Ep#153 Achieving Financial Excellence in the Cloud #FinOps

Host: Jon

Hi everyone and welcome to the Faces of #FinOps, powered by our great friends at ProsperOps. I'm your host, Jon Myer. Faces and #FinOps podcast is all about highlighting the thought leaders in the cloud financial management space and how they're making an impact not only within their organization but within the broader #FinOps community. Today's guest is Matt Mazer, a #FinOps manager at a Fortune 100 company. Matt is an accomplished IT professional with over 20 plus years of experience and holds an undergrad degree from the University of Miami, also an MBA from Franklin University. Please join me in welcoming Matt to the show. Matt, thanks so much for joining me.

Guest: Matt

Thanks for having me. I'm excited to be on the pod.

Host: Jon

Matt, we've got a bunch of questions here for you and really kind of centered around your #FinOps journey and your expertise, but how about you share a little bit of information about yourself?

Guest: Matt

Certainly. So, Matt Mazer, I live in Dublin, Ohio. As you've stated. I have a little over 20 plus years of IT experience and my most current role is a #FinOps manager for the past three and a half years.

Host: Jon

Let's talk about your current role as a #FinOps manager. Really, what does that entail?

Guest: Matt

That is a great question and I feel my current role has many hats, so for those that follow the #FinOps Foundation, I support all of the domains and capabilities, but my current role revolves around much more than just the basics of #FinOps. So first we are, we are managing cloud spend for AWS, GCP, and a small little bit of Azure. Then we also get involved in contract negotiations, so enterprise agreement, new tools, bringing new tools into the environment. Then we also are responsible for the visualization layer, and so we are responsible for making sure that our tool is up and running, making sure that the data is correct, and making sure that our end user, whomever it may be, can get the segmented data that they are looking for. Then we also need to work on our roadmap and that is what #FinOps looks like now.

Guest: Matt

What does it look like a year from now? What does it look like three years from now? Then there's a whole host of other activities that I jotted down just in case, so I'm going to glance down, so data integrity, making sure all of our data is correct, so we ingest data from our cloud providers at different points. Once we ingest the data, sometimes there are data challenges, so then we are firefighters to make sure our end users always have accurate data. We partner with various teams, so we're responsible for assisting with forecasts, assisting with budgets, and also future planning.

Host: Jon

Well, Matt, let me jump in there and ask a couple of questions. Not only about your role but your organization and the #FinOps function within it. How would you rank your maturity level within your organization? Is it you in a crawl, you in a walk? Are you in a run stage throughout all of this?

Guest: Matt

Yeah, I would say for the most part we're in walk or run on almost all aspects of #FinOps. As an organization, we've had the #FinOps practice for a little over six-plus years. So many items such as the visualization layer, our committed spends, and our tagging. Most items we are walk or run. One thing I would say is, and this is probably true for all organizations, it is a continuous process that is never-ending, so anytime I feel like we're kind of close to the sun or as close as we can get on certain things, there's always more that can be done to optimize our spend.

Host: Jon

Matt, you said that are you by having a chance like in a crawl in one cloud or run in another and a walk, or does all of that sync up together where you're in a walk within all three clouds?

Guest: Matt

For our two primary, we're that walk to run, and then the other one that's a much smaller portion of our spend. We're crawling in that environment and the reason is there just isn't as much opportunity or dollars being spent and being candid. We don't have the bandwidth to go after all things. We try to prioritize based on opportunity and where we would see our return on investment.

Host: Jon

Now, how would you describe the process for implementing #FinOps culture for multi-cloud? Was it difficult? Was it easy to achieve following the rules and guidelines for #FinOps or how would you kind of rank that?

Guest: Matt

I would say it is a challenge because ultimately most of our end users, so our VPs or our cost center owners, want to see a single pane of glass and they just want to see their costs. They don't necessarily care if they're in two clouds, they don't care if they have a committed spending agreement. All they want to do is understand what is their cost and what are their levers to optimize that cost, and so it has been a little bit of a challenge to get them to that single pane of glass. We're doing a good job. Again, with all things in #FinOps, there's still room for improvement.

Host: Jon

There's always room for improvement. As you progress through the crawl, walk, and run, you implement different variations of the phases where you know what you can tweak, you tweak it and you improve little by little. Would you agree?

