Transcript for "Steal Zapier's AI Workflows: How 8 People Run Finance for a $5B Company": Alright. Welcome. Welcome, everyone, to today's webinar. I can see some folks trickling in. Hope everyone's having a good day. Good morning if you're on the West Coast. Good afternoon if you're on the East Coast or wherever you might be, somewhere in between or around the world. Would actually love to learn where everyone's from. The chat will be muted for today's webinar, but we do have a q and a feature at the top right of your screen. If you click on that, would love to hear who you are, where you're from, what you do, anything like that. Mahen from California. Nice. I'm originally from California as well. It's awesome to hear. We've got Elizabeth from Raleigh, North Carolina, accounting manager. Awesome. Daniel, VP of finance from Long Island. Very nice. I'm in New York right now. We got John from Franklin, Massachusetts, Christopher from the Bay Area. Love to see it. Love to see it. Alright. Cool. Well, now that you've all found that q and a feature, what we're gonna go ahead and do from this point forward is as more people continue to trickle in, I see that number continue to rise, I want to kick it off and just say welcome to everyone to today's webinar. We're talking about Zapier's AI workflows for accounting. Some really, really awesome guests in Miranda Connolly, the senior director of accounting, and Luis, who is a revenue manager, at, Zapier. So one, again, quick housekeeping item like I mentioned before, chat will be muted. But if you just joined, you should see a q and a feature at the top right of your screen. And you can ask any and all questions. Don't be shy. We actually have Luis and Ezra, another accounting manager from Zapier here to go ahead and answer some of your q and a's by text. So do not be shy in the q and a section. Also, as more people continue to trickle in here in last minute, I want to quickly share a bit about Rippling. So you might know Rippling as just an HR company, but we actually have software solutions not just for HR, but starting from the left, payroll, HR, IT, talent, finance, and custom apps. Now solution, obviously, most pertinent to today's crowd is most likely finance, which itself is an all in one spend management solution. You see, Rippling spend consolidates your expense management, corporate cards, bill pay travel, and very soon procurement into one unified platform with powerful automations and, of course, seamless accounting integrations. And each of these solutions is actually built from the ground up on Rippling's platform, meaning they work, seamlessly together, and they help you automate over 90% of your manual processes and get that end of month close time down. So what really makes Rippling stand out is what follows. You're likely familiar with some sort of diagram like what's on the left. Disconnected, siloed, isolated HR, payroll, and finance tools that really can't talk to each other. They have unique data formats, very brutal integrations that break as you scale. But by contrast on the right side, Rippling is built entirely around your employee data. It's always on real time, and most importantly, interconnected across every aspect of HR, IT, payroll and finance. Lastly, this leads to a ton of unique use cases that save you time and money, like auto locking employee cards when they're missing more than x number of receipt submissions. This is actually super popular amongst controllers using rippling or blocking out of policy spend in real time, or even using AI, today's hot topic, to turn receipt photos into ready to submit expenses. So the last thing I'll say before we dive in to today's juicy content is if you're interested in a one on one consultation to see how rippling spend can be customized for your business, go ahead and click the link that will be posted in the chat right now. And if you attend this demo, we'll send you a $100 gift card. Now with that said, let's go ahead and dive into today's content. Alright. Alright. Once again, welcome everyone to today's webinar. I'm really excited to be joined by the Zapier team. We have Miranda and Luis today, and they've been able to use Zapier and AI to actually reduce their month end close time by 25%, And this is also despite a 43% increase in their technical accounting complexity. And also, it includes implementing external audits with a big four firm. So while there's a lot of things going on, lots of moving pieces, they've actually been able to drive more efficiency through some of the workflows they're gonna talk about today. They've also done all this while improving the accuracy of their financials without having to backfill any roles after departures and with a lean team of about eight full time employees. And keep in mind, this is, you know, a $5,000,000,000 company, so it is complex. There is a lot to work through, and so I'm really excited to share what they've been working on today. So Luis and Miranda, if you wanna give maybe a quick introduction for yourselves, for the audience, and then we'll dive into a quick interview portion with, Miranda first. Sure. I'll go first. My name is Miranda Conley. I'm the senior director at Zapier. I've been here for almost six years and have a very interesting background in a lot of different industries. And, but Zapier is, an amazing automation platform and, pretty much my dream job. Awesome. Luis? I'm