[00:00:00] Speaker 1: Do you want to see how SlackBot can help marketers? Watch this. Hello and welcome back to Slack School. I'm your host, Mike Reynolds. I'm part of the Slack team here at Salesforce. And today we're going to take a look at how SlackBot can help out marketers. We're going to do this with an interesting example. It's something I've never done before, which is PR. They don't let me talk to people. We're going to pretend we work at AW Computing. And there's been a pretty serious problem based on an update that we've made. I've got a lot of impacted customers. My normal PR team is out. And it's now my job to figure out how we should say what we need to say. We're in trouble, but SlackBot's going to save us. Of course, you can try this on your own, but you would need a connected Salesforce instance and a Slack workspace. You can get one of those on Trailhead if you go over to the Getting Connected module that I can link to you here. All right, let's get going. What we know is that there's been a problem. Now, I could open Salesforce to see a case that is part of this, but I don't want to. I'm going to stay right here in Slack. I can see the important details of this case. I have a SEV1 incident. Our A200 model desktop's got a blue screen on bluting. On bluting? I'm going to go with it. On bluting. Now, I don't really know what this means. I don't know how to solve the problem. It's going to be difficult for me to understand because it's way out of my wheelhouse. But fortunately for me, I've got SlackBot. I'm going to adjust the sizes of my screen just to make sure that we can see the bits that matter. And then let's start having a conversation with SlackBot and just, we know where we're going. We're going to see how this goes. I don't have to have a perfect prompt because SlackBot just sort of understands what's going on here. You can also see right here, there's a little ghost image where it says, ask me anything about, and then it gives you the case number because SlackBot is contextually aware of what I'm looking at. So it knows that this is a case. It knows that this is associated with a particular record in Salesforce. That's going to help save me some time. So let's get going. So I've kind of teed up this situation for SlackBot. I've been very transparent. First, I want to understand this issue. I don't know how to do the PR stuff, so I'm going to need some guidance there. And I want SlackBot to just start by doing a lot of research and helping me understand exactly what's going on. So here we can see, it started out with a query to Salesforce. It's checking messages, and then it's figured out that it needs to do a more detailed query around the specific problem that it found in this case. So it's founding a lot, it's founding? It found a lot of messages. Now it's going to pull through all of those. It's doing another query, and now an analysis of that data to give me what is clearly a pretty significant incident. So our computers are not booting up, and it's given me a link here, which is actually going to take me into Salesforce to show me where it's getting this information from. This is a feature of SlackBot that I really love, because what it does is it helps me be an important part of the process. Maybe there's a hallucination. I can go look at this record in Salesforce and check and see if that's really happening. Or I could view the record in Slack. Either way, I can make sure that my AI is not misleading me. All right, now SlackBot's giving me a little bit of guidance around the idea of PR and the messaging. Now, I actually have two separate messages that I need to create. One of them is going to be for our status website. That's going to be a fairly tight message. It's not going to have a lot of feeling to it. It's going to be very factual. I call this type of thing the associated press level of communication. Just facts, nothing more. So we needed to draft that. I also need to understand a little bit more about what a good message is. It's told me some, but I would like to understand a bit more. Now I'm going to want to work on this, and I'm going to want to share it with a lot of other people. So I'm also going to have SlackBot put this into a canvas. The second PR message that I need to create is going to be one that I'm going to use in a campaign that SlackBot can help me build. So we'll have it draft that at the same time, and again, we'll have it added to the canvas. SlackBot is now going to do a bunch of research again, make sure that it has all of the information it needs, create the canvas, and then start populating it with all of the information that I want. That's going to be very helpful. I'll be able to share that with the other marketers that I'm working with, the people that are providing the service, getting the problem corrected. They'll also be able to see it, and then they can put updates. We can all work in the same place. That's going to be very helpful to make sure that the whole process is nice and streamlined. Let's see what we got. Again, did another query of Salesforce, and then it created this canvas, SEV1 Incident PR Messaging, and we can open that up in the sidebar, and now I've got my canvas that I'm going to work on and my conversation with SlackBot. This is going to help it be nice and easy for me. Always love seeing this at the top, reminding me that there could be inaccuracies in this, and I always want to check my work. We've got a draft, it's pending review, under investigation, there's a boot failure, there's a specific product, we state the severity, the date identified, and exactly what we're working on. This seems pretty good. What's nice about this is because it's in a canvas, I can edit directly, as can my colleagues, if I can bring other people in to help me with this. Let's check out the other campaign message. This one's also really good. We've got some placeholders for merge fields in here, and this is a much more empathetic message. Again, it's good on details, with more details in here. Some next steps, and it's also given me some notes to review. Again, it told me about those merge fields with the placeholders. This is great. Let's take this a step farther. If I'm going to be bringing this into a campaign, I'm probably going to want to add some campaign members. Let's see if Slackbot can identify all of the impacted accounts and the people who opened the cases. What I would expect to see here is that Slackbot's going to do another query, although this one's a bit tricky, because it's going to have to check the case object and the contact object and the account object. It's a lot of looking, and it's a lot of things to keep track of. Let's see what we got. It's found 38 impacted cases, and it's doing that using the description. We don't know for sure amongst all of our open cases which are parts of this, but Slackbot is reading them all and finding them for us. It's building a customer list, and it's going to add this into the canvas now. This is pretty fantastic. It did update the canvas. Here's my list. It's got my account, my contact, and it's found their email address and given me the case number that is associated with everything. That's all I would need, and I've got that for all of my cases. I can see I do have some missing email addresses in there, so that's something for me to solve, but Slackbot has saved me a ton of time here. It's also summarized exactly what it put in here, and a couple of things here. It is flagging me, hey, we've got some email addresses missing, and there's a couple of these customers who've got two cases, and it's saying maybe we need to reach out to them specifically and just make sure that we know that we're on the issue. Well, there you have it. Slackbot can help us eliminate our blank page problem, can do a bunch of data research for us, and can really, really help you understand exactly what you should focus on. Notice how throughout that whole process, Slackbot was saying things to me like, hey, consider this too, and have you thought about also maybe doing this? These are things that are going to help you that are grounded in your data. It's not just giving you a wild idea. It's giving you something that's grounded. Hey, some of your customers actually opened two cases for this SEV1 incident. Maybe it's worth reaching out to them. That type of insight can help you save customers, not just your own time, and that's pretty cool. Yeah, it's pretty cool. Don't forget to like and subscribe, and jump into the Slack community at slackcommunity.com, join the Slack School channel, and let me know what episodes you would like to see me do next. Maybe you'll win a HODL record. You never know. We'll see you next time. Hey, you're awesome. Don't let anyone tell you otherwise. So much hair everywhere. Why are you both so furry? Should have gotten naked mole rat cats.
We’re Ready to Help
Call or Book a Meeting Now