Video: GitHub Copilot Billing Is Changing ——— APAC Partner Office Hours: Usage-Based Billing | Duration: 2384s | Summary: GitHub Copilot Billing Is Changing ——— APAC Partner Office Hours: Usage-Based Billing | Chapters: Welcome and Introduction (2.24s), Pricing Model Transition (90.37s), Enterprise Bonus Changes (268.675s), Transition Tools and Support (596.655s), User Budget Controls (723.695s), AI Credits Explained (877.205s), Pricing Impact Analysis (1019.77s), API Usage Reporting (1204.56s), Post-Promotion Strategy (1308.305s), Partner Resources (1479.575s), Q&A and Feedback (1649.095s), Dashboard & Billing Updates (1783.29s), Q&A and Closing (2039.005s), Closing and Thanks (2273.645s)
Transcript for "GitHub Copilot Billing Is Changing ——— APAC Partner Office Hours: Usage-Based Billing": three, two, one. Countdown. Let's go. Alright. Okay. Hello, everyone, and welcome. Thank you so much for joining us today. I'm Irina Zvigelski, and I lead partner scaling within the GitHub Learn team. I'm very excited to welcome everyone to today's sessions, and we're glad that you have that you are with us here for the webinar, which is going to be an open office hour. Our plan here is to answer your questions on usage based billing. But before we get started, just a few quick notes. Please feel free to submit your questions to the webinar chat anytime throughout the session. We will do our best to address as many questions as we can either during the discussion in the next hour or so or as follow-up after. This office hour is also being recorded, and we're going to post the recording in our partner portal usage based billing landing page. So with that, let's get started. Today, we have Evan Rodriguez, who is the director for go to market and business strategy with us to help answer your questions on our announced usage based billing launch. We also have, Geraldine Burns and Camilla available within our GitHub partner team to support as well. To start off with, we actually have several questions that were already presubmitted that we will start off with before addressing any questions that you may have throughout the chat. So with that, I'm gonna go ahead and turn it over to Evan to start with our presubmitted questions. Evan? Yeah. Absolutely. Thank you so much, Irina. So, hi, everybody. So I'm based in California. I lead our business strategy and GTM teams. Been working diligently on the usage based update to our Copilot pricing. Hopefully, we'll get plenty of questions, but got a couple of presubmitted ones. Overall, like, there was a lot of questions about the pricing specifically and justifications and how we are thinking about packaging and value and everything like that. The first one that popped out to me was really like and this was a theme across multiple questions, but the user referred to it as a cost per unit increase. What they said was for power users, the shift from 1,000 PRUs to $39 in credits represents a significant reduction in total monthly AI capacity. How does GitHub justify this value erosion for its most expensive tier? Think that's perfectly fair question. I would probably expand that question if I were and if I were one of y'all on this call. And I would say, like, honestly, that value erosion might be something that customers are asking about at any tier of Copilot, whether they're talking about individuals, Copilot business, or Copilot enterprise. To take us back like a year, GitHub had a very advantageous customer friendly PRU model. The PRU model was based on one request equaled one PRU. A user could ask it to do anything and it would go do that. There was a multiplier effect for certain higher priced models. But as we entered the agentic era in earnest like this year in 2026, what we saw is that one request was no longer mapping and we were no longer able to fix that to a fixed price per request. That makes intuitive sense if you think about the fact that a request might call upon the model multiple, if not dozens of times. It might have very large or very small context. Users might drop whole lists of things to do within one request. Obviously, with the rise of agentic usage, it's no longer feasible for GitHub from a financial perspective to say, we can price all requests the same. We have to have price discrimination across multiple use cases. So it is kind of like a must have. We are the last to the market. GitHub did hold on as long as possible in order to not make this move, in order to not move to like a more token first approach. But just the financial the economic realities of AI today mean that we had to move in this direction. I do wanna stop there real quick and I wanna see if anybody has any follow-up questions, whether it's in the chat arena, or anybody dropping into the q and a or chat, just on the overall why behind why we're making this move. Great. Got I will? pause. there. Keep going. Any other hot questions coming? Alright. Yes. No. So. I will jump into my I will jump into my next question. Next most common theme that we saw was like around the removal of one person referred to it as the removal of the enterprise bonus. In their words, in the previous model, enterprise users received a thousand PRUs, more than triple the business plans 300 PRUs. Why is this