The Cloud Cost Playbook

Over the past few years, SaaS standards have been shifting. As the economy has become more tumultuous, investors have been weighing SaaS companies more on their profitability than their growth. The Cloud Quarterly report by Battery Ventures was the first to capture this shift around June 2022.

The report confirmed that companies unable to improve their margins are seeing their valuations plummet.

The result: These companies may not be able to raise as much capital as they would like, which could stall their growth.

The shift emphasizes the need to balance growth with profitability.

Yet, visualizing, understanding, and controlling cloud costs can be challenging. Figuring out where to start and what to do next can be intimidating, too.

In this guide, we’re not just scratching the surface of FinOps and cloud cost optimization – we’re diving deep. Think of this as your GPS through the often foggy and complex world of cloud costs.

Whether you’re a small team feeling lost in the clouds or a large organization navigating through a storm of expenses, this guide is your beacon to clearer skies and smoother sailing.

Unlike traditional IT expenses, cloud costs are dynamic and can scale up or down based on usage. This is where the magic and mayhem lie.

Imagine driving a car without knowing how much fuel you’re burning. Scary. That’s what unoptimized cloud costs are like. 

The more companies spend on the cloud with untrackable ROI, the more it eats into their profitability, slashes their valuations, and makes it harder to survive — let alone thrive.

From cloud storage to data analytics costs, they can spike without warning, leaving your budget in a tailspin. 

Optimizing these costs ensures you’re not just throwing money into the wind but investing it smartly to propel your business forward. This is where FinOps, short for Cloud Financial Operations, comes into play.

First time learning about FinOps? Or, want to refresh your memory?

Start here:

Efficient cloud cost management leads to:

Yet, to achieve these feats, you need to shift away from traditional cloud cost management . 

Where do most companies fall?

This may seem familiar if you are already doing it. If not, don’t worry, we’ll share more effective ways further down this guide.

Yes, it’s like choosing a route for a road trip based solely on distance, ignoring factors like traffic or road conditions that could affect fuel consumption and travel time.

So, where do you go from here?

Take a closer look at the FinOps Foundation; you’ll notice it moves through three major phases. It is a “Crawl, Walk, Run” maturity model that encourages learning, action, and improvement.

It all does not have to be flawless from day one. 

We’ll break down each phase into real, actionable steps you can take to implement a working cloud cost intelligence strategy. Applying this FinOps maturity model can help you stay on the right track to cloud efficiency.

The Crawl stage describes the initial phase when organizations have very little reporting and tooling, basic KPIs are set, and only basic processes and policies are defined. The goal at this stage is to understand the FinOps capability rather than implement it across all major areas of the organization. 

Would you prefer to start FinOps from scratch?

Start here: 10 FinOps Diagrams To Help You Better Understand The Value Of FinOps

At this point, you’ll want to target “low-hanging fruit”. To put it another way, it’s time to look into traditional cloud cost savings.

Step 1: Traditional cost savings: strengths and limitations

Imagine these as the foundational tools in your FinOps toolkit.

What are traditional cost savings?

You find traditional cost savings in the cloud similarly to how you would reduce your monthly personal bills: Buy in bulk (Reserved Instances (RIs)), take advantage of off-peak rates (Spot Instances), and turn off the lights when not in use (terminating unused or idle instances).

Resource: 13 Key Cloud Cost Management Strategies (And How CloudZero Can Help)

Benefits and limitations of traditional cloud cost savings methods

While these methods can yield immediate savings, they’re a bit like using a hammer – effective for certain jobs but not a one-size-fits-all solution. 

Their benefits include:

However, their limitations are notable, too:

Here’s a snackable overview of common traditional cost-saving methods you can apply right away:

AWS Savings Plans

These are commitments to spend a specific amount in dollars per hour over a one- or three-year period. Think of it as buying a season pass for a service you frequently use. Features include:

Learn more about AWS Savings Plans here:

Reserved Instances (RIs)

These are a commitment to use a specific type of resource for a defined term of one or three years, often at a discounted rate. All three major cloud providers offer RIs, unlike Savings Plans on AWS. So, what do you get here?

