Powered by Salesforce Agentforce, Plative's Application Agent empowers organizations—including the fictional dog shelter in this demo— to assist more applicants without needing internal staff or caseworkers.
Join James as he guides us through the complete AI-assisted applicant journey and demonstrates the back-end case worker's operations in Service Cloud, now significantly streamlined by Salesforce AI.
#BuiltWithAgentforce
So here we have a fleet of Paws Dog Rescue. See, there's all these dogs waiting to be loved and to be taken care of while they've find their forever home. And down in the corner, we have the button for connecting to an agent. So Robert can hit that agent button. And the plaintiff pause agent is gonna join and the plaintiff pause agent could be called whatever you like and it can greet however you like. And all of that programming is done in plain language. So there's no code or anything required here in order to get it to sound and look and feel like you'd like it to based on your brand and what your nonprofit unique identity is. So. How can I help you? Well, I'm interested. In fostering a dog. Makes sense? And so the agent is going to say, alright, so what's the first thing that I need to do? I need to see if I can identify who this potential applicant is. Do we know who they are or is that going to be a new person to us? Agent response is great. I'd love to help you find a dog to foster. And it wants to know our e-mail address. So we haven't identified ourselves yet. So we can say I'm Robert and my e-mail is. Alright. And so now what it's going to be doing is within Salesforce, it's going to be looking to see do we know anybody with this e-mail address and if not, it's going to go ahead and create an account for them. So we can continue to track our interactions with them. So it's this. Thanks, Robert. Great. And we'd like to know what you're looking for in fostering. So a few different parameters are are suggested here. And again, we can make this whatever you'd like to for your workflow. I'm going to go ahead and copy some text into here. Based on Roberts preferences, he says I'm eager to foster a medium sized dog. I'm in Queens, NY and since I don't have a car, I'm looking to find a pet within 10 miles. A dog between three and five years old, weighing between 40 to 60 lbs. It's playful, loves the outdoors, but also gentle, right? Can you help me find a match? So lots of different preferences here. And one of the great advantages of agent force versus a hard coded bot is that it has the ability to think kind of outside of the box and just do some sort of fuzzy matching with these parameters. So it doesn't necessarily need to meet all of these parameters, but it's going to come our database something. All right. What are some possible options for pets that meet some of these needs here or meet some of these preferences? Amazing. So it's gone ahead and looked at the database and said based on your preferences, we've got three dogs that you could consider here. Bailey, who's a bulldog, five years old, medium size male dog, white Brindle coat, 50 lbs and Syracuse and a little description about the animal as well. So why did they think this would be attentional fit? So it's given us three different options. My eyes are immediately drawn to Scout, who's an Aussie shepherd. I have my Aussie shepherd sitting right below me here. And, you know, I'm a big fan of the breed, so this seems like a good fit right in the center of our kind of weight restrictions. And we know that the Aussie Shepherds are an energetic breed. It finally summarizes with a bit of plain text again about why those three dogs were surfaced here, and it says consider meeting one of these. Well, I'm really interested in Scout. And we could type in #3 we could do whatever to indicate that Scout might be the right pet that we're interested in. Awesome. So now the agent is beginning to create an application on the back end and is capturing the details that we've put in so far. So what it wants to know is a few other things to determine suitability and to begin making that application. So it says even there it's going to start to create an application on our behalf. So Robert says I can foster for up to six months. I wanna foster because I love dogs and I wanna help as many as possible, which is a really common reason why people get involved in dog fostering. I also want a companion for emotional support and as a jogging companion as I'm quite active and I have lots of experience with pets as I used to foster dogs for many years. OK, cool, let's give the agent that information. And see what happens next. Look at that. Cool. So actually it's created an application for us. It's given us an application number, which I'm going to go ahead and copy here. And says what we're doing now. So the next couple of days we're going to be reviewing the application and if it makes sense, we'll arrange for a meeting. And it thanks Robert for his interest in potentially being a foster here. Super cool. So Roberts got a meeting coming up so he's gonna go ahead and end the conversation now and he's going to close this out so that he can get on with his day. We're going to go back into Salesforce now and have a look at what it's like when she logs on for the day. So when she jumps on start work, she's going to go into service console and see all of the different pet applications. So we've been testing this for a little while and different parameters. So you can see there's lots of applications in here. But she notices, ohh, cool. A new application came in this morning and I'm going to go ahead and have a look and see some of the details here. So when she opens up the pet applications, she can see a lot of the stuff that Robert punched in has made its way into these custom fields that we have on the pet application record. So at first glance, she says, OK, yeah, some of this looks good. Yeah, the time commitment works. So overall, yeah, I like the look of this potential applicant. I'm going to go ahead and kick this status into pending documentation, and I'm going to save it. And then she can go back and do the same thing with all of her other applications here. We're gonna go back to Roberts experience. So Fast forward a day or two and Robert remembers. Ohh, yeah, I put in my application for fostering a dog. I wonder where that's at, if he's a little bit like me, which of course I designed this. He is not very good at checking his personal