# FinOps Foundation FOCUS V1.4 | The Keys to AWS Optimization | S17 E5

## Метаданные

- **Канал:** AWS Events
- **YouTube:** https://www.youtube.com/watch?v=noQ_wW-_ozk
- **Дата:** 22.05.2026
- **Длительность:** 58:43
- **Просмотры:** 182
- **Источник:** https://ekstraktznaniy.ru/video/51701

## Описание

Join us for an essential deep dive into FinOps Foundation FOCUS V1.4 

In this livestream episode of "The Keys to AWS Optimization," we explore the latest updates to the FOCUS specification and what they mean for your cloud financial management. Discover how FOCUS V1.4's new features—including invoice reconciliation, contract commitments, correction handling, and enhanced commitment programs—can transform your AWS cost visibility and control.

Get your FinOps X ticket $699 using KEYSTOAWS

What You'll Learn:
✅ Understanding FOCUS and its role in multi-cloud cost management
✅ Setting up FOCUS in AWS Data Exports
✅ New features in FOCUS V1.4 (June release)
✅ AWS-specific implementation changes and commitments handling
✅ Hands-on demos: FOCUS Sandbox and Validator tools
✅ Real-world use cases and deployment strategies
✅ Why standardized cost taxonomy matters for your organization

Whether you're a FinOps practitioner, cloud architect, or finance professional managing AWS spend, this sessi

## Транскрипт

### Segment 1 (00:00 - 05:00) []

Hey. Hey. Hello everyone and welcome back to the keys to AWS optimization. The show we share stories, concepts, and solutions to help you unlock your spend at AWS. Today we are going to be diving into the new release of Focus from the Phob Foundation. That's why I've got Matt with me. I'm also gonna be talking about bit of sex uh the conference in oh my goodness how many how far away are we from the conference Matt — we are 18 days away — is that oh I need to practice just talking about my session beforehand uh but don't worry folks if you're coming I'll be well we'll rehearse before then um but yes h so throughout the session please put your comments questions in the chat or if you're watching us on demand feel free to put a comment below and I will catch up with you later and you can always get in contact through uh cost optimization Amazon. com for any questions you guys have. Uh got some hellos uh Hayes in the chat. Welcome, welcome everyone back to the show. Um I'm so excited to get into today's session mainly because so focus well actually let's do some intros and then I'll give you some my brief history of working with focus. But Matt, let's um uh give us your one minute origin story for me. — All right. Uh so I ended up uh falling into FinOps after getting a degree in accounting and then uh I joined the US Army for seven years where I was a budget analyst for the latter half of my time and kind of fell in love with getting the numbers to add up and then uh went and got my NBA at NYU and then got hired by AWS to go uh work on a product called budgets. Uh loved it. uh also was there when we launched savings plans and launched a little uh service called billing conductor which is very partner oriented. Um and I just I've always loved FOPS. I love the community as well like it's a very passionate group. Uh and yeah, now I'm at the foundation and I'm the principal product manager for focus which we're going to talk about a bit more later. Uh, but it's really the utility you get out of having like a normalized uh view of the world that I think is really the thing that gets me most excited. But yeah, I you know, I'm living and breathing it. I didn't realize that I was doing it back in the day. Uh but yeah, when I look, you know, history or uh hindsight's 2020, uh yeah, the stars all have been aligning for some time and I'm just now realizing it. — Yeah. And I mean focus, if you don't know what focus is, do not worry. We're going to dive into that. We're going to have a look at what some of the format is, why we should be using it, and things like that. But it's funny, um, with focus, I always describe it as I don't know if people remember this about me, but I like don't know much about containers. Containers is like one of my big like spots I don't know anything about. And focus sometimes feels that way because focus didn't exist when I was a customer. And so customer, I made my own like version of things like a lot of people had to do. And so I never had to learn it from scratch. And so I know I'm excited to say to like ask maybe some dumb questions which will uh inspire me. — No dumb questions. — Um just as in the chat I love when people find these little emojis. If you're in Twitch, you can see this little emoji of me lifting a dumbbell from years ago. Um uh it says, "I'm not sure what focus is. " Don't worry because we will tell you. Um and someone's asking if we're going to go to Phops X. Yes, I am. I'm gonna be at Phopex. I know Matt will definitely be at Phopex. Matt, do you wanna Before we go into focus, we're gonna talk a little about Phobex. It's a big phrase that we love, which is why we have these foxes uh around our — Can you tell us what the color is this year? Are you allowed to share that secret? — Uh no, I don't think I would make it to PhobX if uh if I told anybody what the color of the fox is. Uh but I'm also going to be uh completely honest with you. I don't know. Uh there's not a whole lot of surprises in life and so this is one of the ones for me. So I'm going to do uh I'm going to be a surprise just like everybody else. — Yeah. I feel like let me know in the chat like what color you think it is cuz like we had purple. — Yeah. I mean you could go for something random. You could do like bright red. But I feel like maybe um another Oh, someone's put blue. Yeah. I feel like blue. I feel like white. Um so you never know. have to we'll let you know when I get my new fox. Um, — but Matt, why don't you give people a bit of context of what FHOPS X actually is? — Yeah. So, FHOPSX is the annual