Learn more about Reserved Instances here:

Rightsizing

“Rightsizing” means using the resources and instance types for the job – not too little, not too much. Yes, it’s like tailoring your cloud wardrobe so it fits just right. And that’s great, because rightsizing …

Learn more about rightsizing here:

Spot Instances

AWS, Azure, and GCP all offer huge discounts on their surplus cloud resources. But, if demand peaks, the cloud provider can reclaim the resources and assign them to other clients. 

Depending on the provider, you’ll only get a 30- to 120-second notice. It’s like flying standby – cheaper, but with some uncertainty. So, why try?

Learn more about Spot Instances here:

On-Demand Instances

On-Demand Instances use pay-as-you-go pricing in exchange for maximum flexibility. It’s the cloud equivalent of a taxi ride; available whenever and however you want it. The downside is that flexibility makes this model the most expensive.

Learn more about On-Demand Instances here:

Data Transfer Fees

These are costs associated with moving data in and out of the cloud. Think of them as shipping fees for your data. Understanding and optimizing these fees can lead to significant savings.

Learn more about data transfer fees here:

Automated Cost Optimization (with ProsperOps and Xosphere) 

These tools automate cost savings, like having a smart assistant to find the best deals for you while you focus on other areas of your business.

Step 2: Getting engineering, finance, and FinOps on the same page

This step is a lot like bringing together chefs, accountants, and restaurant managers to create a recipe for success.

Getting these teams on the same page is not without its challenges. It’s like asking people speaking different languages to work together without a translator, and then some. Common challenges include:

How do you bridge the gap? Start here:

Learn how to align engineering, finance, and FinOps in more detail here:

By fostering collaboration between these departments, you can reduce the barriers to effective communication. It’s a team effort, and each member plays a crucial role.

Step 3: Complete cost allocation: ingest 100% of your cloud spend in a single pane of glass

Imagine having a crystal-clear view of every penny you spend on cloud services. This is what complete cost allocation does — it brings all your cloud costs into focus in one place.

Cloud cost allocation is the process of identifying and assigning cloud resource costs to the specific users, departments, projects, products, and/or any other entities that use them. It’s like itemizing a restaurant bill to see exactly who ordered what.

Cloud cost allocation methods

Tagging: This involves labeling cloud resources with identifiers (tags) to track their usage and related costs. It’s like putting name tags on dishes to identify who ordered them.

Resources:

Code-driven approach: Here, you use software and algorithms to allocate costs based on usage patterns and other metrics. Think of it as using sophisticated AI to automatically identify who ate what.

Resources:

Telemetry: This involves collecting and analyzing data from cloud resources to allocate costs accurately. Imagine it as having sensors on each dish to track consumption details.

Resources:

Tagging is like using a manual ledger; it’s straightforward but labor-intensive and imprecise. Code-driven allocation and telemetry are like using a smart accounting system; they requires some setup initially but offer far more accuracy and insights over time.

Also, why a single view of cloud costs? Consider the following:

The Crawl stage also presents two great opportunities. 

Dealing with shared costs

Shared costs in the cloud are like shared dishes at a dinner party – they need to be fairly allocated among the participants. Proper cost allocation methods ensure that each department or project is charged its fair share, based on actual usage or an agreed-upon formula.

Start here: Solving Shared Cost Allocation With Telemetry-Driven Cost Organization

Improving your tagging

While tagging has its limitations, it’s still a valuable tool in the cost allocation toolbox. Improving your tagging strategy can enhance its effectiveness, much like refining a recipe can improve a dish’s taste. Want a deeper dive into how to improve your tags without endless tagging?

