Join Frank Brogie, Director of Product Marketing at Knowify, for a 30-minute webinar introducing Knowify’s new AI connector. The AI connector allows trade contractors to connect their Knowify project management data to AI tools like Claude and ChatGPT.
Frank demonstrates how generic AI tools are limited without business context, then shows real-world examples from customers who have built innovative workflows in just 2-4 weeks of using the connector. These include Luis Linares, from Crane Monsters who created a daily briefing podcast for project updates and a team scoring system for data quality, Billy Clark from Paint Applicators who built comprehensive margin dashboards for sales reps and project managers, and Michael Haff from Castle Electric who automated bonus calculations based on job profitability.
Frank provides step-by-step instructions for connecting Knowify to both Claude and ChatGPT, emphasizes best practices for using the tools effectively, and announces an upcoming prompt library to help customers get started.
Hello, everybody. Thank you so much for joining today's webinar. I'm really excited to be sharing a little bit more about Knowify's AI connector and running through with you all how to turn your job data into real decisions you can trust using Noify and AI. My name is Frank. I'm your host. I'm the director of product marketing and community at Noify. If you're a member of our customer community over on Circle, then you'll probably be used to seeing my face there. Today's webinar will be about thirty minutes. I'm really gonna be focused on our new AI connector and how to use that to go beyond just general purpose AI and use it to capture insights you can use to run your business. We should be able to keep things to about thirty minutes today, and today's webinar will be recorded. So if you miss anything or wanna refer back to it, we will be sending the recording out after the fact. Please feel free as we go to ask questions in the chat. I've got Vin from our support team here, helping out, manning the chat. So as you have questions, feel free to drop them in there. And with that, let's go ahead and get started. So we're going to cover, as I mentioned, the Noify AI connector. We just launched this a few weeks ago, and we'll talk through some, real world use cases for AI and construction and how the AI connector is gonna enable some of those use cases. We'll cover where generic AI, so tools like Clod, ChatGPT, Grok, where they tend to run out of road when it comes to project and financial management, and how to close the data gap using Noify plus AI. I'm really excited today to be sharing a couple real world examples from Noify customers, from contractors who have connected Noify to their AI tools. And just within the first few weeks of using this, they're they built some pretty cool workflows and just show some of the possibilities of what you can do once you're connected. I think people are gonna have some really strong takeaways from that. I'll then do a hands on walk through of how to connect to Knowify to your AI tools So you can follow along, maybe in another tab or another screen, and you should be able to get connected with me here on the call. At the end, we'll open it up for questions. But like I mentioned, Vin's here, to help answer questions in the chat as we go. So with that, let's go ahead and hop in. So for those of you, who may have missed it, we're super excited to be introducing this Noify AI connector. This is available now, for Noify customers. It is going to allow you to connect your Noify account to AI tools like your Claude and ChatGPT that are gonna enable those systems to give you real insights based on your actual project lineup. So all of the jobs that you're managing in Noify, all of your financials, your change orders, your invoicing, all of that information, this really rich dataset that you're building in Noify, we're now going to be able to make that available to tools like Cloud and ChatGPT to help give you some real insights you can use to run your business. This works for those of you who are technically inclined. This works through what's called an MCP server. So what it does is it's kind of a a standard connector that's gonna allow Noify to connect to any of these systems that support that MCP protocol. And so today, the the leading AI tools that support that are Claude, ChatGPT. Grok actually just announced it's supporting it as well. And it's a very fast growing standard in the AI space. So there are other AI tools pretty much every week that are adding this kind of MCP connection. So if you use another tool that's not one of those three, I'd encourage you to to check and see if they support an MCP server. If so, you'll be able to follow these similar kind of steps that we'll talk through to connect Noify to that connector. But that's that's getting in the weeds. Let's let's back up a little bit and talk about, you know, why we've shipped this and and why we're leaning on this so much going forward and why we think it's gonna have a huge, influence for your business. So the primary reason and and what we've seen so far is that general purpose AI has come a long way in the last twelve months, but it's still blind about your business, about what's actually going on on the day to day in your business. So it knows the concepts. It's able to do things like, you know, helping draft emails. It's it's obviously very good at at writing, but it doesn't know about your actual numbers. So while it can answer questions about job costing, WIP, you know, AIA billing, those kind of concepts that are central to your business, It can only answer them in general, because it doesn't have any context for how your business actually works and, what's going on in your business today, last week, and the week prior. So it can't tell you you know, if you ask it to tell you