Operations | Monitoring | ITSM | DevOps | Cloud

The inevitable happened. A while back, Gartner projected that in 2026, 30–50% of all new SaaS product features would use LLM inference. That meant OpenAI-style costs would become a standard part of SaaS COGS. Today, OpenAI has become one of the most operationally significant line items for SaaS companies. But for many teams, this creates an uncomfortable gap. Engineering sees OpenAI as a fast path to innovation.

In 2025, Oracle shocked the market. Its cloud growth was so aggressive that Oracle’s stock surged, briefly making founder Larry Ellison the world’s richest person. That didn’t happen by accident. Oracle closed fiscal 2025 with $57.4 billion in revenue, mainly driven by cloud services. Oracle Cloud Infrastructure (OCI) grew roughly 50% year over year, driven by enterprise databases, AI workloads, and network-intensive applications migrating from more expensive platforms.

Right now, 48% of organizations say they’re being asked to measure or report on AI-related costs. The problem is that they’re still figuring out how to do it. That was a very telling stat from a recent CloudZero webinar on AI and profitability, and speaks loudly to the reality that many organizations are still struggling to get a grasp on AI spend which our data shows to be rising sharply as a part of total spend in recent months.

Think of a stateful application like a conversation with a barista who remembers your order every time you walk in. They know what you had yesterday, how you like it prepared, and what you’ll probably want next. That memory makes the experience smoother, but it also means that if that barista isn’t around, your experience can break down entirely. A stateless application, on the other hand, is similar to ordering from a self-service kiosk.

On paper, Gemini pricing looks straightforward. You pay per token. Input tokens cost one amount, output tokens cost another, and different models come with different rates. But once Gemini is wired into a production SaaS product, that simplicity disappears. Fast. That’s because token usage compounds across context, retrieval, and output — not across requests. The same “API call” can cost pennies in one feature and dollars in another.

When CloudZero CTO Erik Peterson joined the SourceForge podcast in January 2026, he didn’t just talk about cloud costs. He reframed them as a launchpad for innovation, survival, and competitive advantage. Whether he was describing the “trough of lost innovation,” the “freemium tax,” or why efficiency is the next frontier of engineering culture, Erik’s expert insights go beyond FinOps hygiene.

Consider this: traditional cloud cost monitoring was like checking your fuel gauge once a month — after the trip was already over. That model worked when infrastructure scaled slowly. You provisioned resources predictably and paid for stable, linear usage. AI breaks that model. Today, AI costs behave like a high-performance engine with a hypersensitive throttle. A small input, like a prompt change or a single power user, can dramatically increase your fuel burn in seconds.

With all we’ve seen from AI in the last several years, it can be easy to forget that it’s still in its very early days. As torrid as its evolution has been thus far, it will only intensify. As SVP of Engineering at a B2B SaaS company, I’ve had a front-row seat for much of this evolution. Here are three ways I see AI heading in 2026.

Choosing between AWS and OCI is a common decision for organizations moving workloads to the cloud. Both Amazon Web Services and Oracle Cloud Infrastructure offer global infrastructure, robust security, and broad service portfolios. On paper, the platforms can look interchangeable. They are not. AWS and Oracle Cloud differ in pricing, compute models, storage options, networking, and managed services. These differences affect scalability, reliability, and day-to-day operations.

Welcome to January’s Cloud Economics Pulse, CloudZero’s monthly look at cloud spend as AI moves from vibe to prod. And this related news flash — AI spend keeps hitting new highs. pilots to production. In last month’s Pulse, we explored the compounding effect of AI becoming part of everyday cloud operations. This month, we see that pattern harden into year-end results.

API metrics have long been a core part of building and operating reliable SaaS products. Teams track the likes of request volume, latency, and uptime to ensure APIs perform as expected under load. First: API cost intelligence metrics measure how API usage translates into cloud, AI, and third-party spend — and attribute that cost to customers, features, workflows, and teams so SaaS businesses can protect margins as usage scales. But today, the API metrics that matter most go beyond performance.

OpenAI doesn’t bill per feature, per customer, or per transaction. It bills per token, across multiple models, with usage patterns that can change by the hour. As a result, two API calls that support the same feature can have very different costs. Without a clear way to translate token-level pricing into something product, engineering, and finance teams can reason about, AI spend becomes difficult to forecast and harder to control.

FinOps is evolving fast, and 2026 is shaping up to be a big year for specialization. While these certifications are ranked from beginner to advanced to help you build skills in the right order, one course stands out as the hottest recommendation right now: FinOps for AI. AI spend is accelerating, ownership is getting murky, and teams are scrambling to keep up. That urgency is exactly why FinOps for AI is generating so much interest heading into 2026.

In the traditional finance world, budget variance was a static comparison between actual and budgeted spend. But in the cloud era, where costs scale with usage, experimentation, and engineering decisions, variance tells a much richer story. Done right, budget variance helps you distinguish between healthy growth and margin erosion. It can signal strong feature adoption, rising customer demand, or successful launches. It can also reveal waste, inefficiencies, and weak cost controls.

Why does AWS API Gateway spend rise even when backend infrastructure stays the same? For most teams, the answer isn’t compute. API Gateway pricing is driven by how APIs are used — request volume, retry behavior, traffic patterns, and growth over time — not by provisioned resources. Because AWS reports these costs as aggregated usage totals, it’s often unclear which APIs, environments, or behaviors are responsible for increases.

Think about how airlines assign seats on a plane. Some have extra legroom. Some sit near exits. Some are cheaper, while others cost a premium. Certain passengers also have strict requirements, like families traveling together or travelers who paid for a specific class. Now imagine boarding everyone randomly. A passenger who paid for extra legroom (perhaps for health reasons) ends up squeezed into a middle seat. Families scatter across the cabin. Premium seats sit half empty while the back rows overflow.

Happy New Year, everyone! 2025 was another exciting year filled with impressive AI advancements. As you might expect, some significant cost changes accompanied these new developments. Because reflecting on the past is one of the best ways to prepare for an even stronger future, here’s your end-of-2025 headline round-up, complete with what you can expect going forward into 2026: Get caught up on the details below.

Modern SaaS companies aren’t reporting weaker margins because they forgot to rightsize instances or buy reservations. It’s more because cloud spend now moves at the speed of AI experiments, overnight shifts in customer usage, and automated systems that scale in seconds. That’s why the next generation of cloud cost optimization strategies looks fundamentally different from what worked even two years ago.

“Cut spending!” “Slash costs!” “Stick to the budget!” Poke your head into almost any finance meeting in a SaaS company and you’re likely to hear one or more of the above phrases played on repeat. At first glance it makes sense: Costs are increasing, so we should reduce them. I’d like to challenge that narrative. Mature companies don’t care about cloud costs.