### Segment 2 (05:00 - 10:00) [5:00]

conference that the found the Phops Foundation puts on. Uh, again, it's meant as a opportunity for the community to gather bare best practices. You're going to hear uh what's top of mind from practitioners. You're also like the POPS tooling vendors that you see in the space. Uh obviously the folks that are actually uh generating the data. So you'll see a good mix of um you're going to organizations from public cloud and from SAS and from data center and data clouds. Like it's all of the technology providers that you as a practitioner would use on a day in dayout basis. like they're going to be at the conference and the people that have your jobs, they're also going to be at the conference. And so, uh, you're going to have an opportunity not only to learn from, you know, the best and the brightest. I think the — the, uh, the ambassadors who actually curated and selected the talks that are going to be available, uh, this year did a phenomenal job. one but two like it's the whole point is to like talk about what's currently happening within the industry and so uh I always come away like more refreshed after that which is kind of a surprising thing because it's a lot of time and effort that we put into it but I feel more refreshed after that the conference than I do before and really — um yeah it's just — I don't know like I get a lot of energy by being around other people that like and do the same things that I like to do. And so, uh, yeah, it's a it's going to be a phenomenal time. — Uh, I couldn't be more excited for it. Last year for me, I just started at the foundation in April. So, — uh, it was it was, uh, interesting to try to get up to speed before that, but now, uh, yeah, I'm just, you know, so much enthusiasm. — I, um, I see some people in the chat are going to upex. Let me know in the chat if you're coming, uh, and say hi. Uh, but yeah, I know what you mean. the I think is does Jr. say like returning home is the is what he says something like — yeah welcome home — welcome home that's it um and I know what you mean it's funny because there are people who go to an up sex and I see them like once a year and I've seen them once a year for like seven years or like twice a year and it's just these people and I'm just like how funny that I like you spend like these days together it's like an intense week it's not as big as reinvent so it's quite nice because that like reinvent is intense whereas this feels like you said a lot more like energizing and you get really motivated really good questions and stuff like that. Um, — it almost feels like a reunion of sorts. Like there there's this like such a strong community orientation. Um, yeah. Like — yeah, — there's just like there's a comfort level like Yeah. Anyway, it's uh I don't think I've met a single person that that's gone to FinOps X and didn't feel like they got a just so much utility and appreciation out of their job. uh as well as out of the community that they were able to meet while they were there. — Yeah. And my favorite thing, swag. — Yeah. Lots of swag. — Um but if you're still like, should I do I want to go? Well, we got a discount code for you. If you use the code keys to AWS, you will get a your ticket for $6. 99. Uh wait, is that Yeah, 6 was like not 6. 99. 699. Um, just to clarify, uh, when I say it sounds like a dot, but yeah, use that discount code and, uh, you can get a nice little chunky discount for that and come say hi if you watch the show. Uh, let me know and um, and we can kind of keep that going. Uh, so, uh, we've also, sorry, just want to pull this up. Um, I've heard that the fox can jump over a lazy brown dog if like very clever. Um, but let's I wanted to drop a couple of things in because we were talking before about like what to look forward to with X. Um, and so two things. One is I have a session AI unsurprisingly um but optimization and it's me and Adam. We're talking to Capital One, Brent from Capital One. So if you're looking for an AWS's point of view, an ads specific uh session at Phobex, then please check this out. Also, if you want to come hang out with the AWS crew, then we have a mixer happening. Um, so you have to sign up. This is for customers only. Um, feel free to just like sign up and then uh it should be a nice little lection on the Monday. Yeah, on the first day. So, uh um I was joking because I was like I'm techically on the get on the speaker list of the party. I was like, do I have to sign up? And then they were like, yes, I still have to sign up. Okay, fair enough. But what um we were also talking about sessions that we're looking forward to and we just had a quick look and I was like what jumps out and the first one we saw was this one right Matt which was like running Phips at scale with a small team that sounds really cool. Yeah, I mean I always find it interesting, especially uh with, you know, maybe this