high volume bonus being eliminated in flavor with flat one to one credit ratio? First thing I would say is that in the original model, the more generous plan was designed to be as close to like a one to one credit ratio in itself. If you did the math, 4¢ times a thousand, you were getting to essentially $40 essentially, the $39 that a Copilot Enterprise plan was worth and the Copilot business plan was worth less than that. You're actually getting $12 in value for $19 of the Copilot business seat. I will say that just because the impacts of token based billing or usage based billing, were so wide reaching when we did an analysis across our business, we did want to just make sure that we were providing as much value as possible to our customers. Rather than just carrying over those dollar amounts of 12 and $40 we just said we're going to make it equal to the value that customers are putting into the system for their licenses so that those licenses effectively come back as zero. We We do understand there were a couple of questions about this as well, that this is gonna create a little bit of risk for Copilot Enterprise seats. Because in, many customers' perception, there's not gonna be a lot of like feature differentiation anymore around Copilot Enterprise. And especially if like the entitlement amounts are flat, our customer is gonna move back to Copilot Business. We do understand that that is a risk. From a business perspective, from a total spin perspective, we actually don't think that's going to be a significant risk to how much customers are actually spending with GitHub. Really, we want to start moving as close as we can to moving the seat costs out of the equation and really getting customers focused on the token based aspect of it. Because really, if we look at the overall token spend, that's what's going to be dictating what customers are spending. Because if I have a enterprise with two users that are using a $100 worth of tokens, if they pay for $219 seats and then get $38 in credits back or they pay for two $39 seats and get $78 in credits back, it doesn't matter. Both those seat costs net out to zero and really what matters is I'm spending a $100 in tokens to begin with. Mhmm. I'll stop there. Any questions that have popped up in the chat around anything I've said so far or off topic entirely? There is actually one that came in. I think it's very timely. When you are moving toward usage based billing, how does request multipliers matter now? I could see change in request multipliers in the latest docs. Yeah. Absolute. So in the latest docs so let me actually so this is one thing that I wanna share my screen for. And, hopefully, I don't break anything in sharing my screen because I wanna make sure that we're all talking about the same, item. If I go to the GitHub docs, and here, I'm just looking at this is called Models and Pricing for GitHub Copilot. Really important to note that everything in usage based billing or token based billing, some people might refer to it as, is based on model rates that are industry standard rates. So when you're asking how much do tokens cost, well, the short answer is it depends on how much each individual model and each type of token within those individual models cost. Our goal overall is to keep it as close as possible. Literally, we might not be able to move it same day, but whenever possible, we're going to try to keep it the same as the public API rates from the model providers directly. ChatG so if OpenAI charges something for a GPT model, we'd want to charge the same. If Anthropic charges something for a Cloud model, we want to charge the same thing. And I I say this because the important thing is that multipliers do not matter anymore for our Copilot business and Copilot enterprise customers. There's only one group for whom multipliers will matter as of June 1, and those are for our Copilot individual customers who were on prepaid annual plans already. So you can't serve yourself an annual plan anymore. We actually removed it from the overall pricing page because it was causing some confusion where users were just scrolling all the way down to the bottom. They saw updated model multipliers, and they were assuming that applied to everybody. It's only for this very specific group as model multipliers only apply to annual Copilot Pro and Pro Plus subscribers. You can click through to see those, but for Copilot business, for Copilot enterprise customers, those do not matter anymore. The only thing that matters is the price per million tokens per model per token type. Does that answer your, I think that was from Akshay. Does that answer your question? If you have any follow ups, feel free to drop it into the chat. Yes. I see that Byron Goodman, thank you very much. You asked a question that said, This feels like a very quick transition from PRUs to tokens. What will assist customers to make this transition while they learn the true cost of using tokens? Great question. Couple things that we have. So a, yes, it is a quick transition. We want to make sure that we are providing all customers with at least 30 of notice that this is a pricing change. It will take effect on June 1. The next thing so we notified all customer that's gonna happen. We're hoping