gonna echo on what Miranda said. Yeah. It's working at Zapier is awesome. But, yeah, I'm the revenue manager here. I've been here almost four years now. My background's in public accounting, so a little bit different. But, yeah, I think I think Zapier has been one of the best career choices I've made for sure. That's awesome to hear from both of you. So we're gonna demo some really cool use cases again with Zapier and AI, but before we dive into those, I wanted to speak to you, Miranda, a little bit more about Zapier's shift to like an AI first mandate, how you're running this accounting function, and before we even get into that, I know you're kind of a self proclaimed efficiency and process improvement junkie, which I think is an awesome term. It tells me you can't hop help but optimize everything that you do. So I'm curious, outside of accounting, are there any non accounting related AI rabbit holes that you've kind of gone down the last twelve months that just really had your full attention? Oh, absolutely. I've gone down my fair share of rabbit holes, non accounting related. My favorite lately, think, has to be my soap making side business. It started when I received an order for 600 bars of custom soap. And I was like, oh my goodness, how am I going to do this to scale? And so I asked myself, can AI kind of help me tidy up a few steps, make it easier to scale? And the next thing I knew, I had optimized recipes for better lather, generated mockups of what the soaps would look like, cranked out marketing descriptions for my website and even drafted social posts. So the whole gamut, soap making is creative and it's a hands on hobby, but I somehow turned it into this little mini manufacturing and marketing operation with AI as my copilot. And that's what really hooked me. And it proved that AI isn't just for technical workflows. It can make even the most random hobby, more fun, more efficient, and way easier to accomplish. Absolutely. Wow. 600 bars of soap. That's pretty incredible. That's more than a side hustle for sure. That's like you said, that's a full on manufacturing operation, and I'm sure your your accounting for your soap business is probably top notch as well. So Absolutely. And a lot simpler. Yeah. So are you using ChatGPT to help me with those recipes, like Nano Banana for those mock ups? What are you using? I haven't tried Nano Banana yet, but so far ChatGPT has been everything I need to handle. It analyzes the recipes, it helps me tweak them, it helps me to scale them, and it generates the mockups because we'll, we'll work on a recipe together and then I'll say, well, what, what should this end result look like? And it'll give me mockups for it. And Chatuchipi Tea has basically become my indispensable soap making sidekick. Yeah. One of the use cases for Chatuchipi Tea they probably didn't see coming, but that's awesome. So I also wanna talk about moving into Zapier and the AI first mandate. You now require a 100% of new hires to be AI fluent, and Rippling also did a separate webinar with Zapier talking about your AI fluency rubric when you're hiring. So I wanna talk about this within the context of accounting specifically. So for your accounting team, what does AI fluency actually look like for the audience? What's the difference between, let's say an accountant who casually uses ChatGPT versus someone who's truly AI ready in your eyes? Yeah. When, when we're hiring, we're not looking for someone who just uses ChatGPT sometimes, like as a ramped up Google. We're looking for someone who treats AI kind of like a junior analyst, gives it direction, checks its work, builds it into repeatable, scalable processes. Curiosity and willingness to experiment are really important here. First they need to think in terms of systems. They look at, accounting workflows and immediately ask, you know, what part of this should or can AI be doing instead of me? We want people who don't accept the norm as the standard, but naturally look for ways to redesign processes to run more efficiently. And secondly, they can turn messy inputs into structured outputs. That means giving AI the right context, constraints, and examples, so that the output is reliable and auditable, not just interesting. And they can also see the big picture. They can imagine how AI plugs into workflows across Zapier, NetSuite, Sheets, Slack, not just one off prompts, but real end to end processes. So everyone on our team is AI fluent. Even the degree may vary from person to person, but what matters is that the foundational mindset is there. AI isn't a novelty. It's a it's a core part of how we operate. That makes a lot of sense. I definitely also resonate with that third point about seeing the bigger picture, like looking across applications instead of just solely prompting within ChatGPT, which is a great first step, but I think sometimes people forget that you can actually connect all of these tools and they can talk together now and you can recreate these entire workflows, which I'm sure we'll see a bit more of in a second. So as a company, Zapier, you know, putting the foundation in place or the mandate in place to start using is one thing, but adoption is a whole another thing. And I know Zapier hit 89% AI adoption back in April, which is pretty remarkable. How did you actually play how did this actually play out in accounting? How do you get an accounting team to adopt