### Segment 3 (10:00 - 15:00) [10:00]

is a slightly different flavor on the AI for Pinpops versus PinOps for AI, but I have a sneaking suspicion when you're running uh with such a small crew that like you're looking for any bit of efficiency and scale that you can get out of uh out of your practice. And so, yeah, I mean I always I mean, obviously there's going to be 15 to 20 sessions about PinOps for AI or AI for Pinpops. Uh, I'm also guilty. I have one of those sessions. Um, which again I hope folks want to come to at 5:00 PM on a Wednesday, but I promise we'll make it worth it. It's me and Mike Fuller and uh Bo Nelford, who's our new uh staff technical community architect. But um but yeah, there's just uh so many like very positive uh interesting sessions that I think folks will walk away from with a ton of new insight. Yeah, for sure. There's going to be I think you're right. There's AI is like the hot topic, but the reason why we were talking about this, I was like, let's find something that's not. I'm like I'm definitely going to be looking up positions that have like a little bit of different but like you said, we both have sessions. But — I I do think too uh just for a bit more orientation around the types of talks that are there. Uh the chalk talks are the ones that are not recorded. Uh so you get a slightly different flavor. uh for presentation there's a bit more uh interaction and orientation within the talk um whereas like the breakouts uh and again I'm not saying don't go to the breakouts please do um but if you have to make very hard decisions like some of them are recorded and not so you know use that to your advantage when you're trying to optimize your uh your time at phopics — yeah if you want to get that it's interesting the balance between the two of like if I find as well if you've like experienced something and you kind of want to see how other people approached it and you're like in the mix of doing something, a chalk talk is really good because then you can like bounce ideas off people and have a discussion versus if you have like no idea about a subject sometimes it's better to go to the breakout so you can like level up a little bit. So that's how I try and think of those sessions because — in the talk talks like you're thrown in to discuss and if you've never if you're like never done it before you're going to be like what and so that can be quite like but it can also be really educational every session you go to I'm sure um — yeah no I like that framing though like that uh that resonates for me on the ones that I would be picking on my own so that yeah that tracks — yeah if I haven't done something like I'm curious to see like how people approach things then I might go to more of the breakout sessions or if like new products and releases and things like that. Um but you will be a lot at the focus booth, right? — Yeah. So, uh for folks that have been at X before, the focus booth normally sat outside of the expo hall. Uh so it was very hard to miss. Uh this year we get the pleasure of being within the expo hall, which uh I'm interested. I'll maybe have some feedback after to see if it was a positive or a negative experience with that particular shift. But uh yeah, so we're going to be in the back in the expo hall. Uh I don't want to like say that we're like the maybe I don't make this reference. There's this joke about like the Costco chickens the they put them in the back of the store so you go and uh you go and get them. And I'm wondering is that what they're doing with focus? Like trying to get folks in. — But um no, it'll be a whole lot of fun. Ton of interaction. and the folks that actually help build the specification will be there as well. Uh we have uh data generators that have a focus implementation giving demos there. I'll be there pretty much the whole time outside of the fact of when I'm uh giving talks or kind of outside the expo halls closed. Um yeah, it's going to be a whole lot of fun though. So, if you're focus curious or you're focus interested, uh, I'd strongly encourage you to swing by, say hi, uh, spend, you know, 15 20 minutes kind of getting an orientation and happy to answer any nuanced questions you might have about focus. — Yeah, it's one of those things where like you'll learn a lot from today's session if you haven't used Focus before, but getting there and like having that kind of hands-on and able to ask questions. Uh, but if you have any questions, of course, today, feel free to put them in the chat. But let's transition uh a little bit over to focus elements. So yeah, Matt, why don't you just like what is focus? Let's give that brief summary. — Yeah. So focus uh it's it stands for PHOPS open cost and usage specification. And so the project was founded back in 2023. And the entire premise was like how do I normalize my public cloud cost? uh because this is when um practitioners started to want to operate across multiple clouds. Uh and so there's challenges there is like even at a categorical level like each provider like has a different

### Segment 4 (15:00 - 20:00) [15:00]

interpretation a different way that they describe and again they have different offerings which is all natural and is expected. Uh but as a practitioner you're like you're trying to communicate to your leadership team or you're trying to make optimization decisions and it's hard to do that when you need to like get everything into one format so you can at least evaluate it or communicate or talk about it. Uh and so focus like focus is the specification that we launched back in 2023 and we've been on a two release cycle uh every year. So every six months we introduce a new version of focus. Uh which is has been fun. But the the part of that I maybe want folks to come away with is like focus is a language for how we communicate technology value. Yes, there's a specification where we're uh you know we spent a lot of time and effort around expectations for how the data quality should look like in the data format. again with the idea of making it easier for uh the folks that are consuming the data to be able to actually understand what it's doing and then ideally being able to, you know, union those across the providers that they're using so they can get that unified view of their technology spent. — Yeah. And not having to union it yourself. Like I think that is in like a big nutshell it I it it's like almost what I said at the start but like doing that yourself back in the day it didn't really work and that was where the call for this kind of concept came from and it was so needed especially if you are multiloud and I think that uh we're never going to we it's great that we've all like agreed as providers to do this uh but having our own stars as well so we still have and the other two have whatever their ones are. Um I can't remember off of my head, but it's good. — Well, I would I would say too that uh I and just maybe to preempt some questions that I hear regularly, um — it's not a it's not an or discussion. It's generally an and discussion. So there are very specific details that each individual provider has within their native format that's critical for optimization and for evaluation of usage within that individual provider. But — uh when we're talking about reporting and communicating with stakeholders, you really don't want them to need to learn the nuance language of every single provider to understand the utility that a business is going to be getting out of that technology provider. And so that's where I think focus is really helpful from a if you're operating across more than two or three providers like having one language to communicate that makes it significantly easier. Um you get utility within the PHOPS practice as well as communicating the financial decisions that you're making. — Yeah. Like if you've been hired in to do PHOPS at a company and you come from somewhere that maybe did majority one cloud and now you're doing one or more of a different one. Having that focus at least you're like okay I understand what this means. means that with these things flags um they can have that. Uh we've got a couple of questions and points. Um so uh someone said why twice a year I think when you mention the releases. — Yeah. Uh which I get can be a bit scary on the forefront around like why are you making adjustments to my billing data twice a year. Uh but what we find is that providers have their own kind of release cadence on when they're ready to develop the next version of focus. And what we're interested in doing on this twice a year cadence is we're listening to practitioners every single day tell us like hey I need this additional capability. I need this data formatted in this way. uh can you normalize this thing that currently is sitting only within one provider or I have a new technology provider that doesn't support focus yet but if you had this capability then they would uh and so for us when we're talking about this release cadence it's really working backwards from the needs of the practitioner it's one of the things that I think makes it kind of fundamentally different from focus relative to like a native export is my entire goal and purpose in life is to evolve focus to meet the existing needs of the PHOPS practitioner. So it doesn't mean that like focus 1. 0 wasn't the right thing to do at that particular point in time and if 10 satisfies your needs like you may not need to go to the next version. Uh but what we're trying to do is integrate the use cases and capabilities that we're hearing are necessary now as the roles and the responsibilities of the Phops practitioners grow over time and they're just fundamentally different. They change pretty quickly and we're trying to make sure that we keep up with that pace of change.