we in our pricing announcement, we said that in early May, we are going to be releasing a, cost calculator for all customers. So a pricing estimate that's built into the UI, downloadable CSV, and a sidecar app that will help customers break down their costs using that CSV. We're hoping that that so we said early May. We're hoping that that's actually as early as tomorrow or actually Friday in The US. So it might be this weekend for those of you in the APAC region. But they will have access to it'll say, here was your PRU costs for April. Here are your token costs for April so that customers can see a side by side comparison. Owners and billing managers will have access to that page. And then oh, just to make sure. Did I stop sharing my screen? No. I did not. stop sharing my screen. Yeah. Let me stop that. So owners and billing managers will both have access to that preview of here's your PRU cost, here's your token cost. Then they will access to the app that will help them break down the CSV so they can see cost by model, cost by request type, etcetera, etcetera. Great. We have a follow-up here from Akshay, Akshay. Let's see. Most of the customer questions would be around token optimization. How can we do it? Is there any internal mechanism for the same any docs available, to help aid with this? Yeah. Absolutely. So first off, we so we are releasing a couple of, materials from our dev developer relations organization, both for individuals and for businesses and also from, like, our business well architected framework documentation. So we'll have a customer adoption, path that will help customers adopt token based billing and understand everything. Our well architected framework will have more opinionated articles on how they can do things like prioritize different models. Set budgets is a really crucial thing, though, and we'll have budget specific workshops available for customers. We're also seeing partners in region and globally, set up their own budget focused and optimization focused workshops from their customers because that is something that customers have been very loud and clear that they are looking for. In addition to with workshops for capabilities that already exist, important to realize that we are rolling out one new key ability for all customers that'll be GA on June 1, and that is user level budgets. User level budgets are really important because previously, with PRUs, all PRUs in the entitlement that if I got 300 PRUs as part of my Copilot business subscription, that means that I could only I could use those. With tokens, the $19 or $39 that we're including or $30 or $70 if you're counting the promo amounts that are good for June, July, and August, those are all pooled, which means that if I don't use mine, somebody else can use mine, so there's none wasted across an entire enterprise. Really good for efficiency, but it does make budgeting a little bit more complicated. User level budgets are now going to be enabled so that you can set user level budgets that dictate every individual user, not just cost centers and not just at the enterprise level, has a budget that they can hit even if they haven't yet hit the cost center or the enterprise level budget. Strongly encourage every customer to, activate that. That'll be you can activate that with a single click. There'll be a pop up banner that says that this is a new feature that's enabled and some suggested budgets for users on June 1. There will also be available availability to set those via the API if they want to, and they can customize those. They can have individual user level budgets set at different levels if they want to. K. Great. Looks like, at this point, we're all caught up with the questions submitted. If you wanna go back to the to the presubmitted ones. Yeah. Whoops. Where did my presubmit questions go? One good question I liked was, what is the rationale of GitHub introducing AI credits when providers like OpenAI and Anthropic are to use tokens as the standard user unit of measurement? I think this is a really good question. I think that AI units do add a layer of complexity to this overall thing. That's not designed to confuse our customers or to provide GitHub any pricing advantage or anything like that. It is simply for accounting reasons. Like, I just pulled up the list of all of the different model providers. We have to have a layer between all model providers that is you that is, unified across all. So whether I'm using a very expensive Claude Opus four seven model or if I'm using a much more lightweight GPT Nano model, I need to have one unit that's billable without having 40 new line items on a customer's bill. We decided to introduce AI credit. It also is for accounting reasons. We can't pre sell US dollars to customers if they want to prepay for something like this. We have to actually have a fundamental unit that we're pre selling. I myself and my recommendation to all of my sellers is to just simply refer to everything in dollars. There's a reason that we made AI credits worth 1p, so it was as natural as possible to do the conversion in your head without any extra steps to US dollars. So I would frankly just not refer to AI credits. I would refer to the total spend in dollars. That's how budgets