the margin on your latest job or maybe how you're tracking towards labor budget. Obviously, it doesn't have any of the connection to, to help you, answer those questions. So it's very quickly gonna run out of road and kinda leave you with these generic answers. And one of the things that that we've heard from a bunch of customers and even ourselves as we're using these AI tools is as you're able to get generic answers, those are the answers that everybody's getting. And so very quickly, those answers are no longer providing a competitive advantage. And so our goal is to equip those tools with your Notify data to help provide that competitive advantage. And so as you can see here at the bottom, right, that missing piece that we're talking about is context about what's actually going on across your business. If you try to provide that context, you might get stuck, you know, uploading spreadsheets and, you know, copying and pasting information about your business into the chat, but you very quickly realize that providing all of that context on an ongoing basis over and over and over again starts to take up a lot of your time. And so we'll talk through how you can avoid that going forward. As you know, as you're all are using Noify, Noify is your hub for your project execution and your financial data. So as you use Noify and you use Noify to manage your jobs, you're actually building this this hub, this single source of truth for everything that drives your margins. So, you know, from the entire job life cycle, you know, estimating and proposals, your project management, all the way through to time tracking and job costing, All of that data is being built for each individual job and housed in Noify, which is incredibly valuable, especially for for AI tools that are constantly looking for data to analyze. And so you can almost think of Noify today as not only the software that you use, but it's also this hub of data about all of the work that you do. Right? So you can see here on the right just some examples of the type of data that is housed here in Noify. So if you're using Noify, you know, on a regular basis, you understand this because it's, you know, maybe how you do some of your reporting today, and it's, you know, the the data that you're feeding in. But what we're able to do now is use all of that data as context to give your AI tools what they need to work smarter. Right? So as you can see from the flowchart here, your nullified data is being built in this no in this nullified system. The AI connector is then gonna provide access to tools like Cloud and ChatGPT so they can access that nullify data. And on the end, you're now able to get real answers and analysis that you can trust and that you can actually act on because it will have to do with what's going on in your business. So I think we've we've covered this conceptually. Let's take a look at what what it looks like in practice. Right? So I think we've all been at this point of a a conversation. This is an an example from from ChatGPT. I just simply ask, you know, what's my WIP? Maybe I'm trying to understand what my WIP position is at the current moment. But on its own, just as a generic tool, it can't tell you much. Right? It can define what WIP means, but as you can see, ChatGPT tells me I don't actually know what your current WIP is based on our conversation. However, once I've connected Noify to my systems, I'll now show you an example from Claude, the game completely changes when these systems have access to my data in Noify. So now when I ask what's my WIP and I'm connected to Noify, I can see some of the analysis that's, you know, that's taken part. So Claude is here looking at all of my active jobs. It understands what's, you know, the the job status, what's bidding, what's pending, what's already closed. And it can give me a a summary of everything going on for all those jobs based on the latest real time data from nullify that it's pulling. Infinitely more valuable than what we saw here just from a generic AI tool. And this is really easy to get to. Obviously, I just asked, you know, what what WIP looks like. This is fantastic for any sort of, you know, ad hoc reporting. If there's any question that you have about your business, all you have to do is ask. And by connecting Notify to to your AI tools, you can get answers like this, you know, in seconds when previously you might have to be digging through dashboards or or pulling reports or building those reports maybe from scratch. So let's take a look, you know, beyond this kind of, ad hoc reporting and, and having questions answered this way. I'm gonna talk you through some of the kinda real world examples from from contractors that I've spoken with in the last couple weeks. All of these examples have been built in the you know, refined over the course of, you know, two to four weeks. No one that we're gonna take a look at today has has actually had Noify connected to their systems more than a month. And so you can just imagine, this is just the very beginning, and I just simply asked, you know, hey. For those of you who have already connected, what's the most impactful workflow that that you've seen so far with Noify connected to your AI tools, and this is what they shared. And it's some pretty creative stuff. So let's jump in first. This is an example from one of our customers, Lewis. He's the president at Crane Monsters. One of his challenges is, as president, he's got a lot of moving parts in the business, and it's not realistic for him to, you know, get a daily snapshot of what's going on on every job site. You know, if he was to do that, it would take him a little bit of work. He was previously at the end of his workday going into Notify, digging through the dashboards, checking on each individual job, and doing this analysis usually between, like, five and six o'clock and