### Segment 5 (20:00 - 25:00) [20:00]

— Yeah. And what's the process like for someone to change the uh focus version? How does that work different for provider? or is it like an update feature like what does that look like? — Yeah, so generally and so this is something that generally sits within the provider uh on how they deal with their report configuration. So I don't want to speak too broadly but generally what we're asking uh and again you'll hear me use the term provider and data generator uh colloquially it's the same thing just one sounds it's easier to say. Um, but when we talk to the folks that are generating focus billing data, generally we're asking them to specify the version, provide documentation for the version that they support. And so it's a natural progression for them to put that into their reporting workflows, whether it's uh you know something as specific as like an object format where we're talking about is it a CSV or is it a parquet file or even if they have their billing API uh and that's how they're vending focus data uh there's generally an optional parameter where they're specifying the version that's being vended. — Excellent. So make sure you check what version you are working with on your provider and what you have set up folks. Um and someone's kind of asked is this a universal translation to just to summarize in like this thing it normalizes that's a better way of describing it right it normalizes the data across cloud providers so that you can understand it all in one element. Um I also like this question. Um when pred when practitioners are running multi cloud providers uh what does that cost data situation usually look like before focus enters the picture? So what's it Yeah. And what's the kind of like I mean you've kind of described the value but what does that transition look like? What does it normally look like beforehand? Yeah. — Yeah. Uh so my recommendation uh is folks kind of make this a pull in an evaluation between like am I ready to adopt focus? Does this make sense for my organization? There's generally a core set of columns or metrics that you're interested in tracking uh within the native data set. And so the first thing that I would do is say are those also present in focus? And if they are that's great that the next step is then am I getting the same values by using the same calculations that I was using beforehand. And that does take a little bit of time. And so I generally recommend folks run it for a month or two or another pattern we say is like turn it on now so when you are ready to get into it uh you'll have more historical data to work with. There's a couple of different patterns there. But uh I'm always a trust but verify. So I you know I'm glad that Matt's working on this thing but I want to see the documentation and then I want to run the files check it against what I've done historically. Like I think those are all super solid practices uh and things that I would encourage uh practitioners to do before they do a wholesale swap over. — Yeah, definitely. Val, yeah, like that trust and validate that combination. Um, excellent. So, uh, do you want to show folks where they can see this data and information on the Finance Foundation website? — Yeah, I can go to a couple of different places. So, um, — bring up — here. I'm gonna I'll maybe start with uh I'll go to the very beginning of the Am I sharing my screen? Can you guys see it? — Yeah, I see. Okay. Sorry. Uh so this is the the focus website where you'll see a list of all of the existing providers that support Focus today. Uh so we have a healthy list. It's growing uh seems like by the week sometimes. So you know we talked a lot about the public cloud story but uh you're also seeing other providers that are not just public cloud. So the nebases and the verscels of the world we uh graphfana just launched GA not too long ago along with Reddus. Uh you're going to continue to see this list grow. We have some exciting announcements at FinOps X. Uh so keep your eyes peeled there. But uh back to like hey where do I actually learn more about this? If you go into adoption, we have this list of data generators and you'll see this list broken out on what version they support, if they have uh a gap report or not, which again just says the difference between the spec says this, but my implementation may be slightly different. You'll see that continue to grow over the course of time. And I'll go into to AWS specifically. uh you'll see a breakdown here of this is the exact steps uh to be able to uh create a focus export and so you can do this in the console with the cloudformation template I think you

### Segment 6 (25:00 - 30:00) [25:00]

can even do this with the C uh CLI though I haven't had a chance to do that one personally just yet uh and then we also link to both the announcements the versions uh and the documentation so um this is where you can go to check to see hey how does this actually represented within um the providers that I'm interested in working with. — Yeah. And why like so cloud obviously I'm a cloud provider um I'm not you know what I mean I come from the cloud provider point of view. Um — what why are the like what do the other ones entail? the other ones that aren't necessary like the other providers like that aren't cloud providers that are rather kind of data generators as you put it. — Yeah. So I mentioned before when we kind of started the project was very much focused on uh public cloud orientation of normalizing billing data but what we're observing over time is that uh now practitioners are responsible for managing SAS spend and data cloud spend and you know AI spend is you know obviously a very um a very relevant topic right now as well and so you know it's not just hey which like how many different public clouds am I on but what they're and I'll just kind of share colloially or like anecdotally what I'm hearing from these data generators that come out to me they say hey — we either don't have an external format yet to uh vend our billing data and it sounds like you guys spent a lot of time and effort — uh trying to try to think about um you know what this like what this format should look like why would we go introduce our own thing like this seems to be catching on and has support so why don't we just start with focus and we've had a number of providers that have done that — um Nebius is probably my favorite example uh and I'll maybe go back to the screen for a second uh so Nebius uh which is uh a Neo cloud uh so they heavily work with like um AI and GPU specific workloads Um when I was talking to Nebius about supporting focus like oh you need to go through this like native column mapping to focus column mapping to kind of so your customers understand what's going on. — They're like Matt we don't have a native column mapping and I was like oh my god this is incredible like we're making progress. Um, but it was such a cool experience um to be able to — uh to hear that. And again, I think it's going to be more and more common like uh you know, I can point to the other uh to some of the other providers like uh Verscell has a billing API. It's only in focus format. Reddit's uh Reddit's the same. Uh Graphfana just released their GA of focus 1. 2. Again, I'm not aware of another um another format that they have. And I think we're just going to see this more and more common to whereas like today it's novel. Uh you know, two years from now, I think it's going to be weird not to be supporting focus whether it's the only uh format or not. But we're definitely seeing — um we're definitely seeing additional adoption. And again, like the more coverage we have from the provider base, the stronger utility the practitioner community is going to get. Yeah, 100%. And it's the like it's like I said, bringing awareness, but also the scope, no pun intended, to the Philips scopes, but the scope of a PHOPS role has expanded. And this was a J talk about this at London Finex Day and I'm sure we'll talk about this in a couple of weeks. But this role has gone out because it's shown how important it is to have a centralized point of this data and to normalize it and be able to do these reports and things like that, but also just like to simplify it. If we think about all these previous versions of all these different things flying everywhere now just to centralize it and ask questions and then you add AI on top like being able to ask questions in like normal language to understand these kind of things has become even simpler once you have all this data because data drives decision you can see how much you're using something whether you need use more less and all these kind of elements which is very cool um I will what I'll also do just because you mentioned like getting started — um just uh in the as console This is what like it looks like in here. So if you go into your billing and data exports, you come down here and this is where you could choose one of your reports. The one you want to choose is the focus one. And so you can see currently we have focus one and one. 2. So whatever is relevant for you and you can go down into like seeing the different details park everything is normally like set to default like keep it default I would say. And you can see all these different columns things like that. So some of this is like familiar