work. Budgets are all set in dollars, not in AI credits. And our pricing on our web page is all also in dollars, not in AI credits. Let's see here. If we don't have any more questions, there was another question about billing calculator availability. As I mentioned, that will be coming out, hopefully, this Friday. It might get pushed to Monday. That'd be Friday US time. For most of the people in APAC, that'd be over the weekend that that'll come out. Another question I thought was interesting was, has GitHub conducted any analysis or forecast on Copilot adoption patterns, specifically identifying which customer segments are likely to increase their usage under the new billing model and which segments may reduce or limit usage? I think this is a good question. And it's important to understand, like, what is the scope that we're talking about when we're talking about the impact that customers can will see with these pricing changes. I wanna be really candid with everybody that this for most customers, this is not only just gonna be this is gonna be a price increase, but it's also gonna be, like, a very significant price increase. As I mentioned at in the very first question, we simply add, like, a entirely unsustainable model like GitHub. Literally, even with Microsoft backing, like, we literally couldn't afford the amount of AI throughput that our customers were loving putting through for each 4¢ PRU. So when we switch everything to tokens, costs are gonna go up. What we're seeing is that the average Copilot customer is going to see over a 100% increase to their bill. Now that's just based on, like, fundamental math, like the underlying tokens for the requests that they did in April. That's significant. Now one caveat to that is that if you just straight convert the underlying tokens to all requests that were made in April. If a company or a user has a budget set up, that would prohibit them from hitting a 100%, of additional spend. Obviously, that budget's gonna take precedent, and they're not gonna hit those larger numbers. They're just simply going to be able to do less. The other significant way that this is a price increase is there's no others there's no zero x models anymore, which means that there's no fallback models that when you hit your PRU limit, you can just continue to use unlimited ejectic requests. That being said, you can still use unlimited auto complete and you can still use unlimited next edit suggestion. Those two capabilities are still unlimited with Copilot plans. But in general, everybody should absorb that for all of our customers, it is going to be a significant price increase. That being said, as I mentioned, GitHub is trying to align itself as closely as possible to all industry standard models. So price increase doesn't mean that we are more expensive than the market. We're really just last on board to these significant like market wide changes. So we are gonna be on a very level playing field as far as every single individual model at token price. When you factor in the fact that our seats, if I look at certain competitors, they might charge for and pay for what you use model at some competitors, they might charge $12.20 dollars per seat and those seats include no usage. Then you get into model costs, whereas our seats always end up at net zero because the cost of the seat includes the same amount of model usage by default and then you pay for any additional overage. Great. No additional questions right now in the chat. Just to remind everyone really quickly, feel free to submit your questions as Evan, continues to answer the presubmitted questions. If we can't answer any of the questions, we're going to, send them over to some of our stakeholders internally, and we will follow-up after the office hour. But please do continue to submit your questions. You can get as many answered as possible. Absolutely. I think I have one or two more here. One was, is GitHub planning to introduce a high granularity API for Copilot usage reporting, maybe access to a raw data endpoint, especially input, output, cache, tokens to enable customers to analyze and optimize their usage under the new billing model? So short answer is yes. So like I said, we will be introducing, like a backwards looking CSV, start starting as soon as possible. Friday and Monday is when when we expect that to come out. And that'll be so that customers can compare not just, like, their usage when it starts on June 1, but, hey. We already got this bill for PRUs. What would that bill have been if it was billed in tokens? So we're gonna provide that. We're gonna provide that for May, customers are and then going forward, once the billing's actually live in the billing platform, they're gonna have access to all, like, the billing platform tools. That includes the API. Right now, we have an API that includes that a REST API to get billing usage information for PRUs. All of that PRU capability is automatically gonna be transferred over to tokens once we're actually billing for tokens. And I think I think I have one last question. So post this question is regarding our post promotion strategy. And just to be clear so everybody knows, we are running so as I alluded to, this is a fast transition. And, we are giving