Guest: Matt

100%. When we come up with our yearly goals or modify our quarterly goals, oftentimes we remark amongst ourselves, Hey, we're not bringing anything groundbreaking or earth-shattering or really like net new. It's we kind of do this, we need to continue to improve upon this, or we've been going at it and we're at 80% against our committed spend. Can we go to 85? Can we meet with additional teams? If it's something in storage, can we drop it to lower-class storage because we don't necessarily need it? It's always tweaking, but it's not necessarily anything earth-shattering or brand-new.

Host: Jon

Matt, would you say it's continuous improvements and it can be small chunks at a time, but you're always moving forward and trying to improve? Little by little

Guest: Matt

It is continuous improvement, and again, I've been in my role for three years, so the optimization opportunities when I first started were much greater than what they are today, and the reason is we've refined, we've evolved as a company, we've gotten rid of a lot of zombie resources. We've implemented start-stop tags, and we've committed spending. All these things have decreased our costs or optimized our spending, so as you go forward in your #FinOps journey, it becomes more and more of a challenge to save the big dollars.

Host: Jon

Mary, are you using any type of automation to complete some of these tasks not only in the optimization efforts but say in the reporting structure?

Guest: Matt

Yes, so we leverage certain automation, so the simple answer is yes. We also do some anomaly detection so people know we've also automated reports, so an end user can see when they have spikes and they can address those promptly versus waiting, let's say till the end of the quarter, so they now get the highlights or the reports every week they can take action. Where we'd like to head as an organization is to take it even a step further where you get the anomaly and then it's almost like an opt-out where the change will be automatically implemented if you do not opt-out, we are not there yet. That is on our roadmap for the future. Right now you get an alert, but then you still have to take action.

Host: Jon

I like that you have to opt-out before the action's taken for optimization, so you have to validate that the action is worthwhile and you have to invest, kind of look into it, and dig into it a little bit more.

Guest: Matt

Yes, absolutely.

Host: Jon

Now, the reporting that you were talking about, is that daily reporting? Is it a weekly report in real-time? I mean, I know you kind of went and touched on the anomaly detection, but is a report generated automatically or is this something that you're looked at weekly?

Guest: Matt

Yeah, so it depends on the subscriber of the data. So some of our cost centers and VPs would like to see their data daily for one needs to see the data daily, so I get a daily report. I can see if there are any spikes in our overall cloud spend. I have some end users who want to see it every week, some who want to see it on a biweekly basis, and some who want to see it every month. What we do is we set up standardized reporting, but then the frequency that they receive the data is really up to them.

Host: Jon

Do you find it challenging that the reporting might be out there biweekly, monthly, or something like that to perform this action? Is reporting supposed to be daily or is it really kind of based on your environment? In the beginning, it was done daily, but as you progress and you mature a lot longer, it's done monthly.

Guest: Matt

No, it varies by business segment or cost center, really what their spend is. For the most part, most of our spending is somewhat stable at this point. We do support projects, we do support mergers and acquisitions and there is testing and that kind of stuff, but our cloud spend is much more stable than it was in the beginning, and so most of our end users do not require the daily reports. I still want to see the daily reports because I want to know if there are any major spikes. I also want to know if I have a data integrity issue before I have 15, 20, or 50 people saying, Hey, what's wrong with the reports? My spending is out of whack. That probably wasn't the best grammar, but spending is not what it was supposed to be, and so I want to be preemptive so I can send out a mass email just saying, Hey, we know we have a data integrity issue. We understand this. We expect to have it resolved within the next 24 hours. Look for the next update, and that way I'm not feeding questions. I'm also building trust amongst my end users so they know what to expect. They know that we're on top of our data sets.

Host: Jon

I think you just nailed it right there. The data integrity and building trust amongst the users is that if the data is incorrect, then they're not going to trust the actions that need to be performed to optimize or reduce spend or even scale or grow an environment

Guest: Matt

That is 100% true, and I will say that early as a team, we at different points had data integrity issues and sometimes there were small instances where we found out about it when a team was drilling into the data. There's no worse feeling than having a team come to you. It's our data, our reports, and they're saying something doesn't look right. I always want to be ahead of it, understand if I have a problem, and again, build that trust so that people feel empowered that the data they have is correct, it is accurate and they can take action on it.

Host: Jon

Matt, let me ask from your expertise, what are some of the biggest mistakes that you see maybe an immature #FinOps team implement or try to achieve?