### Segment 7 (30:00 - 35:00) [30:00]

to what we see but it's like slightly different. It's like par parallel universe occur. It's like slightly different naming and things like that. When people um get started with focus, do you see them kind of like struggle with anything or find anything really easy between the two kind of like sets of data? Especially on the AWS side. I I hear a lot um and I'm thinking back to the reinvent talk that uh Justin Markx did with Jason Woo back uh back in November I guess maybe it was December of last year uh and they like very brave of them like they pulled up the Athena uh console and were just hammering away on two different files at the same time and one was uh one was the curr and one was the focus uh the focus export report and they shared this query for uh like trying to understand your RARI are your RIS uh like hey how well uh was something being utilized and uh like effectively or we call it in focus uh like what's the effective cost of a particular resource that's covered by a reservation and uh when they were showing the SQL query on how to get to that value uh with the cur was I think it was like 30 or 40 lines. It was it was a whole host of like nested if statements and case and everything else — and then they pulled up the focus export and I think it was two lines uh that they ran and it got the exact same value. — Yeah. And so like that like built cost and effective cost and kind of the relationship between those two particular cost metrics I think are probably the thing that folks immediately gravitate towards. Um and again like we're effectively asking the providers to do this computation on behalf of the practitioners at the point of computation rather than needing to do that analysis after the fact. I think that's always I think it's probably been one of the most like durable positive benefits that I hear from folks that adopt focus or kind of moving from a from another format. But — yeah well because as you're saying that it's kind like we now there was two different reasons behind having these data sets. when I'm sure Amazon created well the Kr was after what was that report before the Kur — the D D D D D BR D DBR — detailed billing reports — and like so there's like the these had just like spends data spending users data really granular like millions of lines long and it was just kind of like spinning it out so you had access to everything versus focus is for a job like there's a reason why you have these things and it's the experience over time like you said to be able to be like okay what we actually need is this we don't want to do all of this work beforehand. It simplifies this kind of thing which is really clever. Got a couple of questions. Um — is focus a free tool? Now I can go with this one first if you want. — Uh yeah, go ahead. — I mean the concept is free but in the cloud in ads specifically you still pay for data storage. Like that's probably the simplest version of this is that whenever you deploy a focus data export that concept is free but you are putting it into an S3 bucket and so therefore you are incurring storage costs but anything you want to add? — Yeah. — Uh I want to maybe touch on the uh we distribute focus as a specification in a royaltyfree manner. when I was mentioning uh or maybe we were talking just before the start, we're in this uh intellectual property review phase and I've learned all this new terminology for how open source and open specification development happens. Uh so bear with me for a second, but um yeah, so focus as a specification is freely available. Um, everything that we do around the specification for like literally every single PR that I've ever uh added to the focus spec repo is publicly there. Everybody else's contributions as well, but every single contributor signs a it's called a contributor license agreement and that's how you actually contribute to the spec, but you give your efforts, your contributions in a royalty-free manner. — Uh, which allows us to be able to distribute as widely as possible. I mean it doesn't change what you just described but I wanted to make it clear like — uh the project in and of itself exist uh it's like a very strong community orientation there's no — there's no uh yeah there's no cost associated there. Yeah. — Yeah. So if businesses want to implement it like if there was another newer cloud provider they could get the spec and then utilize it. There's no like cost to them. Yeah. That's — Yeah. No distribution freeze nothing.