everybody, like, the thirty days of notice in or before we actually take effect on June 1. However, still very short, so we wanted to make the offer a little bit more generous at the start. So rather than starting with $39 worth of entitlement credits for enterprise users and $19 of entitlement credits for our business users, we're actually starting with 70 for our enterprise customers and 30 for our business customers. So we're actually getting more than what they're paying into the system. We think it's a really good deal. But this question is post promotion strategy. After the temporary $70 promotional credit expires in September 2026, will enterprise customers receive a permanent multiplier or discounted token rate to restore the previous 3.3 x throughput advantage? Again, I would categorize this as it's not that we're diminishing or not entirely that we're diminishing, the in enterprise advantage because as I mentioned, if you calculate that, like, the nominal dollar value was $40, now it's $39 What we're actually doing is we're just significantly bringing up the business value just so that everybody, we understand this is a significant change. We just wanted to help more of our customers. We realized that that's going to disincentivize some enterprise customers. But I will say that this is not going to be the last offering that we consider from GitHub. We did want to leave ourselves some room to get additional SKUs. There are things. We should have more information this summer about road map items. So by the time that we are out of the promotional period, you should hear more from GitHub about potential road map items. You can look at where the market's going, where our competitors are going, and they're going towards, usage only models where there are no seat costs. Everything's just built purely from a usage only model. They're going towards max and even ultra plans that are higher than our enterprise model. They're going for, like, upsell, like, quote, unquote power packs, so whether you can prepay for usage at a discounted rate. We're looking at all of those things in addition to looking at further tweaks to the Copilot enterprise plan, as far as both feature differentiation and included value. So I would say that, like, this is where we are now. This is the change that we had to make in order to get to a sustainable business model for Copilot business overall, but it won't be the last change and the last, like, value oriented offering that we're gonna come out with, for Copilot overall for our customers. Great. Great. Thank you. Please do continue to submit questions. While we wait for a few more questions to come in, just a quick plug. In case you haven't noticed or you haven't seen, there is a usage based billing landing page on our partner portal that is available right now. There are some really, really good resources that we already have posted on there. There is a full on partner guide with some additional deep dives on FAQs. We also have a partner offering toolkit, a conversation guide with some high impact customers. Earlier this week, we did post a customer workshop for understanding budgets. So that's new as of, I wanna say, two days ago. And then we also have some previous recordings for previous webinars that we've run. So continue, please, to check that page. We are gonna post any and all information we possibly can as quickly as we can just to make sure everyone is is overly prepared to have customer conversations. And thank you, Geraldine. She has a link here that directly feeds into that page. So we have a question that came in. Do you have any visuals or screenshots showing what it looks like after the changes? We'd like to use visuals or diagrams to help explain this to come to customers. For example, the usage budget or the user budget layers. So that is a good request. Yes. You can hear me right, Yeah. Irina. I'm not on mute. Okay. I was briefly muted. So that is a good request. We are working on those both for internal sales teams, because user level budgets is a brand new feature. It's literally just like in early in early private preview right now, so it's not even, like, widely available anywhere. But as we get closer to GA, we expect to, one, we're gonna be updating some dot we're gonna be making some unique moves in terms of updating public documentation with pre GA amount for everybody. Usually, you only see documentation for a new feature that ships if you're actually in the private preview. Because so many customers are highly interested in user level budgets, we are going to make some tweaks to public facing documentation with the big asterisk that says this isn't for GA product that ships on June 1, just so that customers can prepare because we've seen plenty of questions around, hey, what are the actual, like, API fields that I rely on today that for budget setting that might change with things like AI credits or when we're shifting from PRUs or for user level budgets. So we are working on additional materials. And once once we have those for GitHub internal sellers, we can also, put those onto the partner portal. Yeah. Exactly. But a good point. Our our goal is to get information over to our partner teams as quickly as Hubbers get them. So we have been. doing