then heading home. But connecting Noify to Claude, he's been able to build this daily briefing podcast for his construction projects that gives him an up to date rundown of where every project stands that's generated from his live Noify data that day. So instead of digging through his his project updates and reports, all he has to do is he runs this prompt each day. It pulls it together into a chat, and then he hops in the car. He hits play, and Claude just reads it to him. So he gets his update on his drive home, gets him home on time, and it's it's it's much easier than what he was doing previously. So you can see he's been nice enough to to share a little example here for us. You can see some details on, you know, the day at a glance. Looks like he had seven employees working. He talks about the the time that they've logged, how many photos were uplated and who his most active texts were on that day. Claude's been nice enough to call out flags that, that need to be addressed, and and certain flags that needed to be addressed that were already corrected. It's really condensed. I think he said that that typically, this is, about four to five minutes that he runs daily, and then he does a weekly recap as well on Fridays, but typically runs maybe seven or eight minutes and gives him a, you know, a really detailed look at every individual job. And this is just one prompt. Obviously, it's it's hard to hear, hard here to show, you know, what that audio looks like. But you can imagine this, just kinda being read by by Claude. He joked it's, you know, it feels nice to hear it be read by, some guy in a British accent makes him feel a little bit more official as a business. So this is, you know, one prompt, five minutes. Lewis is getting these updates on his way home every day. Now one of the big challenges let me back up. One of the big challenges for this, obviously, in order to get this kind of daily briefing, your data that's in Noify needs to be really solid. And so he very quickly in the, you know, the first week or so of having this up, he said, you know, this is great conceptually, but I'm actually finding a lot of of data gaps. It turns out that my team isn't uploading progress updates. They're not uploading photos from the job site. They're not adding notes to their their time entries as much as they would like. So his daily briefing was actually revealing a lot of gaps. And so he decided to use Noify and Claude to help close some of those gaps to make this daily briefing even more effective. And so what he did is he he he built a scoring system with Claude that scores his his crew out in the field on their data uploads. So it gives them points for uploading photos. It gives them points for updating job progress in the daily logs. It gives them points for for having accurate time entries and things like that. And they they've been running this on kind of, like, a four week cycle. And you can see here, they've kinda have the data that they need to, you know, reward the best and coach the rest. And so they have a reward system in place that that's constantly reinforced to to have their team kind of improve the data that they're sending in from the field. But, again, this he built this scoring system totally in Claude. Every week, he just says, you know, Claude, run my my my rep scoring for for data accuracy, and it updates their numbers. They send that out on a weekly basis to help keep everybody motivated throughout the month. And then this is an example of of what they shared at end of this kind of four week cycle. So across the board, you can see they're they're making great progress. They're getting much better updates on on phase updates. They're getting, you know and overall, their team score has gone up by by six points. This is something that he's mentioned wanting to do for a really long time and just never had the time. But now there's no extra work that's required. Claude is just looking at the activity that takes place in Noify, assigning these scores, and then sending the update out to the team. So, again, a really creative way to solve, you know, a data problem that he's now able to do, you know, on a on a weekly basis with with just a simple prompt. Now let's jump over, to another business. This is, is Billy. He runs the back office, at Paint Applicators, and he put together a really, comprehensive dashboard that helps measure the margins for all of, their sales reps and their PMs. So it's based on their live Noify data and helps them measure by individual the impact that they're having, on their margins. So let's go ahead and and dive in and take a look here. As I mentioned, he kinda has two versions of this of this dashboard that we'll take a look at, one for his sales team and one for his PMs. So you can see here a really comprehensive dashboard. He's got a a couple of his team members here on the left and some high level summary around contract values, profits that that their jobs have brought in, and they're tracking all these against the estimates. And so this is actually clickable up top, and he can say, you know, I wanna look at just the jobs that have beat their estimates or maybe, you know, take a look at the sixteen jobs that are underperforming based on our estimates. And at a glance, he's getting some quick highlights on on the jobs here. So this is helping his team, direct their attention so they understand what jobs they need to focus on and also understand what jobs are performing, really well. And he's able to kinda create this custom view, for each member of his team. And then a final use case that we'll talk through is from Mike Half. He's the owner and project manager at Castle Electric, and he wanted to put together a bonus calculator similarly based on job margins and profitability for his team. And Mike very quickly ran into one of kind of the roadblocks of of