Guest: Matt

I would say not spending the time upfront to establish tagging governance, so everything as it comes back to any time you want to segment the cost, it always comes back to tagging. We are, as I said, many years into our journey and we're still doing tagging cleanup, When you're trying to find an owner, if a tag's incorrect, it's a pain, it requires manual intervention, there's lost stuff where you don't know truly where to assign the costs, then my guess is your organization will need a dumping ground, a centralized for anything that you don't have. An owner goes to that cost center, but my advice to anybody just starting, invest in tagging governance and make sure you're tagging everything upfront and you have good policies around

Host: Jon

It. #FinOps culture and tagging are proper and required for a successful environment. That was just a little bit kind of digging into it and talking more about it, I just asked you about some of the mistakes you think in immature, but I feel that tagging even a mature #FinOps culture still makes those mistakes. Is that one thing that you see out there or is there a recommendation where you can see maybe a mature one can make or kind of learn from?

Guest: Matt

Yeah, so you mentioned it. So yes, we still have tagging challenges in our company, so we have automation. We have a governance team that is checking all of the tags. That being said, there are limitations to our checks and balances, so bad tags still come through the system. Then my team has to clean them up whenever we're doing the chargeback. So yes, tagging is imperative. It seems like a never-ending battle, so I would just say good luck. If you could force tagging to be almost like a dropdown where only valid cost centers and valid tags are utilized, I think that would be great. We as a company aren't there where it could be just a total dropdown so that it would avoid all incorrect tags, but it's never-ending.

Host: Jon

I think even cloud providers could offer that as a dropdown rather than an open field where you can enter any text up to like 50 hundred characters. It accepts all of them, but the problem with that is that it also accepts uppercase and lowercase, and those are unique in some cases. So Matt, before we wrap things up with our first part, you were the third #FinOps hire within your organization. What were some of the key things that you did once you entered the company? What were some of the things that you wanted to implement or tried to do?

Guest: Matt

I come from a PM background, so I've always been, I don't want to say taught the three key project management principles, so I was concerned that we were consistently as an organization missing our budget and missing it by over double, and so what I wanted to do was understand the budget. I wanted to have metrics in place to understand exactly where we were at against budget and then try and provide solutions to get us closer to budget. So that was my focus to understand the cost and get us back on budget.

Host: Jon

What were you doing before joining your current company?

Guest: Matt

Yes, so my prior experience actually was as a program manager for another Fortune 100 company, and then I was traveling a lot and family and personal situation, elected to switch companies, came to my current employer, joined as a project manager, and quickly excelled as a project manager, became a program manager. Then from there became a people manager of project managers. Then we did a major initiative at my company, became a program manager again, and then from there I supported a lot of cloud initiatives where we were moving from on-prem to cloud solutions and it was a natural transition to get back into people management, and I viewed #FinOps as the future. I always like to tell people it's a unique opportunity to influence the company's bottom line, and I get excited about the financials and the opportunity to influence the financials the #FinOps manager role seemed a great fit for somebody with my background in project and program management and interest in the finances and an interest in people management.

Host: Jon

I think your project manager background helped you because you did the cloud migrations, you went through some of it, you've seen what they look like and you were part of the budget and you've seen the cost for it, and it was a natural progression into a #FinOps manager. It just feels like a normal progression to me.

Guest: Matt

Yes, I would agree. Several of our team members got their start in project management, project management, and program management gives you a diverse background and allows you to see how a large company works, see what's important to the company, and then see how you could fit in and help and influence positive change.

Host: Jon

Well, Matt, speaking of your team, what are some of the things that your team is responsible for?

Guest: Matt

So my team has, again, on part one, discussed some of the many hats that I wear. So my team is responsible. We have a billing arm, so they are responsible for the chargeback, the POS, and supporting our sourcing team. Then I have a team of advisors, they are your frontline. They are providing support to various cloud champions, and VPs. They're assisting with our visualization tool. They're helping with optimization opportunities. They're helping to set up reports. They also have an education component. They're teaching teams #FinOps basics. They're the frontline. And then I have an engineer team and the engineers are maintaining our visualization tool and they're also working on enterprise solutions. So when I say enterprise solutions, it's multifold. So you have the roadmaps, you have your enterprise agreements with your major cloud providers. Then you have your secondary agreements with your third-party tools. Then you're also looking at enterprise committed spend discounts that they're assisting with, so they are also wearing many hats. So in the current role, again, it's many hats, and many responsibilities, all with the ultimate goal of optimizing our cloud spend.