well so far. We'll continue to drive that, at the fast speed there. There's I don't see any current questions, so we'll do one last call for any questions or even feedback. We'll take any feedback that you may have or anything that you've heard, from customers that you want help kinda how to navigate. This is a great or an opportunity to to talk about that as. Absolutely. I'd be real I'd be really curious about anything. that customers are saying to you, questions. I heard we heard about the optimizations. We heard about like more visual guidance are both really valuable pieces of feedback for us. But, also, just like what are objections that what are your customers saying to you? What are objections? What are the mean and not always nice things that our customers are saying to us around a price change like this? Happy to help navigate those as well because we've heard quite a bit in the last couple days, as our internal teams have also been navigating these conversations. And we know it's very active internally at GitHub as well. Yep. We have a question. Do we get token usage details in the usage metrics API? So yeah. So I will set let me actually find this. So in the billing usage API, once we actually start billing via tokens, we will be providing it will actually be in, like, AI credit. That is one place where you will be getting information in terms of AI credits. So you'll be able to see, like, who was the requester or what org requested made the request, what was the token, what was the volume, and what was the AIC, and when when did it happen. So you'll be able to get all the same metrics essentially that you're able to get for PR use today. And but they will also be, like, model specific around tokens once this goes live on June 1. Great. Does that in does does that answer your question, Akshay? Amazing. Thank you. Yes. Alright. So last call for questions, feedback. In case you are also having, challenges or struggling with some of the conversation, please do check out the conversation that that we've made available in the partner portal as well. That's been a good resource. We've gotten some really good feedback on that recently, so just wanted to call that out as well. Actually, he has a question on the dashboard. Let's see what he's asking. Was there a question on the dashboard, or should I just start talking about dashboards? We we can do either. I don't know if he's trying oh, here we go. So I I think the question is around, like, at what point will organizations get access to their token usage in dashboards? I will say that in general, when I'm talking about dashboards, I use I'm generally referring to our or I would typically be referring to, I should say, our budget dashboards and, like, our actual, like, billing dashboards in the product. Important to note that, like, our standard billing dashboards today in on May 6 for the rest of this month, we're still billing via PRUs. So none of that's gonna switch over from PRUs to tokens until June 1. What we are working to make available is, like, what we're referring to as a sidecar app. So the sidecar app will basically do three things for customers that will, one, and this is available to all organization owners and billing managers, within enterprises, that we manage. So first, it will give, like, a preview, like, dollar value. As in in April, you spent this much total, under our PRU model. And in April, we think that amount would have been would have converted to this amount in our token based or a usage based model. And really importantly, like, that's not like a standard conversion. One customer could have spent a $100 in our PRU model and spend a $105 in our token based model. The next customer could spend $100 in our PRU model and spend $250 in our token based model. Because really it's about the complexity of the usage that was underlying those PRUs and that varies dramatically and really that's why we have to make all these changes. So first, they'll get that preview. It's just two numbers side by side. It'll pop up for everybody, for every organization. I'm sorry, enterprise owner and billing manager once that's available either Friday or Monday, is what we're hoping for today, once that's available. They'll also be able to click through to download their CSV. The CSV will have row will have row based information on every single request and the AI credits that that request consumed, what models that actually hit, etcetera, what user or what organization actually requested that, etcetera. So all the same information that you would already be able to download a CSV for for PRUs, but for tokens and, like, the actual AICs and how many of those were spent on each of those requests. And then the third thing that'll be offered is like a temporary what we're referring to as a sidecar app that's just designed to help customers easily parse the CSV. We know many, many of our customers will go and start hacking at the CSV themselves and build custom analysis, etcetera, Many other ones, especially like billing owners or procurement managers, they won't want to do that. They'll just want to be able to easily slice and dice it by model or by user and see what their highest usage is. And that app is designed to do that. And all three of