using, like, general purpose AI. So we worked with Claude to figure out, okay. How exactly should I build these bonuses? What does you know, what's an appropriate bonus structure for my team given all these things? And so Claude could could build this SOP. It could build a a recommendation for how to pay out bonuses. But without the AI connector from Noify, he would have just had to stop there, and it would have taken a lot of manual work for him actually to calculate his bonuses. But with the AI connector, he could then point Claude at the actual data in Noify. So Claude now reads each job's financials right in Noify, calculates the bonus that's due for each of, you know, each of the PMs, you know, based on the profitability of the jobs that they manage. And then they actually have an additional step here where they push this into monday dot com, which is kind of their source of truth for their CRM, and that's what they use to then trigger payroll. So it's all kind of automated within this within this one prompt that he's built, which is pretty elegant. So you can see this is an example summary that he set up here. So he could just basically run this bonus review on a quarterly basis. Claude is gonna go through every financial from Noify, and you can see here some examples there. They're they're of the jobs they're looking at. They're looking at the budgets versus actuals, checking out the variance and and the revenue for each job, and then using these based on the SOP to calculate bonuses. So this is something I've heard from from a bunch of of Noify customers is they're interested in bonus payout, sales commissions, and things like that. And so Mike had a really creative solution here. So this is obviously complex. Right? It in this case, it's, you know, three or four steps, and and they actually end up pushing to another system as well. But to close, I wanted to kinda touch on, you know, again, going back to the idea of just having really easy access to data. Not everything needs to be a complex workflow like this. It could be just as valuable, if not more valuable, to just be able to come up with a question about your business, ask it in plain English, and get an answer right away. And so one of the first things he did, it actually may have been the first prompt that he ran once he had know if I connected to Claude. He just wanted to know who his most profitable client was. So he just quickly asked, you know, in the most eighteen months, who's made me the most money? Claude took a look, ranked every client by their profit based on, you know, the latest data from Noify. You can see here about a hundred and eighty six projects. And he told me his his the, you know, the client that ranked most profitable actually surprised him a lot And it was somebody that traditionally was was quite difficult to work with. And he said he would probably treat that customer, you know, maybe a little differently now that he he knows that they're actually his biggest profit driver. So a really meaningful insight, something you could not get just from chatting with with Claude or ChatGPT without the Noify connector and just shows how quickly you can get to insights once you have these systems connected. So we'll spend the next couple minutes actually walking through how to get connected. Hopefully, as we've been running through these these workflows, the gears are turning in your own minds, and you can imagine ways that you might use the nullify data when it's connected to to your AI systems. So let's go ahead and get you connected so you can you can start working. Two, just quick disclaimers as we go. In order to connect your your Notify to to your AI system, you need to have admin permissions in your Notify account. This helps us just ensure that data is only shared with those folks who have permission to to view it. It's something that we'll, you know, be in investigating, making more accessible in the future. But at this stage, it's gonna be just limited to admin users. And two is at this stage, you can see from all these examples, giving no Claude and ChatGPT access to your nullify data at this stage is going to be limited to reading and analyzing your nullify data. We've had lots of requests to let these systems take action in your Notify account on your behalf, things like, you know, create a job, create a follow-up task, things like that. Our team is actively working on that, and we'll be speaking again on another webinar like this in just a few weeks when that's live. But for now, just be thinking about analysis for this. So with that being said, let's talk about how to get connected. First, we'll we'll run through the steps for Claude. So if you are if you are on a team plan, you need to be the, like, the admin of the Claude account. If you're on an individual plan, you can do this on your own in Claude. But you're gonna open up your settings menu and then go to connectors here, and you'll see a little plus icon at the top right with the option to add a custom connector. If you don't see the add custom connector option, it means that you're on a team plan, and you're not the admin of that team plan. So you'll need to find out who the team admin is on your account and and have them take these steps. But you should be able to add the custom connector if you are the administrator of your own plan. From there, Claude is gonna ask you to give your your connector a name and to drop in the URL. This is like the MCP URL. This is the same for all Noify users, so you should just be able to copy this. I think Vin put in the chat a link to our knowledge base article, and this URL is right there. And this is what's gonna give Noify and Cloud that connection. So just give it a name, add the URL, and click add. Once you've done that, you should then see if you open up a chat. You can just confirm, but you wanna make sure that the