Start here:

The clarity you’ll get from implementing Step 3 here can be a game-changer for making smarter cost decisions. In addition, you’ll have what it takes to proceed to the next FinOps stage; the Walk stage.

The Walk stage is the intermediate phase where your teams deepen their understanding of FinOps concepts. It is now easier for engineering and finance to understand how their activities affect cloud costs. And leadership is actively encouraging the teams to routinely assess the business value and cost impact of their decisions. 

Ultimately, the Walk stage is about getting everyone on the team to understand the cost implications of their actions and to take responsibility for them. 

Let’s explore how to achieve this.

Step 4: Go beyond traditional cost savings with next-generation cloud cost intelligence and optimization

Cloud cost intelligence is like having a financial advisor specifically for your cloud costs. It extends beyond mere cost tracking to provide deep insights into how and why resources are used, and their impact on your bottom line.

The main focus of this step is to understand the cultural shift you need to make to reap the benefits of cloud cost intelligence. And it pays. Here’s how.

Resource: What Is Cloud Cost Intelligence And Is It Right For Your Business?

Adopting cloud cost intelligence offers the following benefits:

With precise data, you can make smarter choices about resource allocation, like a chef carefully selecting ingredients for a dish to reduce waste.

The table illustrates the transition from traditional, reactive approaches to next-generation, proactive optimization. It highlights the evolution from simply managing costs to actually understanding what drives your cloud costs so you can maximize their returns.

	Traditional Cost Management	Cloud Cost Intelligence
Focus	Reducing bills, managing committed discount contracts, and rightsizing resources	Understanding the who, what, and why that’s driving your cloud costs so you can tell exactly where to optimize usage (thus costs) without sacrificing performance or engineering velocity
Approach	Top-down, centered around procurement strategies	Holistic, involves technical and financial perspectives
Engagement	Mainly a finance concern: better buying, not better building	Takes a cross-functional approach involving engineers, finance, and FinOps
Visibility	Limited to contract and usage metrics	Deep insights into usage patterns, cost drivers
Cost Analysis	Based on historical data, reactive in nature	Proactive, data-driven, supports forward-looking strategies
Flexibility	Rigid, often locked into longer-term contracts	Adaptive, aligns with dynamic cloud usage

Step 5: Use engineering-led optimization to promote a cost-aware engineering culture

This approach integrates cost considerations into the engineering process. It’s about giving engineers the tools and insights they need to understand and manage the cost implications of their design, coding, and deployment choices.

Here’s what to do here:

You can also offer incentives to foster cost-conscious engineering. These can range from recognition to bonuses to priority project considerations, to name a few.

Resources: 

As you might have noticed, the key component of cloud cost intelligence you’ll need here is understanding your unit costs. And that takes us into the next stage of your FinOps journey.

In the “Run” phase of your FinOps journey, the focus shifts to maximizing profit. This is where the strategies and practices you’ve implemented start paying off in practical ways. It’s about turning efficiency into a competitive advantage and cost optimization into increased profitability.

At this stage, your teams should have a solid understanding of their capabilities. You can allocate over 90% of your spend, address difficult edge cases, set high KPIs for measuring success, and make automation your preferred approach. 

In addition, key metrics include achieving 80% resource-based commitment discount target coverage (higher is better) and about 12% variance between forecast and actual spend (the lower, the better).

Step 6: Make unit costs your North Star metrics

Unit costs show the costs incurred for each discrete unit of cloud resource consumption. This could be expressed in terms of cost per customer, cost per transaction, cost per feature, cost per product, or any operational metric relevant to your business’s services or products.

Traditionally, organizations divided the total costs by total units to find the average cost per unit. Because that approach averages out every activity’s usage and cost, it doesn’t actually show how each service is impacting your cloud costs.

A unit cost approach shifts the focus from viewing total and average costs to pinpointing how those costs relate to specific aspects of your business. This is what unit economics is about.