this new kind of technology? And as a leader, how do you coach them up? Yeah, absolutely. And adoption is the biggest thing. Accounting teams don't adopt new technology just because we tell them to. They adopt it when it genuinely makes sense for their work, when it's easier, faster, more accurate. But equally important is giving people the right tools and the bandwidth to explore. So you can't just expect innovation if everyone is buried in manual work. For us, really accelerates when you build confidence. And that usually comes from starting small, little tasks, a few minutes here and there. But they make a big difference. And, and they make things slightly less painful. Those small wins, they really compound. They help people get comfortable using AI. They show what's possible and they lower the perceived risk of trying the next thing. We also model the behavior. So as leaders, the leadership uses AI openly. We share the examples of what we use AI to build, and we celebrate progresses, not just ours, but across the whole team. And we share things that work, but we also share the things that don't work. So the things that don't quite land because those are learning opportunities. And so in accounting where precision is the norm, creating that kind of psychological safety is really critical. And so I would say that's how you coach an existing team. You start small to build confidence. You give them the tools, that are worth using. You create breathing room for them to experiment and you reinforce that AI enhances judgment. It doesn't replace it. Once people feel supported instead of evaluated on it, they lean in. And that's how you can really get an existing team to dive in head first. Yeah. I love a lot of those points. I feel like it also echoes what I've heard from other AI leaders in finance, which there's actually a lot of companies doing kind of hackathons and creating that room to go experiment and I feel like hackathons alleviate the pressure, like you said, to produce something that's gonna work right away. I know at Rippling, we even did like an AI hackathon for the marketing team and so I think creating that space, like you said, is really important. I'm curious what are some maybe concrete examples of starting small that you mentioned to build that confidence for using AI within your team? Like something that audience could go try within a few minutes of this webinar. Yeah, I think the simplest confidence builders are really small automations that remove that everyday friction in your work. Think about those tedious things that you hate to do or that you forget to do. If you're like me, like a zap that looks at your calendar and one day before a particular meeting, it will drop a Slack message into a channel to remind everyone to add items to the agenda. You know, it's a manual thing that, you know, we forget to do, but it's really important. An agent that scans your emails and as it comes in, the agent can quietly get rid of the noise for you. A Zap that catalogs, Slack messages where you're mentioned, and then AI can go decide, Hey, is this something that you need to do? And I, and then go put it on your task list. These are really tiny automations that are easier to build with Zapier, but they create a lot of outsized impact. They're perfect confidence builders. And once people see how easy these wins are, then it gets their juices flowing in their mind about, okay, what can I build next? Yeah, absolutely. I think something, you know, as, as a person who creates content for this accounting and finance audience, something I've done at Rippling is try to create docs that have like step by step instructions on how to use ChatGPT to build a GPT bot for like answering T and E questions. So you can actually plug in your T and E policy into this bot, which we give you step by step instructions for. And then your employees can go to this bot and ask the question and it'll answer their question based on your policy and cite it. And if it really can't answer the question, then only then will it escalate to you. So it saves a ton of Slacks, I think, similar to some of those examples that you mentioned. So the next question I wanna get like a little bit more tactical. You've implemented AI across your accounting function, but what was the single most impactful or important lever that you pulled as a leader to yield the most AI efficiencies for your team? Like, was it a specific tool, particular process or change in how you structured work? No specific tool other than Zapier, of course. But I think the single most important lever wasn't necessarily a specific tool as much as it was just changing the mindset of how to do a task from that perspective to, okay, let's design a system using AI. Once we stopped asking, how do I complete the task and started asking, how can I get AI involved in completing the task? Everything pretty much changed. That mindset shift along with creating bandwidth for the team to experiment. And it created more efficiency than any one tool ever could. Yeah. So when you say like, how can I get AI involved? I kind of imagine your team again, going beyond just that prompting. I'm curious, what are some of the more maybe advanced techniques one could say that you're using? Like maybe agent mode or tying things together with some API calls? Yeah, absolutely. We didn't, I don't think anyone on our team stayed in that just