connector is turned on in that chat. So you can click the little plus button down here at the beginning of a chat, go to connectors, and then make sure that Noify is toggled on. And it's that simple. One, you know, little step that that I missed actually, as I say, is once you hit add here, Claude will prompt you to confirm your like, authenticate the connection by confirming, your nullify credentials. So it'll redirect you to a page where you're you will log in with your nullify credentials, and hit authorize. Once you've done that, then you'll see the option, to toggle the chat, toggle the connector here in the chat. Great. Now let's take a look for for Notify and ChatGPT. So to do this, there's just one additional step, but for the most part, it looks the same. So you're gonna go into your profile settings. You can click on your little avatar and then go to settings. Once you've done that, you're gonna go to apps here on the left, advanced settings, and then you're gonna wanna turn on developer mode. ChatGPT will give you some some disclaimers here about about risk and security. Basically, they're they're letting you know to make sure that you're only adding connectors to systems that you trust. So once you've turned that on, you'll be able to go back and then add a new app. And, again, you just name this one Noify. You can add a description if you'd like, and then the connection URL is gonna be the same as that URL that I just showed there. So assistant dot no fi dot com slash a p I slash v two slash m c p, but you can grab that from the knowledge base article that is linked to the chat. Click understand. Click create. It will then, again, redirect you to confirm your your Noify credentials, you know, basically prove that you're a Noify customer that has admin access. Once you've authenticated that connection, it will then be live. And so similarly, you just wanna enable this in each individual chat, hit the plus. You can go down to more and make sure that Noify is toggled on. And it's that simple. You know, turning these connectors connectors on really takes three or four minutes. Everything else that comes from there is a product of, you know, testing and imagination. So one of the things I'm excited to to share that we'll be launching towards the end of this week, and and I'll share as a a link and a follow-up email, but is a a shared prompt library. And so what we're doing at Noify is chatting with as many customers as we can, getting a sense of how they're using these tools, and then putting together a comprehensive library of lots of different types of prompts that perform analysis on your Noify data. And then with just a couple clicks, you can copy those and paste them in into these kind of tools. The idea is we want to be not only the hub for all this data, but also a resource for you, helping you get the most out of it. So that should be launching at the end of this week, and you can look out for an update on that in an email from us. So we're gonna jump over to to q and a. Before we do so, I I can see some conversation going on in the chat here around around different different AI systems and kinda how to use them and how to get the most out of them. I saw, you know, Ben just answered there about the recommendation to use Claude. That is far and away the the system that we have seen the best success with. It it does the best at at handling and parsing data from these MCP servers. They're actually the ones who who invented this m c MCP protocol and and adopted it first. So they're they're the early movers there. But as you as you use these systems more and more, a couple of best practices. As you are doing analysis that you found to be successful and you wanna run this on an ongoing basis, Claude, you know, other systems have other words for this, but Claude has a a thing called a a skill, which will allow you to save these prompts that you use over and over again just simply as a skill, so that you can just say, you know, run my job profitability skill. And without having to create the whole prompt from scratch every time, Claude will just run that for you kinda remembering the outputs that you're looking for. So we definitely recommend creating some skills as you start to see the outputs that you're looking for. Another useful tool here is using projects in Claude to hold context about your business, about what it is that you're trying to accomplish. If you just work in individual chats, Claude loses context about the way that your business works and and what you're trying to accomplish each time. Putting together a project for maybe, you know, certain function functional areas around your business, whether it's profitability reporting or, you know, maybe, for the examples, you know, sales commissions or AR and AP, putting those projects together will allow Claude to reference the other chats within that project so that it has all the context to be a little bit smarter about the way that that it works. And so I'd recommend, you know, creating some kind of standard templates and and key documents, creating those in a project, and creating a couple of those to manage the way that you work. Obviously, we recommend working from live data wherever possible. So instead of uploading spreadsheets or reports that you pull from Notify, just tell, you know, tell it to view the WIP report, or, you know, measure your job profitability. So we recommend, you know, always working from live data where possible, and be specific with your prompts, and then always verify the information that comes back. Have Claude maybe summarize before you take any action to make sure that that all the data that you see there is is looking correct. Especially in the early days as you understand what data Claude has access to in your Notify account, this is a really important step. So with that, you know, we'll kinda