### Segment 8 (35:00 - 40:00) [35:00]

Uh it's just like we think this is the best thing and I don't mean like we the foundation I mean like the group of 56 contributors that are actively in the GitHub going to the meetings again like bringing forth the use cases and capabilities that they need or the challenges that they're trying to solve for on their day-to-day basis. Like that's how we drive this consensus orientation. — Yeah. that community and that's always been a big thing I think about the Phops world is that we're one of the most like open and sharing kind of groups. Um, but it makes sense by nature. Like security, you feel like they're trying to keep things private and secure to the main. — Yeah. But Finnops were like, "Hey, we're spending money. What are you spending money on? " I'll tell you what I'm spending money on. How do you stop spending money on that? — Yeah. It's funny. It's been a bit of a different orientation just as a, you know, as a product person. Um, like normally it's like, "Oh, is this on my road map versus not? " And it's like, "Oh, I can tell you exactly what I'm working on, exactly what we're interested in um in exactly how we got there. " And for me, it's like a very that level of transparency. Like I truly appreciate it and I hope the community appreciates as well. — Yeah. Yeah. I think they do. Uh so one more question and then we're going to dive into version 14. Um so off topic but uh but sparks curiosity. What are some challenges in open source versus closed source that were a surprise? Uh yeah so for me personally uh you know I came from AWS where we have this whole like planning cycle of like you write your we call it OP1 and it's like hey write your plan uh why are you doing this what's the what's going to be the expected utility you know what are the headcount that you need or how do you carve all that stuff out uh funny enough when you work with an all volunteer set of contributors uh that type of a topdown approach doesn't exactly work quite the same. And so what's been a fun orientation is to say like bring me what you're interested in. That's really the whole spirit of the project is like — it's not Matt's idea or Sean's idea or Mike's idea or even JR's idea. like we are conduits for the community, but at the end of the day, the how the work gets done is folks are interested in solving a very specific problem — and for them to be able to bring those types of problems to bear is how the specification advances. And so for me, it was a bit of a come bring your favorite thing and like let's have some fun where uh beforehand I was a bit more uh I had a bit more control or uh influence over how we were thinking about what comes next. — And I'll um just put up where's the link on? Uh that's the link to get started in case anyone missed that in the chat. But let's dive into this new feature, this new version of uh Focus 1. 4 uh which will be released in June. So it's not out yet. So it's coming out soon. But what what's it looking like? what's in the new version. — Yeah. I'll uh I'll go back to Sharon's screen for a second. Um so, Focus 1. 4 uh has this like we've always had these points around um we've always had this point around supporting invoices. Uh so it was like invoice issuer name and we did that in 0. 5 and we had invoice ID and that was in focus 1. 2 and that's really helpful. It's like top level granularity for like hey where are things roughly sitting? Um but we introduced these two new data sets and so uh there's a cost and usage data set there is a contract commitment data set which we launched back in 1. 3 and now we've introduced these two other data sets. Um one's called invoice detail and another one is called billing period. And the idea here is that you can get to a very fine level of granularity uh to be able to match your uh the details that you're seeing on your invoice and tie that back to the actual usage that you're observing within your cost and usage report. And so um that level of granularity is a as a net new uh thing for focus. — Wow. — Um yeah, I'm sup I'm super excited about it. Um it's going to be really helpful again when you think through the accounts payable role, the finance role where they're really trying to understand like is this invoice actually closed? Is it still open? Uh what are the boundaries and you know when and I'm just going to make some assumptions that the folks have seen invoices from some of the like public cloud providers. They're pretty

### Segment 9 (40:00 - 45:00) [40:00]

lengthy and the level of granularity that you're going to get on there uh is different than the level of granularity you're going to see within your cost and usage report. And so — uh sometimes those things tie out very neatly and sometimes they do not. Uh but that traceability is really what we're getting after. — Brilliant. So that means the so personas that are going to be using this I'm assuming is for things like when they're actually putting in POS into their finance system and like validating. Is that the kind of work that you see like connected to this? — Yeah, I mean I definitely think there's a PO angle here as well. Uh but I view it more as like before I'm comfortable paying this particular invoice. I want to make sure that I understand how those details are connected between what you know what a provider is saying that I should be owing them relative to the actual cost and usage data that I'm observing um from them directly. — Okay. that like you said the more granular information. So I guess and with things like tags and stuff like that does that mean I'm just kind of branching it across because you can see that talks about this invoice you can be like oh this is percentage wise like what portion was this project and things like that and balance those kind of things out. Yeah, you yeah, you'll definitely like internally, you know, like tags are meant to represent like your business taxonomy and so yeah, you could absolutely uh use some of those additional details to tie those things out. Um, but yeah, like again, most invoices or at least the invoices that I've seen like generally don't have a like a representation or an orientation on tagging. So, — um, just again trying to again if all the numbers match up, but you sliced it three or four different ways, like I think that's really what we're trying to get after is like giving an additional level of certainty uh to the practitioner community in a way that we've alluded to on a rough basis, but sometimes you do need those fine level details uh for — for the accounts payable team or for finance to feel comfortable with paying those invoices. — Brilliant. Okay, cool. What else? Yeah. Uh so the next piece is I touched before that we'd launched a contract commitment data set in 1. 3. You could think of this as like uh this can track your rais your savings plans uh any sort of spendbased or enterprise level commitment you may have with an individual provider. Uh the idea there is we were capturing things like duration and uh amount of the particular commitment that that's being made. But now we have a much deeper level of categorization and payment terms and discount terms like — um so it it gives you much more uh depth into uh the actual contract commitments themselves in a way that just didn't exist beforehand. Yeah, because I'm guessing if you wanted to do that and I'll use the advice example. If you wanted to go find out what was purchased and like you said if it's a three-year commitment or a one year commitment of a compute or an instance, you'd have to go into that console and see that. And often like — I hear more and more these days that people get access to this raw data because they don't have access to anything else. Like we're taking people out the console who don't need to have access to the console, which makes sense. and then they're getting this and then they can actually see this information and you're like, "Okay, well then I can use that in my commitment planning uh cycles of when these going to be up to date and no more are you having to like we said at the start like manually join all these kind of sets of data and stuff like that. " Um but yeah, any other big kind of again work that you see this helping? Yeah. So the way that this was it and again uh just for folks to kind of understand the the project itself, we started with this cost and usage uh orientation and everything kind of branches off of the cost and usage piece. What we were hearing from practitioners is, you know, I'm making these very large commitments with my providers, but I don't always understand like uh how am I doing? like how is my progress against these milestones that I'm making and uh what bit of my cost and usage ties uh based off of the commitment that I've made. What's excluded? Uh I don't actually have my usage data anymore from when I had made this initial purchase. Uh I don't know when things are starting versus ending. Uh I don't know if uh I've actually reached the threshold for me to be eligible for these new credits. If we're talking of something like uh migration assistance, there there's a number of programs, right? Like we're not what we're really just trying to do is look, each provider is going to have their own kind of programs and sets of incentives, but we want to drive transparency between uh the commitments that are being made and then their performance against those commitments from a practitioner side. — Yeah. I mean, like you said earlier, it's all coming from finance practitioners to drive this kind of engagement. But I love that that's going to be in there just people can like make