Resources:

CloudZero emphasizes unit costs to align cloud spend with business outcomes and activities. It’s not just about how much is spent on the cloud, but about assessing what that spend achieves and how it drives value for your business.

This perspective allows for more nuanced and strategic decision-making, such as enabling you to see exactly which service or usage you can tweak to reduce costs without sacrificing performance or engineering velocity. This is what true cloud cost optimization is about.

It’s like a restaurant knowing the cost of preparing each plate, so they can tell how much to charge for each to protect margins per plate.

Examples of cloud unit costs and how to use them

You may find the following unit cost metrics useful based on your business model.

Cost per customer

This metric indicates how much you spend serving one customer. It is a great way to identify your most profitable and most expensive customers.

Cost per customer lets you set custom pricing that is fair to your customers (considering how much it costs to serve them) and profitable to you (healthy margin for each contract).

By identifying your most profitable segments, you can develop better go-to-market strategies to attract, convert, and retain more profitable customers. 

Results: Higher margins. Increased profitability. Improved valuations. Ideal investors. And you’ll be able to substantiate healthier metrics, get better financing terms, and have more opportunities to grow. 

Similarly, cost per customer can help you determine how much discount to offer a specific customer to encourage renewals.

You can also better predict future costs when you know your costs per customer. For example, you can more accurately estimate how much your costs would increase if you onboarded five or ten similar customers within the next week.

Resources: 

Cost per feature

Cost per feature is an insightful way to measure the economic viability of a particular feature. It enables you to compare the cost of developing, maintaining, and upgrading a feature with the amount of revenue it generates.

You can use it to determine whether a feature is popular enough to warrant investing in it, such as rolling out updates. 

For instance, if it is popular but expensive to maintain, you can consider moving it to a paid tier. If not, you may refactor it to serve a different purpose or decommission it altogether and focus on more sustainable features.

Cost per environment

Cost per environment helps you assess the cost of an environment relative to the number of environments. 

This metric is great for tracking the cost of maintaining an environment over time. It can also help you identify which environments require the most resources and can be targeted for optimization at the architectural level.

Cost per team

Cost per team can help you tell which teams are using more resources than others, allowing you to allocate resources more efficiently. It can also help identify which teams need additional resources or training to help them become more efficient, such as using more automation.

Ultimately, cost per team can be an indicator of the efficiency of your engineering team. It can help you identify projects that are running over budget so you can make early changes to ensure they are completed on time and on budget.

Cost per deployment

Cost per deployment is particularly useful for engineers who want to track how changes to their code or deploying a new feature affect their costs. 

We’ve used this metric at CloudZero to identify which engineering decisions aren’t as cost-efficient as we would like and then tweaked those to build more cost-efficient solutions right from the beginning.

Cost per project

Finance will love this one. It is an indicator of how much a project costs to complete. 

Tracking cost per project can also help you be aware of costs to expect in similar projects in the future, enabling you to give better estimates and quotes to potential clients –and then follow through without surprise budget increases. 

You can also use it to make decisions about where to allocate resources and which projects to prioritize. It can also be used to measure the success of a project. 

We get it.

Most contemporary cost tools do not offer this level of cloud cost intelligence.

In AWS Cost Explorer, for example, you can get resource-level reports — even if for a few days at a time.

A more robust platform like CloudZero offers over a dozen unit cost metrics by default — the first platform to do so across all cost areas. CloudZero also lets you set up custom metrics to organize cloud cost insights however makes sense for your business.

Step 7: Accurately control, forecast, and manage cloud spend

Now is the time to take what’s working for your business to the next level. At this stage of your FinOps maturity journey, you will likely encounter challenges that can impact your cost visibility, forecasting, and cost management.

Here are a few challenges you might face. We also offer tips for overcoming them with confidence.

Complexity

Cloud providers offer a wide range of pricing options, making it tricky to manage and predict costs.