e-mail, very good at checking his work e-mail, but he obviously put his info in as a personal applicant. So he's curious, like what's the status of my application? So he's going to pop open the agent. And here's the agent again, ready to help. And he says alright, I'm inquiring about the status of my pet foster application. Simple as that. The agent comes back and says, hey, we can help you for sure. Just give me your application number. And if you didn't have it, he could punch in his e-mail as well and it would be able to find the case that way. So I have the application number, I wrote it down, so I'm going to put that in. And they just go back and see, alright, let's look at Salesforce, where is the status of this right now? So it says it's currently pending documentation. This is interesting to me. I wonder what documentation that they're looking for so I can actually ask that question. What documentation is outstanding? If I check my e-mail, it would be all in the e-mail as well. But in this case, I haven't looked at it. So I can just ask the agent to tell me what do I need to be doing in order to move this along? And here we go. So the application is currently pending additional documentation. We need a few different items to continue with the review. Some references, a rental agreement to make sure it's OK for pets to be on the premises there, and employment verification to make sure that, you know, they have the means to look after the pet while they're in their custody and says OK. So once we received those documents, we continue with the evaluation of your application. And one of the cool things that we could be doing with this as well as we could actually be using the agent as a way to grab this documentation as well and to pop it into SharePoint. Or box.com or Google Drive or whatever your organization uses. We could be sending it in this way. The sky is really the limit. It could also be living in another system altogether, in which case we'd be able to pull that information in with Data Cloud or with Meal Soft. But anyway, Robert now has the idea of what he needs to be working on in order to go further with his application. So we're going to go ahead and end the conversation here. We're going to go back to Cassidy's experience here and let's Fast forward a little bit and say Robert has submitted a bunch of his applications. We could have that visible here on the record. So you know, here's his landlord letter, here's his references, you name it. And again, we could be tying that in from multiple outside systems if we like. Basically, Cassidy looks at this and says, hey, I really like the look of this application. I think we should elevate this to our senior leadership and take it over to the next step, which is to introduce the pet. Assuming that our leaders are on board. Our leaders are gonna end up looking at a whole bunch of different applications. I don't have the time as Cassidy, to type in a whole bunch of information and kind of like make the case for this particular foster. So what we can actually do is use the AI functions to generate a pet application summary. And what this is gonna do is it's going to pull in all of the different details about Robert, including scraping through his letters and other things like that. And it's going to create a four paragraph summary basically about why we think that this is potentially a good application that we should proceed with, right? Couple of sentences about the applicant himself, some considerations. We actually also came up with the idea of a compatibility score, which is super cool. It kind of matches the pets parameters with the applicants and says, hey, we think this is a good match and why? So if you hit finish, that's going to attend it to the record. And we can also submit this now for approval to the leaders to say yes or no to the application. Or come up with any other steps here. Another really cool time saving element that we've included is because obviously Cassidy's working through a ton of different applications. She's gonna wanna be able to send her yes e-mail directly from the the record rather than going and jumping over into Gmail or wherever to send it off. So she's gonna click this e-mail button. So Robert is right here. We can leave the subject line blank because Einstein's going to take care of that for us. What we can actually do is when we click draft an e-mail with Einstein, we can go down and say, let's use premade instructions. And again, all of this is in the service of making Cassidy more efficient so that she can work through more of these applications and get dogs matched to their foster parents as quickly as possible, right, with the minimum amount of friction and the best experience for the applicant. So we can double check that we have the right record in there. To generate this from, we're gonna hit continue and Einstein's going to do its thing right. It's gonna pull in the important details about the decision and it's going to go ahead and write this for us with just a couple of clicks. Amazing. So it would pull in his name here and say Congrats on your approval and talk about the pet that's involved just to make sure we're all on the same page. And then we can actually just go ahead and send this directly from Salesforce. And so again, you can see the different ways that this is completely deflected any kind of like phone calls or other things like that so that the application moves along quickly and it allows Cassidy to do a lot more with her day in service of the need and the mission of the foster.
Powered by Salesforce Agentforce, Plative's Application Agent empowers organizations—including the fictional dog shelter in this demo— to assist more applicants without needing internal staff or caseworkers.
Join James as he guides us through the complete AI-assisted applicant journey and demonstrates the back-end case worker's operations in Service Cloud, now significantly streamlined by Salesforce AI.
#BuiltWithAgentforce
Are you a nonprofit that’s deals with a lot of applications for your programs?
Well, it turns out that an #agent reccomendation engine would not only enhance your constituent experience, but also reduce your cost to serve. Check this out below with an example of #agentforce in action!
Powered by Salesforce Agentforce, Plative's Application Agent empowers organizations—including the fictional dog shelter in this demo— to assist more applicants without needing internal staff or caseworkers.