### Segment 10 (45:00 - 50:00) [45:00]

more educated decisions. Uh what what's next? — Yeah. So uh the next one is focused on commitment programs. And this is a this is an umbrella term that captures both commitment discounts as well as capacity reservations. And so what we were hearing from practitioners is that they're trying to calculate their effective savings rate or their coverage rates. uh and it wasn't always clear to them what an individual row was eligible for or if the information did exist it was generally in the console not with like not nestled up next to their cost and usage data. — Okay. And so we've launched this new column uh called commitment program eligibility details that allows the providers to specify if an individual usage row uh is eligible either for a uh commitment discount or a capacity reservation. — Ah interesting. Okay. Tell me more. — Yeah. So uh again I think it's going to be really powerful. This is actually uh just to kind of reiterate some points with uh how much is changing across versions. Uh this is actually the only new column that we're adding to the cost and usage data set between 13 and 14 — is this uh commitment program for eligibility details. I have a here I have another slide that I can share that touches on this one. So you kind of see some of the additional details here, but it's really around tracking eligible spend that can be covered uh rather than total spend. So again, we're kind of doing that back to that point of like doing the calculation on behalf of the practitioner like — that's a very strong orientation here. — Uh being able to do like a cross provider comparison of opportunities for commitment if that's of interest to the community. And we may introduce like additional program types, but the two that we saw most commonly uh in the market were again commitment discounts that can span across multiple uh services and usage types and then the capacity reservations themselves which again are not oriented towards discounts but as much as ensuring that the capacity is there when you need it. — Excellent. This sounds great. I feel like just giving more people the understanding of the opportunities that they have with this data again all in a central place. Uh Victor agrees. He thinks it's cool. Uh thanks Victor. Um okay, last one. — Yeah. So the last one is really on like how the data is being delivered uh from providers to practitioners. So I'll just kind of briefly touch on this. I don't yet have a cool infographic for this one yet, so we'll get there. But uh this is really around just driving transparency. So we're not as much um we're not trying to take a stance on like what's the right way or the wrong way. Like our approach is that you should be explicit around whatever method you're using as a provider. And so uh like a replacement uh is what I'm expecting from a correction. And again, some of this is based off of how your reports are configured within your provider. So uh whether this comes in as a correction as a replacement or uh this is a delta so you're trying to like make the difference or there's some providers that have like a ledger style approach where everything's immutable so you just need to add things and append it at the end. Uh the stand the delivery mechanism again kind of mirror the correction styles for whether it's an overwrite or it's a an append to what already exists. And then um we have this question at times from practitioners on what am I missing by moving to focus relative to what I'm getting within my native data set. And so we do have this uh this attribute that says hey if this is a critical column within your native data set — we expect you to include it in your focus data set as well. And so there's this pattern for without getting kind of too deep into the weeds, there's this pattern for providers to what we call custom columns. There's like an X, an underscore, and then it would mirror the same — the same name that it would exist. So — Oh, that's cool. — Yeah. So going back to like the AWS example, like cost categories is a very AWS specific thing. Uh cost categories is an X underscore column within the focus export. Ah, I see. Oh, that's so cool. So, people can add these elements in so that you're not like miss except not missing out on something. — Yeah. Uh, and again, if you don't feel like that's the right list of things, like then you have a have an approach to be able to bring it back to your provider and say, "Hey, uh, I'd really like to use this, but I need you to add these other columns. " And then again, they can take that feedback in. — Brilliant. Now, somehow we've almost finished. The time has disappeared. Um, but I do want to make sure we talk about the focus validator before we finish. — Yeah. All right. Um, yes. So, let