close by by letting you know there's a a a special offer we have right now for those of you who are just getting started with AI tools. Maybe you're on a free plan or you haven't explored Claude. We're giving away Claude Pro. We would really like to help our customers get onboard with with these tools as easily as possible. And so we're giving away six months free of Claude Pro to anybody who basically refers another business who is willing to go and and take a look at nullify. So if there's another contractor you know who you think would get value from nullify, simply refer them to us. You can do so on our website. You can scan the QR code here. They'll have a little form. Tell us a little bit about them. We'll reach out. And when they show up for their demo of Noify, we'll reach back out to you and give you a gift card for six months of Cloud Pro on us. It's our way of saying thank you for sharing the word about Noify, but also our way of helping you hit the ground running and and get started with with AI and Noify. We're really excited about the the future here and about all the possibilities of giving, these systems access to your to nullify data and helping you work smarter with them. And this is our way of, of giving back and helping, helping you get there faster. So scan the QR code there. You'll see the referral code. With that, let's I'm gonna take a look here at the chat. I know Vin's been using the q and a to to help answer answer some questions. Let's see. Yeah. Question here about leveraging the Noify MCP connector with a SharePoint MCP connector for even higher value. Yeah. That's one of the beauties of the the whole MCP concept is that you can use your AI tool. Let's say you're using Claude as your, you know, your hub, basically. And the same way that Mike was doing, you can do some analysis from nullify. And then if you create a document that you then wanna, you know, upload to SharePoint, you can just ask Claude to do that, and it's effectively connected nullify, to your SharePoint. So that's that's some of the beauty of the MCP, and I think it's a really good idea. Do you know if agent skills can interfere with the MCP connector? Specifically, I use KTMAN with Cloud Code. For token usage reduction, I'm worried of possible edge cases resulting for my agent use few token talk like Cloud Code. We haven't explored too much about additional agent skills. I know that those things that are kind of natively built into Claude because Claude works so well with with the MCP, unless there is, you know, a a direct request to interact with the nullify MCP. You know, we we haven't seen instances where, you know, any other system is is is speaking to the MCP through Claude. I'd be really interested, though, if if you're exploring more in that direction and and you run into anything or there are specific things that you're worried about, send me an email. My email is frank at knowify dot com. I'm kinda spearheading a lot of the customer comms around this AI connector, and I'd love to learn more about your specific use case. So please don't be shy. Thanks for the question. Best example for for a GC, one of the I think the beautiful things is gonna be an answer, but but not an answer at the same time, is that there is no single best example. I would start by, you know, one way to get to know what's what's possible is to connect the systems and then just say, tell me a little bit about my Notify account. Tell me a little bit about my active jobs today. And one piece of advice that we got from customer who connected was to just simply say, what's one thing I might not know about, you know, the the way that my jobs are managed or or one thing I could improve about, you know, my jobs in order to improve my margins? And just start small. From there, you'll start to understand where there's data, where there are gaps, and what kind of prompts kinda yield the best results. I would say keep an eye out for our prompt library. I think that'll probably help you get started a little sooner. As Claude Cowork is virtualized, does it have MCP access? Yes. Claude Cowork does it works on the same MCP system as as Claude Chat. So if you're using Cowork, you will be able to to connect it to to the MCP. At this stage, it it still won't have those kind of right capabilities. And so for, like, you know, kinda creating an agent to then go work in nullify, You won't be able to do that yet. And I'm I'm it looks here like you're interested in creating POs from receipts and things like that. We haven't tested specifically, like, using Cloud Cowork and and, like, Cloud and Chrome. I would say sit tight for just a couple weeks, and you're gonna start to see us kinda bring some of those capabilities natively to the MCP. So you don't have to use these kinda workarounds, and you can just ask Claude to do it, and it will just use our MCP directly to do. I can also share, you know, a little bit of a sneak preview for those who have stuck a lot around for so long. We are also building some native AI into Noify that will be, like, a native AI assistant that will be able to take on some work like that directly in your Notify account. Eventually, Claude will just be able to talk to that assistant to take those actions. So it will streamline some of those, like, most common use cases that we see from our customers and actions they wanna take in Notify. But to circle back to to answer your question, I would say, sit tight. We'll be kinda releasing those options in the in the near future. And how does the MCP connections work when when using APIs? As I understand it, the MCP here is it's it it kinda stands alone. And so, you know, you would be using and and accessing your data through the MCP here instead of of through an API. So they're two they're kinda two separate streams. Another is API and the and the URL for the MCP, but but they work