using ChatGPT as, as Google for very long. Once, once you shift into a design, the system mindset, you just naturally move beyond the one off prompting. And the team starts building in workflows where AI handles real steps in the process, not just drafting the text. We use agents to extract key data and make calculations. Luis has an excellent example of that one. We have Zaps that connect our tech stack and act on real data. And we have webhooks to pull and pull, pull and push data instantly. So our automations are always reacting real time in the moment. That's amazing. That sounds really cool. I'm sure we'll see some more of that in the demos very soon as well. Last few questions here for you, Miranda. I think everyone loves to talk about AI wins and all the efficiency that they're driving, but I guess for people getting started, I love to talk about the failures. Like what are some of the most common AI projects that look really promising from an accounting and finance perspective but are actually terrible ideas where you're kind of just waiting for a disaster to happen when you see someone trying to do this. Absolutely. Especially in accounting. There are plenty of AI ideas that sound great for accounting, but they can fall apart fast. So anything that tries to bypass judgment or controls gets really risky. AI can book the entries, but someone still needs to review and approve them in order to make sure that we have proper controls and that we've got, you know, integrity in our financial statements. Any AI initiative where financial decisions or actions are being taken still needs a human validating the accuracy. And that's why Zapier has a human in the loop step for those control freaks like me. AI is there to assist us, not replace us. And we are still accountable for what's in our books regardless of how it got there. Yeah. That makes sense. I think the human review layer is probably something that won't go away for a while. You always need that extra set of eyes and kinda goes back to your first point of like treating it like a junior analyst and making sure that, yeah, it can still do the work, but you always still review it before anything goes out the door. So we're about to dive into some Zapier use case here, but I'm curious before we do that, what are some non excuse me, Zapier. What are some non Zapier AI tools or use cases that you love for accounting? You know, what's your personal AI toolkit that every accountant should be using regardless of whether or not they're on Zapier? Yeah. And that's a hard question for me answer. But honestly, my personal AI tool toolkit is really simple. I use two things, Zapier and ChatGPT. And the reason why it is that way is because between the two of these, they cover everything that I need. So Zapier covers ever anything for me that is like repeatable, predictive, has some sort of a trigger to it. And anything that I can write and, and, like an if then statement for, which is how my mind thinks. I think in terms of if then this statements, And then ChatGPT is my ad hoc Swiss army knife. That's what I use to help me draft memos, summarize policies, convert messy inputs into structured outputs. If I need to think through a process, improve some clarity or turn a vague idea into something concrete, chat GPT is my go to for that. Love it. Alright. Last question before we dive into the demos here. This is for everyone listening who feels maybe a little bit behind or overwhelmed or not sure where to start. So excuse me. If someone can do only one thing after this webinar to start their AI journey in accounting, What's the single easiest to start thing that you would tell them to do? Not go run a hackathon or get executive buy in, but something they can do Monday morning, 9AM. What do you think it is? Think if anybody feels anything remotely like I do, you probably get overwhelmed by the comms. Email is all over the place. Slack is all over the place. So if you can do one thing after this webinar, I would start by automating your communication traffic. As the years passed, I found that I spend more and more time in my email and now Slack too, so now I've got two of those. So in reality, if you can just automate that piece, that will give you a lot of time back. So building that automation of an email traffic control on Zapier, it's simple. It's a high leverage first step. You can use it to route filter, summarize file messages away so that you don't even hear the noise to begin with. You can even have it draft responses to emails that you need to respond to, and then you can go send it later. So that single change gives you back bandwidth and you can reinvest that more strategic and judgment heavy work. It's an easy lift and it delivers instant impact. Yeah. I feel like it's almost a positive feedback loop too, because as you're experimenting, you're getting that bandwidth back from this first project you're working on, then you can spend more of that time to work on more projects and get even more time back and so it really compounds. So with that said, let's take a look at how account teams can use Zapier to automate accounting, specifically how they can streamline insights, improve collections and accounts receivable, intelligence, and more. So Luis, I'm gonna hand the stage over to you. Would love to see what you have to share with us today. Awesome. Yeah. Let me go ahead and share my screen. You guys