, they had a great opener, they had a great hook, it was really compelling, it was really specific. And so then we started grading human emails and compared that to agentic emails with a goal that within 30 days we wanted our agent emails to be better than our human emails consistently. Does that make sense? And so that baseline was super helpful, but frequently we go into this and we don't have a good baseline. We don't know what good looks like. We're not measuring that. And so an interesting outcome here is we're providing better feedback and metrics of what the human processes are and the agentic. And by the way, our agentic emails beat human performance on day one. We had to the point where we had to rebaseline how we measured the email performance and what good looks like. It's been really interesting.
Marcy talked about this about how teams are working together. And what I wanted to share with you, a really important takeaway is the fact that you should have the business owner like Marcy and the technical delivery team like me and the end user like Adam all together on that planning session of what are we trying to do? What does good look like? How are we going to measure it? And then rapid prototyping where we're sharing that information directly to the end-user that's going to be using the agent or the product. Right? Again, this stuff you're like this is simple. We weren't doing it. And when we meet with customers, they aren't either. So, I'd really encourage you to adopt this method and do rapid prototyping. What I challenge my team with frequently is to share early. I frequently find that technology teams want to wait until it's better. Like, no, we got to bake it. We got to make it better. Just do a simple mockup. And then I would love to encourage you to get these two QR codes where we give you in-depth details about these steps that we've taken, the lessons that we've learned, with even more information. That's what the QR code is on the right. And then the workforce innovation playbook on the left is how we've been thinking about how do we enable and transform our teams. I'll give you an example, but we call it the four Rs. One thing Salesforce, I think, has done really well is embracing AI and leaning into AI shame. The whole idea of like kind of feel guilty that I use this tool to help me be better, so I'm not going to tell anyone. At Salesforce, we embrace it and I say, 'Hey, Marcy, check out this really cool thing I did.' And Adam says, 'Andy, look what I did.' And we learn and grow together.