differently. If I'm misunderstanding your question or or missing part of it there, feel free to follow-up again. Or if there's a specific use case that you're thinking about, feel free to email me directly at frank at knowify dot com. Also, looks like tons of of info in the chat. I'm gonna give a quick scroll through the chat. I know we're over time. So if if you need to drop, feel free. Really appreciate you stopping by. Feel free to reach out to us if you have any questions and need any help getting connected. We'd be more than happy to help you. If you have exciting use cases that you'd love to share, again, email me at frank at noify dot com. I'm always curious to see what our customers are building, and would be really interested to see what you do next. So if you wanna drop, feel free. In the meantime, I'm just gonna take a look through the chat here and see what everybody's been been saying during the webinar so far. But awesome. I'm seeing William, Marcus, Liz. Great to see you guys. Aya, thank you so much for joining. Chris, don't worry. We'll we'll send the the video of this presentation after the call, and it'll also be posted to the Noify website. So it'll be visible there. Lucas, I see a question here about LiveSwitch. I haven't used LiveSwitch. I'd be curious to know if, if you know on their end if, if they support the MCP connections. If so, let me know. What we've seen is each AI system, whether even if they, you know, they support MCP like Cloud and ChatGPT, they support the MCP with varying levels of proficiency. And so we would definitely rank. Cloud handles that data coming through the MCP in a very proficient way. ChatGPT and Grok run into some stumbling blocks sometimes, so they would be a little bit less proficient, but they can still access the data, and you'll start to kinda understand where they're strong. So LiveSwitch, I haven't looked into specifically. If you're testing there, please let me know, and I'll I'll make a point to take a look, you know, separate from this as well. Mike, curious about ad hoc reporting. I used, yeah, that question that term to kinda just mean, like, if you have a question on the fly, it's not something that's systematic and built into your process. But it's four o'clock, you'd like an answer by four zero five, and you just need something quick. That's what I meant by by ad hoc. So not something that's that's part of your maybe recurring process. Just a question about your business. Aaron, I hope you're able to get some step by step help on getting the workflow set up. We we provided the the link to that knowledge base article that should help you actually get set up. If you're interested in setting up any of those specific workflows that that I shared from from those customer examples, Let me know. We'll probably, over the course of the next couple weeks, be be sharing a little bit more details about those, and I'm hoping that some of those customers might be willing to share some of the prompts that they've used. For now, they're all still works in progress, so they're not that easy to share. But we hope to be adding those to the prompt library. So I'd say keep your eye out for that prompt library at the end of the week. Aaron. Okay. Got it. So you're looking for that specific, you know, prompts that that Lewis is using to read out that that daily or weekly reporting. I can tell you from my conversation with Lewis, his prompt is actually not as it's not as complex as you might think. So if you're already connected, I'd say in the background, ask it for, you know, a recap of of what's happened, you know, updates that that have happened across all of your jobs. I would say maybe start with the week, and just give it some instruction on the things that you're specifically interested in in seeing. I know everybody tracks different things when it comes to their job. So if you're interested in production tracking that's coming from your daily logs, tell it to focus on that. If you're more interested in how your margins are trending or, you know, the amount of labor that you've consumed against your budget, ask about that. And you can really be that simple. Like, it just could be give me a weekly briefing on you know, across all of my active jobs, focus on these five things. I think you'll find that that will get you eighty percent of the way there, and then you can make some small tweaks over time to, you know, to to hone it in for exactly the type of report that you're looking for. Once you've done that, I would encourage you to just ask Claude, hey. Save this as a skill for me. That way, you don't have to run that prompt every time. You just have it saved as a skill. Awesome. Looking out through, Corey, Marcus, great to see you guys on. Thank you so much for joining. Marcus, looks like you've a a great stack, OpenClaw, or mez, OpenRouter. You know, that's great. You should be able to find that URL in the knowledge base article, but that's an interesting stack. It's helpful to see what other tools people are using. Great. Yeah. We definitely recommend Claude over chat GPT and Croc. If you haven't used it, like I said, feel free to use that referral link. I will back up here so you can see it. We'd love to give away, Claude. So, if you refer a friend, we'd love to get you set up for free. Awesome. Well, that's about, all the the questions that we have time for. Thank you all for, for joining, for spending a little bit of extra time with us. Like I said, my name is Frank. You can reach me at frank at knowify dot com, with any questions or specific, workflows that you're interested in in getting set up. And you can look for a link from us towards the end of this week with the recording and a link to that prompt library for getting started. It's been a pleasure speaking with you all virtually and look forward to our paths crossing again. Thank you so much.