let me know if you can see it. Yep. Awesome. So yeah. So my name is Luis, the revenue manager here at Zapier. And, yeah, today, I'm gonna present a couple use cases, one for streamlining for faster insights, and then one for an AR intelligence automation that we use to just handle all internal, like, stakeholder questions that we get, related to our billing system. So I'm gonna start off with the one related to, streamlining for FasterInsights. So at a simple level, what you see on your screen is, like, a real time monitoring workflow that looks at every journal entry that's posted into NetSuite to ensure that it meets our standard, for our internal chart up counts and all associated rules. So it validates such as things such as department, memos, projects, and just coding accuracy in general. So this review process used to take us twelve hours a month, to do. Now it only takes three hours a month, and every alert, that generates, creates a permanent record in the internal Zapier table for SOX and audit compliance. So how does this work? Basically, it's triggered every time we get a new, journal entry that's posted into NetSuite. We have a a table that keeps, track of all these, records that have already been reviewed and posted. So it filters out these specific records. It filters out other high volume entries like revenue, and then just any entries that are posted in prior months. Luis, do you think could I ask you to maybe zoom in a little bit? Yeah. For sure. Thank you. Is this better? Maybe one more click if possible. I don't know. It'll just be easier to see that way. Cool. Thanks so much. And then yeah. And then it runs into this agent that basically has is given a set of rules such as, like, removing spaces from employee names, obtaining the Slack ID for the for the employees that posted and approved the entry. And the reason we do this is we want to send this notification in Slack and tag the approver and the reviewer so then that way they can go in, see what needs to be updated, and just, like, action it immediately. Once it does that, there's this looping step that basically takes all the information from the journal entry in NetSuite and just standardizes that journal entry for further review down down the down this workflow, which is the next step, in this AI formatting tool that basically does that for us. It formats all the standardized journal entry information, and then we use that data to look look it up against certain rules that we have in an internal Zapier table. And then if these rules don't meet if if, basically, the stand this journal entry doesn't meet the standard rules in this standardized table that we have for our internal rules. It goes down one of these paths. Like you can see on the screen, it can be like a department mismatch, a missing department, or a missing project. And then once it goes down one of these paths, it'll send a Slack message in, in Slack, tag the approver, the reviewer. It'll have, like, a link to the journal entry, and then people can kinda reply in that thread directly and just kinda, like, triage what the question is, get a response, and then go in and update that journal entry that, needs to be updated for for it to meet kind of our internal, rules for for, here at Zapier for our accounting team. So this is what the workflow looks like. Here is an actual example of one of these notifications. So for example, you see the approver tag, you see the, poster tag, you see the link to the journal entry, and you see specifically what's off here. Like, for example, this was a department mismatch for this GL account. We can't have a top level department, so, you can see kinda like the 10 replies in the threads of everyone trying to figure out what needs to be fixed and and what adjustments have to be made. So that's kind of like high level what our Streamline for Faster Insights workflow looks like here at Zapier. If we're good, I'm gonna move on to the second use case. Well, hang on, Louie. I I would love to jump in here if that's okay with you, Matt. I have to say this this is one of my favorite Zaps that we have. And it goes back to the statistics, Matt, that you were giving at the beginning of of the conversation. Right? So we close, we hard close and we saw close in five days. We hard close in six days. We have added over the years, significant amounts of, subsidiaries, complexity to our financials, everything. And yet we still are able to close facts because this helps us be proactive instead of waiting until we're reviewing financials after close. We're fixing things proactively throughout the month because of this zap. Yeah, a 100%. I think that's like a really big theme as well for any part of your finance tech stack. Like, it help you be more real time, more proactive instead of waiting for that avalanche of things and tasks than stuff to chase down at that end of month. Think rippling spend is definitely aligned with that as well. You know, trying to give you those real time notifications when things are expensed or when there's, you know, you're missing receipt or even even proactively like locking a card if someone's already missing more than, you know, say four receipts to prevent like further buildup. But yeah, I love the productivity of this Zap and this workflow. So excited to see more. Yeah. Thank you for letting me interrupt, Luis. Yeah. No worries. So, yeah, that's that's that