those things should be coming, hopefully, Friday, Monday is what we're looking at today as a ship date. Okay. Great. A question came in around the cost center concept. Does that still apply to the new model? Yes. So cost centers are cost centers are still applied to the new model. So all the same capabilities that you got there. There is one significant caveat around cost centers. Cost centers themselves haven't changed, the model itself has changed, and the significant thing about the model that's changed is that entitlements are no longer per user. There's a pooled entitlement. Now cost centers before, they always only controlled, quote, unquote, overage above entitlement. They are still only going to control overage above entitlement. So there is going to be some concern about users to make sure that, like, different cost centers are not using too much of the entitlement that maybe other cost centers are paying for. Because you can imagine if one cost center is really eager to use a lot of, tokens and a lot of agentic use cases early in the month, they could be the first people to use that entitlement. And another cost center might not get as much, and they might go into overages sooner, or have a higher percentage of their overall usage in overage. So cost centers are still only covering overages. However, our new budgets, the new capabilities that we're rolling out are gonna cover both overage and entitlement. So if I'm a user and I'm subject to a user level budget of $50 for the month, I can only spend $50 overall no matter whether that is in entitlement or in overage, it doesn't matter. I'm going to be limited to $50 overall. So we encourage all customers to implement user level budgets with one click or to do a little bit of customization on June 1. Great. Great. Well, I don't see any other questions right now. Camilo put a great note in there. Even after this office hour, as we head into June 1 and even after June 1, you can always reach us through the partner portal. There's a link right there that Camilo added. We're happy to answer any questions, again, feedback. We know more are gonna be coming. We're fully prepared to do that. Please feel free to reach out at any point. If we don't have any further questions, just a quick reminder that we did oh, one came in. Starting on June 1, are there any settings or configuration changes we need to do in the GHCP or GHEC instance that it's connected to, or they're gonna continue to function like before. Yeah. Abs absolutely. So I would say the only thing that I would strongly encourage people to do is that's that's all new, is go implement a user level budget. Even if it's even if it's arbitrarily high, go implement a user level budget on June 1. Even if it's a even if a customer says, I wanna be I don't wanna limit my users. I just wanna have cost center budgets. Those same customers will be mad when one user blows through a thousand dollars worth of credits in their first couple days. So go even if it's high, implement a user level budget. Everything else, there's no additional configuration necessary. Like, if you already have cost center budgets set up for overage, those will carry over. If you already have enterprise wide budgets, those will carry over, because they were already set in dollars even though it was PRUs, and they're still gonna be set in dollars even though we're talking about AI credits now. So no changes there. Like, the other significant thing is just that people are gonna burn through these a little bit faster. So we do expect that, like, in the June, some people that are not used to hitting their budgets that early will start hitting their budgets and they're gonna have to reevaluate what their spend thresholds are. Good question. though. Definitely. Alright. Well, I'll give it thirty more seconds for any questions to come in, and then we can close out our office hour a little bit early. Really appreciate all the questions though. Really good stuff. Mhmm. These are very much things that we've heard from our internal sellers and from. our direct customer conversations that we've had so far, so we'll be really meaningful. Yep. Definitely. Definitely. Alright. Well, I don't see anyone coming in. So I think we're gonna close it out a little bit early. Thank you all for the for the participation. Really appreciate your time. Hopefully, you got some good guidance and, greater confidence on how to discuss usage based billing with your customers. As a reminder, we welcome your feedback. Please continue to reach out to us via the partner portal. And last quick reminder, this office hour is recorded. We're going to go ahead and post a recording on our partner portal as well as another resource to go back to. We have an office hour that we're also, having tomorrow, for our AMER and EMEA partners. That's also gonna get recorded, posted in the partner portal. So if at any point in time you're curious what other questions were asked, you can go back and listen to that recording as well, for additional guidance. And thank you, Evan, for sharing the the docs page. Perfect. Alright. Thank you, everyone. Thank you, Have a. good rest of your, day. Irina. Yes. Thank you so much, Evan. Appreciate it. Thank you, Ginny, and thank you, Camilo.