Host: Jon

Matt, is your a centralized #FinOps team or decentralized or are you using a variation of both?

Guest: Matt

Yeah, so we are central. We're considered an enterprise team. That being said, at our company, we do not have at a lot of other companies, there's a Cloud Center of Excellence (CCoE) for all of the cloud. There are still silos within our company. So I guess you'd say a little bit of everything,

Host: Jon

A little bit of a mix, but it still works for your company. There's now, while centralized is recommended, you could do decentralized, you can do a hub and speak, but whatever works best for your company as long as it's efficient and achieves the results that you're looking for. Right,

Guest: Matt

Yes, I would agree 100%. The other thing I would add, and you see it in most of the #FinOps and high-performing teams, really having strong executive support. I think that's a requirement when you have key executive buy-in, they help influence change, they help get the attention and they help push for optimization.

Host: Jon

What does your typical day, week, or even month look like within your organization?

Guest: Matt

Yeah, so at the beginning of the month, it's making sure that all of our chargebacks from the previous month are complete, making sure all the costs are being distributed evenly. So the first week to two weeks, it's all about chargebacks and data integrity, and then it quickly shifts. Once the chargebacks are done, then you're kind of going on what I'll call the roadshow or the disseminating of all the information. So we have governance reviews. That's at the executive level. You're showing the cost breakdown of all their expenses, including all of the previous month's charges. You're partnering with finance, you're partnering with sourcing, you're showing everything that's coming up as well. So first part, get your data right. The second part of the month is presenting the data and in between you're always maintaining that data integrity, so making sure your reporting's always up, always working, and all the data is correct. You're constantly on a basis looking at additional committed spend purchases. You're also doing vendor meets because you want to make sure that there isn't a new tool that could significantly aid in your #FinOps journey. It's nonstop, but it's a fun space to be in because every day is slightly different and every day allows you to learn something new.

Host: Jon

What are some of the biggest challenges you or your team are facing today?

Guest: Matt

Right now we have a few. So our data normalization for our visualization tool is one. So we have manual workarounds to get all data from all different providers to look the same so they can see a single one-page view of where they're at with cloud spend. Another area is just how can we continue to push on optimizing our costs and is it additional committed spend discounts? But we're pretty good there, so we're running out of options there. So then it's looking at additional tools. Is there anything else? It's meeting with our primary vendors and seeing if they have any suggestions. It's scouring like Google Recommender a w s, kudos. It's nonstop. We are a low-margin business and we are consistently challenged with optimizing cost, so we're really, well still meeting the business needs. Our clients have stringent requirements, so it's getting creative and finding ways to optimize.

Host: Jon

Thanks, Matt. Let's talk about what's next. Do you see that possibly AI or automation will be impacting your role or #FinOps in general?

Guest: Matt

Yes. I'm a firm believer in that. So funny story, I was just talking to somebody at lunch about this. In the early part of my career, I was a provisioner for a telecom company, and so that's how I got my start in it. And so what used to take me a day, my whole workload in a day now is completely automated. It can be done in less than a minute, and that's only 20 years. So I think we're headed in that direction. That's probably not good for me to say, but I imagine a lot of what we do today and a lot of what takes a lot of time for us will not be the same case in 20 years. I don't see the role as it currently is structured. I don't see it in the future in 20 years, 10 years, yes,

Host: Jon

I think automation or even AI is going to enhance the #FinOps role and make it more fine-tuned that we're going to be able to do a lot more. Man. Let's talk about some advice that you might give somebody or a company that's just trying to get started on their #FinOps journey.

Guest: Matt

So my advice I guess would be executive sponsorship. So really make sure oftentimes #FinOps doesn't have true authority to force change. So you need that executive sponsorship. So executive sponsorship would be one of the first things. Then tagging governance, which we've discussed. And then the other thing would be to just get started. You're never going to find the perfect solution. There are going to be different times you're going to over-purchase. There are going to be other times you under-purchase. My experience has been if you work with your vendors, they will oftentimes work with you even if you've committed because oftentimes you're still going to make an additional larger commitment and they'll work with you. So don't be so afraid to get started that you can't make a decision start. Just get started, and set it up. Don't worry about doing the whole Big Bang. Maybe do a small bit in September, then another bit in October, but get started.