Solution: Pick and stick to specific cost metrics, only changing them when necessary. Choose a robust cloud platform to get comprehensive cost insights and eliminate blindspots without needing to manage multiple cost management tools.

Resources:

Variable usage patterns

Cloud usage can fluctuate dramatically, influenced by business cycles, new projects, or unforeseen events, leading to unpredictable costs.

Solution: Once you have your unit costs and Cost of Goods Sold (COGS) down, you won’t have to worry about fluctuations, since you’ll be aware of why your costs are changing (what and who is responsible), showing you precisely where to optimize.

Resources:

Misaligned cloud cost management

With different teams or departments managing their own cloud resources individually, gaining a common view of overall cloud spend becomes challenging.

Solution: Align your teams to optimize costs using a common understanding as we suggested earlier. Each team can have its own metrics, such as cost per deployment for engineers. But all teams will want to use the common unit economics approach to optimize their spending.

Resources:

Rapid technological changes

The constant evolution of cloud technologies means that what’s cost-effective today might not be tomorrow.

Solution: Use your unit economics to understand your spend and tweak your resource allocation to prioritize the most profitable areas. Keeping efficiency high is key to maintaining a competitive advantage in any business environment.

Resources:

Ineffective cost allocation

Difficulty in correctly allocating cloud costs to the right projects or departments can lead to inaccuracies in spend tracking and forecasting.

Solution: Instead of relying on tags, embrace telemetry-driven cost allocation. This can help you account for the costs of tagged, untagged, untaggable, and shared resources in multi-tenant environments. 

Resources:

Inadequate forecasting tools

Some organizations rely on outdated methods or basic estimations to predict cloud spend, which leads to inaccuracies and surprise bills.

Solution: Understanding the cost drivers influencing your costs is the key to forecasting accurately. For example, when you know how much each feature costs to build, deploy, and maintain, you can better estimate how increased usage will impact costs.

Resources:

Inadequate FinOps expertise

Understanding cloud billing nuances requires a specific skill set that not all organizations possess internally.

Solution: Embrace cloud cost intelligence, and seek a platform whose package includes a designated FinOps expert. The approach empowers your team to treat cost as a first-class metric, and also gives you an extension of your team who is experienced in FinOps and constantly monitoring your environment. With granular, immediately actionable insights on costs, you can successfully build a cost-aware culture (FinOps at work).

Resources:

Unexpected growth or scaling

Sudden increases in demand or rapid scaling can lead to unanticipated spikes in cloud spend.

Solution: Verify with your unit economics. It is easier to tell if the changes in your costs are caused by growth or by overspending when you already have an understanding of the people, products, and processes that drive them. In any case, you can still pinpoint where and why the increases are happening so that you can fix them and prevent budget overruns. 

Resources:

By addressing these challenges with effective strategies, partners, and tools, your organization can gain better control over its cloud spend, leading to more accurate forecasting and efficient budget management.

We’ve covered the strategy part. In the next section, we look into the tools and partnerships you can use to automate cost optimization and maximize your cloud ROI.

Choosing the right FinOps partner is a pivotal decision in your cloud cost optimization journey. It’s like deciding on a co-pilot for a cross-country flight; you need someone reliable, knowledgeable, and aligned with your destination.

To achieve this, you’ll want to complete three actions:

Let’s break each down into actionable insights.

Step 8: Choose the right cloud cost optimization platform

So, why not just build a cloud cost optimization tool in-house?

Build vs. buy

The decision to build versus buy software is often a source of conflict for companies. The rule of thumb is to build solutions that differentiate your business and buy everything else. Yet, the decision is not always straightforward.

Factors like opportunity cost, complexity, technical debt, and long-term implications such as ongoing maintenance and updates often weigh heavily on the decision. 

Think of it as deciding between custom-building a car or purchasing a tested, feature-rich model. Building software in-house can lead to flexibility but may result in technical debt and lock-in, while buying off-the-shelf solutions can offer immediate benefits but offer limited customization.