Join James as he guides us through the complete AI-assisted applicant journey and demonstrates the back-end case worker's operations in Service Cloud, now significantly streamlined by Salesforce AI.
#BuiltWithAgentforce
What is User-Facing Analytics, explained in under 60 seconds ⏱
Our Head of Developer Advocacy, Viktor Gamov, talks you through user-facing analytics and how it differs from real-time analytics.
With StarTree Cloud's new Free Forever Tier, you can start equipping your users with instant insights at no cost. Get started here: https://stree.ai/3qu9fOY#StarTree#Video#RealTimeAnalytics
🫨 You can't be an "AI-first" customer support company but only give AI to the customers that can shell out an extra $30-50/seat.
If you're on Service Cloud, Zendesk or another tool that charges usage-based pricing or access to AI features, check out this demo on how easy it is to set up Customer Agent.
Customer agent comes with Service Hub (at no additional cost).
If you're paying $1-2 per ticket resolution or some extra AI fee, you can instantly start saving money on HubSpot.
Watch this short demo to see how simple it is to set up.
🧡 🎩 h/t to Jessica Park from HubSpot for this walkthrough.
How an Agent works ? 🤔
🌟Agent generate responses that are consistent with your company’s brand voice and guidelines using trusted business data, handling external data from Data Cloud and much more.
💫Excited to share that I recently completed the Trailmix of Trailblazer Quests and created Custom photo overlay which deepend my understanding on Agentforce...
🌟It was really fun to build first Agent using Agentforce and excited to learn new skills with Salesforce . 🚀
#BuiltWithAgentforce#SFgiveaway#Salesforce#Agentforce#Trailmix#Trailblazer
#Dreamforce already got you dizzy? Don't worry! Here’s your Tuesday guide to all things Certinia! 👀 ⬇️
1️⃣ Stop by Booth #1117 in the Dreamforce Campground 🏕️ to connect with our experts and learn why we won the Salesforce Partner Innovation Award in the High Tech category! 🥇
2️⃣ Chris Petty & Joe Thomas chat at Campground Theater 4 how Salesforce drives services revenue, controls costs, and elevates customer satisfaction using Certinia CS Cloud ➡ https://bit.ly/4dWMWVO
3️⃣ CEO DJ Paoni + PwC Managing Directors Amy Richmond & Jason George discuss transforming into a skills-based organization with Certinia PS Cloud ➡ https://bit.ly/4ee3HM7#CertiniaDF24#DF24#PIA24
Welcome to the 100th edition of our Salesforce newsletter!
It's been an incredible journey. Your feedback is golden, so if you've got ideas for improvement, just hit reply and share your thoughts.
Dive in and explore the curated links I've gathered.
Here's a top 5:
✅ Einstein Copilot vs Complex Queries
✅ How To Use Subflows On Fault Paths
✅ 9 Free Salesforce Apps You Need To Know About in 2024
✅ How to Build a Predictive AI Model in Data Cloud
✅ Salesforce Fact #886 | Check if account description is blank using flow
But that's not all of them.
Check other 15 manually selected links via link in my profile.
Click a Like button or share it if you find it useful.
As we continue to share more insights as proud launch collaborators bringing Salesforce’s Agentforce into the market, we wanted to revisit our very own David Geisinger speaking about Salesforce Data Cloud—the technology that underpins everything Agentforce does.
🎥 Watch the session from Dreamforce 2024 and learn about how we leverage the power of data to drive success with improved data quality, segmentation, and automation by connecting product and sales data in Data Cloud. ⬇️ #DDSalesforcehttps://delo.tt/6045tL9LS
When your team's productivity is suffering due to a lack of information, you need a solution, pronto! SaaS content management and GenAI can quickly help you revolutionize how info flows to your knowledge workers & into your business processes. Our 'coffee chat' webinar series is a great resource, featuring a range of topics to help you get the most out of these tools. Sign up below and kick back for the recent session with Neha Jha & Sriram Akella to see how OpenText Core Content and Core Case can overcome siloed processes and fragmented data. https://lnkd.in/gubs2NkS
(Image credit: Vitaly Gariev, unsplash.com/@silverkblack)
OpenText Content Cloud#knowledgereimagined#GenAI#infomanagement
🐧🚀 - Dreamforce 2024: Key Updates and Insights
Did you miss Dreamforce or would like a summary of the key announcements and insights?
✅ - Agentforce, Prompt Builder, Data Cloud and updates across the platform
✅ - Information on the free AI Certification and Training
✅ - The latest information on the Salesforce Lightning Design System 2 rollout
✅ - Summarising key insights from AI thought-leaders
Check out the post in the comment below for more info! 👇🐧
Senior Nonprofit Client Partner | Salesforce | $12MM SB to Enterprise Sales | 3x Plative "Platee" Award Winner Cultural and Sales Impact
1moSalma Ferdowsi Bryan Rees Michael Shroff Rebecca Temple let me know if you'd like us to show you just how easy this was!