Host: Jon

So, Matt, I feel like I got to throw data integrity into some of the advice that you would give for teams that are just starting in their #FinOps journey. You've mentioned it a couple of times throughout this episode, and I think that's key. As much as you emphasize having the correct data is have the trust in your engineers. What do you think?

Guest: Matt

Absolutely. So data integrity is paramount. If people do not trust the data that you are providing. You are a check the box, Oh, we have to talk to #FinOps. You never want that. You want to have a partnership and you want the data to result in success stories. That way when Jon does something great, he could say, Look, we started at a thousand dollars a month. I followed #FinOps best practices, turned off, and did some auto stops on some machines. I resized this. We went from a thousand to 800. Now, not only do you have a success story, you have a success story with data points and everybody has a success story. So now engineer two, Bob wants to do the same thing and he's coming to Jon and saying, how did you do that? I want all the same kudos that you did that you received. So data integrity is paramount.

Host: Jon

It's like internally evangelizing the process and the journey to implement it. Then you have the trust of the engineering teams. I like it. Matt, let me ask you that question a little bit. In Reversive really, you have a #FinOps, a mature culture, an established team already in place. Is there any advice that you would give that team

Guest: Matt

Celebrate successes and continue to push the envelope. There are always more opportunities and do not be afraid to fail. If we fail, let's have our data behind us. Let's demonstrate why we're making certain decisions, but let's not be fearful. Let's continue to try and push and just be ready to explain if it goes wrong. But so far, knock on wood is calculated. I don't even want to call 'em gambles, but our calculated decisions have paid off, and I'd like to continue that mentality. And then the other thing I would push even for us as a mature organization, continue to work to nurture that executive sponsorship and get them on board when they are on board and they are excited. It just seems like the whole process runs smoother and it's good to do stuff ahead of the fiscal year. So don't wait until the last six months of the fiscal year. It's a lot easier to make change influence change and see real results when you have a longer runway.

Host: Jon

Well, Matt, this has been a very exciting and informative discussion, not only between your journey but your company's journey and the teams implementing the #FinOps culture. I'm going to switch gears a little bit and let's have a little fun with a couple of interesting questions. What do you think?

Guest: Matt

Sounds great. Let's do it.

Host: Jon

All right. So Matt, let me pick a question. I got one for you. So Matt, have you ever had the OG of iPods? Did you have the original one, like the Nano or the original?

Guest: Matt

I did have the Nano, yes.

Host: Jon

Okay. All right. So imagine that you have the Nano and you're on an island. What music do you have on this?

Guest: Matt

Oh, can it be current music or do I have to think back to what was originally on the Nano?

Host: Jon

That's a great question. I don't even remember any of the music that was the Nano at that time. You can pick any kind of music, but what is on it?

Guest: Matt

So right now, believe it or not, I am a huge country fan. So it would be Morgan Wallen. I know he's super popular. I know he is selling out stadiums everywhere, but that is who I would be listening to. And if you went back 20-plus years, I think again, probably you wouldn't guess it, but probably like Jay-Z or Ice Tea or some Wu-Tang Clan.

Host: Jon

There you go. Hey, you know what? I enjoyed the country and your list on the OG back 20-plus years ago. So I'm right there with you. Matt, I'm going to wrap things up and ask you who are some of the most influential practitioners in the #FinOps community now?

Guest: Matt

Yeah, so being local to Columbus, Ohio, my first recommendation for anybody interested in #FinOps is always Joe Daily. So he's dynamite. He is very responsive, very engaged, and very active in Columbus. And then other people that I follow more from just on my LinkedIn feed would be JR. Who's in charge of #FinOps Foundation, then Ben De Mora, and then just somebody that I met at #FinOps X.

Host: Jon

So, Matt, I'm going to wrap things up with our latest episode of Faces in #FinOps everyone. Matt Mayer, #FinOps manager at a Fortune 100 company. Matt, thank you so much for joining us.

Guest: Matt

Thanks again for having me. It was a blast.

Host: Jon

Alright, everybody, this has been another awesome episode in discussion around faces in #FinOps, powered by our good friends at ProsperOps. Be sure to hit that like subscribe, end notify, and check out the latest ProsperOps blog because guess what, we're out of here.