workflow. I will say it was created by someone else in our team. Ezra, he couldn't join, but shout out to him. He created this. He's awesome. So I'm gonna move on to our next use case. So this use case is basically, an AR intelligence bot. So, it's been able to save us, like, three hours of, like, manual research and reply, time a week, and it just provides, like, consistent responses and, like, will potentially save us licensing costs in the future related to our billing system, which is Chargebee, since we won't have to add these users into kind of our billing system. We can just have them self serve their questions, via this workflow. So, like, before this automation, like I mentioned, our AR team was handling these questions manually. So we were gonna we would have to, like, get the question, research that in Chargebee, craft a custom response, which was taking us a good bit of time every single week. And basically, what you see on the screen right now is our Canvas tool. Canvas is a product here at Zapier that basically allows you to, create any processes that use several, you know, Zapier products so that you can see you can see, like, a bird's eye view of how everything works. As you can see on the screen, the way this works is we get a question that's asked in this specific channel, and there's four different swim lanes. So you triage the question based on what kind of question it is, whether it's a subscription related question, a customer related question, an invoice related question, or we actually have a swim lane for our prorated credit related questions. So, what happens here at Zapier will, when customers are upgrading, we have to provide, like, an estimated credit on their order form. And, basically, what this, swim lane for prorated credit is doing, it's calculating that all for us without us having to, like, pull up an Excel sheet, get their end dates, start dates, the amounts they paid, and, like, calculate that. It's it's doing all that for us internally. So I'm gonna walk through how one of the swim lane works. I think we'll focus on the prorated credit one because I think that's, like, one of the cooler ones that I've built. Before we dive into this, Luis, can we maybe go back to the canvas view Yeah. And just kinda, like, zoom in and and so is this swim lane that we're about to look at just a a zoomed in version of Yeah. It's just here. Yeah. It's just a zoomed in version of this, swim lane, the bottom, the the top, the far right. The right hand side. Yeah. Far right. Awesome. Also love the reaction emoji customization there. There. Yeah. Exactly. That's kinda like the trigger for all of these. They each have their own like reactions. So when we get a question, you react with a certain type of emoji, and then basically this kicks off the entire workflow. So in that message, we usually get an account ID. That's, like, internal for us here at Zapier, but, like, our billing system doesn't doesn't have this account ID. So that's why this next step is, like, this AI by Zapier step, which takes the question, summarizes that, but then it also takes the account ID, pulls that from the message so that we can go into this internal table that we have and use the account ID to find the customer ID, the subscription ID, what are the other Chargebee details that we need. So then this agent can then run based on the given actions we've given this agent to go into Chargebee, locate that customer. For this specific example, it's a prorated credit, so it would locate that customer, find their subscription, find their end date, find their most recent payment, paid payment that's not like a metered invoice or a a onetime invoice, and then just calculate this prorated credit for us. And one thing that I have added to all of these workflows is this human in the loop step. So just to give us more control over the responses we're giving since since it's, like, a new process for us, we can either approve it, decline it, or we can, like, modify the response if we don't like what the agent spit out. And then once we finally approve it, it'll reply directly in the thread for that message, and we can just close out that request entirely. So that's kind of, like, how this automation works. Although we do use Chargebee here at Zapier, like, we have so many other billing integrations, like Stripe, Xero, QuickBooks. So, like, you can easily replace whatever billing system you're using in in replacement of Chargebee, and you could kinda run a similar workflow for whatever it is you need. So that's kind of, like, what this looks like. I can show you a quick response live. This is what, like, a prorated credit response would look like. Someone drops in the account ID, you know, the estimated upgrade date, and doing the calculation all for us. And, yeah, we we double checked that as this is, like, semi new for us, and, yeah, that's spot on and and didn't need any, like, modification. So that's both of the use cases that I wanted to share. Hopefully, that's useful, and thanks for having me on that. Yeah. Of course. I mean, this is amazing. Miranda, did you have something to add there? No. It just every time I see that bot working in the channel, it makes me smile. Yeah. I'm I'm sure that every time you see it, it's just like all of that potential work that you would have had to do just disappears. You know? It's probably a really good feeling. Yep. Yep. Amazing. It's