Ultimately, the decision hinges on factors like budget, specific needs, quality concerns, and internal teams’ expertise. Are you torn between building or buying?

Start here: 

Whichever method you deem best for your organization, you will need to decide what capabilities you want. We have already touched on some of them in this guide. 

In this section, we’ll offer a snackable guide to choosing the right cloud cost platform (or the capabilities you’ll want to build into your own solution). 

Look for the following capabilities when evaluating cloud cost platforms

Any cost tool worth its salt should have all of the following (no exceptions):

Tools that cannot handle all of the above may not be the best you can have, regardless of your size.

Resources: 

Step 9: Team up with the best FinOps partner

Some of the best aspects of a good FinOps partner include:

1. Expertise in cloud platforms: A good FinOps partner should have experience with the specific cloud platforms your organization uses, such as AWS, Azure, or Google Cloud. 

2. Cross-functional collaboration: A Certified FinOps Certified Practitioner should be able to work effectively with your finance, product, and engineering teams, ensuring alignment and collaboration.

3. Trust-building behavior: They should practice behaviors that build trust, such as sharing data, owning decisions, and being transparent about what they can do better.

4. Cost visibility: The FinOps partner should help improve cloud visibility across the organization, ensuring that teams understand how their activities impact cloud costs.

5. Cost optimization strategies: You’ll want a partner who provides working strategies for cost optimization, such as using Cost per Customer data to set profitable service prices and optimizing your tagging policy.

6. Financial accountability: The partner should help bring financial accountability to the variable spend model of the cloud, enabling your distributed teams to make sound business trade-offs between speed, cost, and quality on the fly.

7. Cost intelligence platforms: You’ll want a FinOps partner who is familiar with cost intelligence platforms like CloudZero to help you build a FinOps program that actually works for you, not against you.

With a good FinOps partner by your side, you can streamline your financial operations in the cloud without cutting corners or burning yourself.

How CloudZero measures up

CloudZero has the right mix of FinOps partnership and comprehensive cloud cost optimization tooling. So, if you are sick of trying to find a platform that gets you, using CloudZero will be a breath of fresh air. 

CloudZero delivers:

Want to see why engineering, finance, and SaaS leaders love CloudZero?

Start here:

Step 10: Practice continuous improvement best practices

After the Run stage, the next step is often to “Fly.” Here is what flying in FinOps looks like.

Cloud cost optimization. And yes, it is an ongoing process, not a one-time event. Continuous improvement in this area involves regularly evaluating and adapting your strategies to ensure they align with your evolving business needs and technological advancements.

Here’s a practical guide to encourage this approach in your organization:

Follow these steps and your organization will revel in a culture of continuous improvement for optimizing cloud costs.

Conclusion

As we wrap up this guide, let’s revisit the key takeaways and insights we’ve gathered on our FinOps maturity journey.

We started by understanding the basics of cost management in cloud computing. From the initial steps of traditional cost savings methods, we gradually moved towards more advanced strategies like cloud cost intelligence and engineering-led optimization.

Although traditional approaches provide a solid foundation, they often lack the depth and flexibility needed for dynamic cloud environments. They often fall short in precision, over-rely on tagging, and miss engineering’s crucial role in cost optimization.

By contrast, next-generation approaches offer a holistic view, involving cross-functional teams in a more proactive, data-driven strategy. The ultimate goal here is not just to reduce costs, but to maximize value — getting the most out of every cloud resource for your organization.

By embracing the principles and practices outlined in this guide, your business can do more than reduce unnecessary costs. It can also thrive in a culture of efficiency, accountability, and operational excellence (a major competitive advantage today).

One more thing. The journey to cloud cost maturity is ongoing. It’s about constantly learning, adapting, and optimizing. With the right mindset, tools, and partner, you can turn cloud costs from a challenge into an opportunity for sustained growth and innovation. 

To your success!