### Segment 11 (50:00 - 55:00) [50:00]

me come up here. So, you'll see uh we've added I want what I'm going to show you is the web uh version of the focus validator. There is a GitHub repo that that's linked down below. Uh that will you can run this locally on any focus cost and usage data that you have. And apologies for scrolling up and down, but um what I'm going to show you right here is the readout from a sample data set that's um just meant to show you that the capabilities that the focus validator does. And so, uh what we're looking for here is measuring how well does the data set that you're putting in. So just think of this could be uh internal data that you're formatting into focus if that's uh of interest to you or this could be data that you're getting from your provider but you're going to check it against a specific version. Uh we have this concept of applicability criteria. You don't have to get too wrapped around the axle around files and some things we'll add into the future to make this easier. But uh a provider is saying like hey these are concepts that exist within my data versus not. And so with the example I'm going to share with you, availability zone is not something that's supported. Capacity reservations are not a concept for this particular provider. Um, and then some of the sub account adjustments aren't there. And so rather than show you like a failure, uh, we're just going to skip those checks because they're not necessary because they they're not relevant to the data that you're looking at. Mhm. — And so you as a practitioner can take your sample data, put it in here, run the validate button. Uh again, this is only this is done in memory. We're not storing any information here. So like I hope that gives people some uh some comfort level. And then we're going to have this entire breakdown of like how does this data set match up against what the spec says for this version for this type of data that we're reviewing. And so, — uh, for folks that aren't as clear on kind of how the specification works, there's these, uh, requirements of like a must is something that has to be there. Uh, a should is like it's optional for the provider. A may is like a softer statement to say like uh it's more optional than the should is, but it's like it would be nice if you did this, but there's not uh there's not a way to do this. uh or like you know there's no mandate um from a check type there's things that I can check from the data set those are called static there's things that are dynamic whereas like um I we can't check it from the information that's available in the data and so uh I'll show you a couple of examples that we have here and then you can filter the rules by what's passed what's fails what was skipped uh so for instance like if we just wanted to look at the things that were passed you'll see these adjust adjustments come down. So this data set from Nebus, which is what I'm showing you here, fully supports the build cost uh metric that that's there. And so you can see — um it's present, which is great. It's it has a decimal and it's following this numeric format and it's not null, which is important. Um, — and so again, I won't go down the full list, but I'll also maybe show you something that is that is passed or partially supported. Uh, so something like a consumed unit is a good one. Um, where it's saying, hey, like, well, this top level rule failed because this underlying unit failed. And so, you know, I would need to go back and look and see like, hey, uh, this data set that I'm looking at, the unit format expectation that I'm seeing like is missing like it's it doesn't have uh it doesn't have values that are matching what's already published within the spec. And so could be like not a huge issue if it's very bespoke or specific or maybe it's a call back to us as the project that says hey we need to expand uh the unit formats that we're actually allowing in. But uh this for from a practitioner side, if I'm trying to, you know, marry up my AWS data and my Nebas data together into a single unified data set, I'm going to want to understand where there's differences across those values. And this gives you a very deterministic way of doing that. — Excellent. Wow. What a good tool. And I think it's like you said, good for people to be able to ballet and not have to spend — all the time themselves — just trying to like um — Oh no, me and someone keeping to take the screen down. Um uh keep having to war on doing that. So let me just pull that up again. So uh if you want to check that out, have a go. And with that, we're almost at time. I can't believe um how far the time has

### Segment 12 (55:00 - 58:00) [55:00]

gone. Um just maybe like one last thing if you really want to suggest people to like go and try focus maybe who haven't done it before what would you say to them? — Yeah. So again if your provider supports focus today I would encourage you to go and follow the steps for enabling that export. If it if it's an export or if there's a billing API that they're vending focused data. I would encourage you to evaluate that against what you're currently doing today within your practice. If you're if you don't if you're not on a provider or like maybe your company's not ready yet or maybe you're blocked on permissions or something else, uh we have a set of training within the foundation at learn. finops. org uh where we have focus analyst training and we update that uh once a year. So my next activity after X is updating our training to account for 13 and 1. 4. before. Um, but like that's a very practical way of both understanding how the specification works, the expected utility behind it, the relationship between the columns, why that's meaningful, what right looks like, what raw looks like. Uh, we're really trying to make it a more interactive and engaging uh, training experience. Um, but yeah, those are kind of my two primary points. like I'm a very hands-on guy. So like I want to see the data and play with the data and — slice it and dice it in all the ways that uh that make me feel comfortable with the quality and safety of the data. Um but yeah, like I think those are the two primary paths and again if uh if any of this is of interest to you and you're going to come to Phops X, please come swing by the focus booth. I'd love to have a discussion with you. we may have some Rubik's Cube and all the swag stuff uh that's focusoriented but um and if you're not coming to FOPS XX like I would encourage you to consider coming uh I do think there's a ton of — uh again I don't think I've met a single person that hasn't enjoyed their time there. San Diego is pretty beautiful in uh you know early June. So there's a couple of opportunities depending on where you're coming from. — And Matt, if they want to give me another one of these water bottles, I love it. So just a shout out. It's one thing if I if anyone wants to give me one of those uh nice but yes if you are coming uh make sure — if we're doing uh if we're doing this is my old AWS I remember those love it. Uh so quickly before we finish uh if you are coming to Adam and Brent and mine session uh use the discount code keys to AWS to get your ticket down to 699 uh which would be amazing. Um also make sure you check out the uh as party come say hi chat all about your phop journey. If you want to see summaries of the episodes check out builder center. This is where we post uh a summary of each episode that we stream. Uh you can see password ones and my Monday show that I do every other week with large customers all about how to build with AWS. Uh best way to do this is to scan the QR code, follow me and drop a like on any other content that you enjoy. And finally, check out the customer acceleration team uh LinkedIn group. This is where we post all new and upcoming episodes and content from our team including this show. That is it. Uh whistle stop tour at the end. Thank you so much everyone for watching and uh hopefully me and Matt will see you at Phopsex. Thank you so much. — Awesome. Bye everyone.