great for us too. Yeah. I think we actually or I'm actually trying to work on maybe doing this externally, so hopefully that can also work. And we just don't have an internal automation for this. We can also do it externally for our emails that we get from customers. That'd be Yeah. Super Very cool. I also love how you had, like, the human in the loop step. Like you said, it's a new process. So again, that goes back to what Miranda was saying about making sure that we're reviewing everything. We're still getting eyes on everything before it goes out the door, especially for a new process like this. But again, probably so rewarding to see those those correct, like down to the cent, you know? Exactly. Yeah. It's awesome. Cool. Well, thank you so much to both of you for coming on, showing what Zapier has been up to and what the internal team is building to save so much time, stay so lean as a finance team, especially for such a large company. Any parting wisdom, parting advice for the audience here today when it comes to, you know, using AI, getting in there, not being afraid to get started? The only parting advice I would have is just what I had said earlier, start small that gets you comfortable and builds confidence and, and, give bandwidth for everybody to do it. Cause I think that's, what's different about our team. Everybody on our team builds apps. There's not just, there's not a gatekeeper that kind of is, is the roadblock for everything. Everybody has the tools that they need to, to build Zaps and automations. Yeah. And I will say like Zapier does a good job of enabling us to like go out there and experiment with this. Like we have like DFW weeks where like we don't have any meetings. And recently we've spent those weeks, you know, playing around with AI, playing around with automations and, like, just having the time to do that. It's just great. So yeah. Can you can you actually touch on those? You said DFW weeks? Yeah. It's like Deep focus week. Deep focus weeks. Yeah. Okay. I love that. I love that. And is that something you implemented like before this AI mandate or after? It's always been around, I think. Right, Miranda? But we just Yeah. Deep focus week has always been something we haven't have Louise. What do we have it three times a year, four times a year? It's kind of like a hackathon. It's time for, for teams to be able to just go heads down on a project or, you know, work that they want to get done, but that haven't been able to. And so what we do is, company wide, we cancel all reoccurring meetings. And so the past two DFW weeks, we've made it very AI focused, either streamlining something that we've already built because you know, a lot of these things we built, you know, a year ago, two years ago, whatever. And so just taking a look at them to see, you know, how can we make them even better? And some of it's just building new. So like a hackathon Yeah. Yeah. For accounting. Yeah. It's like a a is it is it just within the accounting org or is it it's company wide. Right? Yep. Amazing. Yeah. So it's like a almost like a decentralized hackathon for the whole company, you know, depending on how you use the time. Yep. But it seems like a really strong use case for a DFW is to build some of these workflows. Mhmm. Yeah. That's that's awesome. I think a great alternative to a hackathon, like you said. Well, amazing. Thank you so much for joining us today. Really, really appreciate it. I learned a ton. I think our audience did too. And hopefully to the audience, we'll see you all in the next webinar. Awesome. Thanks for having us, Matt. Thanks, Matt. Thank you so much again to, Miranda and Luis. And before everyone heads out, just wanted to share one last message here from Rippling. I want to remind everyone that we do have a special gift for you. If you do want a one on one consultation to see how Rippling Spend can be personalized to your company. And in exchange for your time in signing up and attending a Rippling Spend demo, we'll send you a $100 gift card. So once again, if you go to the chat, you'll be able to find a link, that will take you to this page where you can sign up, attend a demo, and receive a $100. And, in case you missed it at the beginning, once again, Rippling Spend is a unified platform to manage all of your company's spend, including expenses, reimbursements, corporate cards, bill pay, and payroll. Its powerful automations and consolidated nature have been able to help companies like Andros save a $100,000 a year on staff hours and software licenses, dropping their close time to just three days, and Pepsi or Worcester closed their books seven times faster with a 100% compliance. And Victoria Beck and Beauty, doubling their monthly spend and still being able to analyze and track everything in just fifteen minutes. So again, you can use that link in the chat if you'd like to check out Rippling's spend. And also, we don't want you leaving empty handed. So we have a Google Doc here where you can download some powerful ChatGPD prompts for accounting inspired by actual prompts used by the former controller at OpenAI. Now that'll be in the docs tab on the top right of your screen. I hope everyone enjoyed today's webinar. Thanks again so much to the Zapier team, to Miranda, to Luis, and to everyone behind the scenes who helped make this webinar happen. I hope you all enjoyed, and we'll see you in the next one. Bye bye.