
Cloud Costs
I kept thinking “we have heard this cost visibility, cloud tagging and attribution story one too many times.” For me, the game changing moment was when Aran began talking about reducing risk, proactive planning, and creating a secondary marketplace.
There is a counterintuitive truth at the center of cloud financial management: the faster you grow, the more dangerous cloud cost unpredictability becomes.
During hypergrowth, everything accelerates; customer acquisition, infrastructure scaling, product development cycles. The instinct is to prioritize flexibility over financial discipline. Lock in as little as possible. Keep optionality. Grow first, optimize later. It's an understandable instinct, but it is also a gross margin trap.
SaaS companies with gross margins above 80% command a 38% valuation premium over peers below that threshold. Cloud infrastructure is often the single largest line item in cost of goods sold for SaaS businesses. When cloud spend scales linearly with revenue; or worse, faster, it compresses the gross margins that investors, acquirers, and CFOs use as a proxy for operational discipline. A well-designed commitment strategy is one of the most effective ways to protect and improve those margins while you are scaling hardest.
Archera helps hypergrowth companies build commitment strategies that scale with them. Book a demo →
Growing companies face a specific version of the cloud cost problem. On-demand pricing seems like the right choice because it eliminates commitment risk. You only pay for what you use, and you are never locked in. But on-demand pricing is also the most expensive way to run cloud infrastructure, and the "flexibility premium" compounds rapidly at scale.
A company spending $500,000 per month on AWS compute on-demand pricing could be paying $300,000–$400,000 with a well-managed commitment portfolio, a difference that flows directly into gross margin. Over a 12-month fundraising runway, that is $1.2M–$2.4M of additional gross profit. For a company preparing for a Series B or C, that is a materially different financial story.
The second dimension of the trap is unpredictability. A mid-stage SaaS company spending $1.2M annually on cloud infrastructure can see bills swing 15–20% month over month due to variable usage patterns. That variance makes forecasting unreliable, clouds burn rate analysis, and forces CFOs to carry excess cash reserves as a buffer against billing surprises. Commitment-based pricing eliminates that variance, replacing a moving target with a predictable cost structure that finance can model with confidence.
When we talk about cloud commitments protecting margins during hypergrowth, we mean three distinct things:
Lowering your blended compute rate. Reserved Instances and Savings Plans reduce the effective per-hour cost of compute by 40–72% compared to on-demand. This directly reduces COGS, improving gross margin without any change to your product or customer experience. At scale, a 3–5% improvement in effective blended compute rate translates to a significant and compounding margin improvement.
Converting variable cost to predictable cost. Commitments replace the most volatile component of your cloud bill, on-demand compute, with a predictable, fixed-rate structure. This predictability has financial value beyond the dollar savings: it enables accurate forecasting, simplifies investor conversations, and reduces the cash buffer required to absorb billing surprises.
Building a defensible unit economics story. Investors at every stage evaluate the trajectory of cost per unit of revenue; cost per customer, cost per API call, and cost per active user. A commitment strategy that reduces infrastructure costs as a percentage of revenue strengthens the unit economics story that matters most during fundraising.
The objection we hear most often from hypergrowth companies is simple: "What if our infrastructure needs change? We don't want to lock in and get stuck."
It's a legitimate concern. Hypergrowth inherently involves architectural uncertainty. You might pivot to containers. You might migrate regions. You might add GPU workloads as AI features ship. A commitment that made sense at the start of the year might not align with your infrastructure six months later.
This is exactly why the design of your commitment program matters as much as the size of it. A few principles:
Commit to your proven baseline, not your projected peak. Model your minimum consistent usage, the floor of your compute consumption over the past 60 days, and commit to 70–80% of that. The rest stays on-demand, giving you room to grow and change without stranding capacity.
Use Compute Savings Plans rather than rigid RIs as your primary instrument. Compute Savings Plans apply automatically across EC2, Lambda, and Fargate, regardless of instance family, size, or region. If you migrate from m5 to m6i, or shift workloads from EC2 to Fargate, your discount follows. That flexibility is worth the slightly lower discount depth compared to EC2 Instance plans.
Prefer 1-year terms during periods of high uncertainty. The additional discount from 3-year terms (typically 5–10 percentage points more savings) is valuable, but not worth the risk exposure before you have a clear view of your 3-year architecture. Graduate to longer terms as workloads stabilize.
Archera was built specifically to solve the commitment risk problem that holds hypergrowth companies back. Through our Guaranteed Commitments structure, we provide protection against unused capacity, meaning you can commit to the level that maximizes your savings without carrying the full financial exposure of a stranded commitment.
This changes the calculus entirely. Instead of asking "how much can we commit to safely?" you can ask "what commitment level optimizes our gross margin?", and then commit to that level, knowing that downside exposure is managed.
For hypergrowth organizations approaching a fundraise, preparing for a Series C, or building the unit economics story that will define your next chapter, a commitment program built on Archera is one of the highest-ROI infrastructure investments you can make.
The SaaS companies that win on gross margin during hypergrowth share a few common characteristics. They treat cloud commitment management as a financial discipline, not an infrastructure afterthought. They build commitment programs early, before the spend levels make manual management untenable. They optimize for predictability as well as savings. And they use the right tools to manage the complexity of continuous monitoring and portfolio optimization.
High-performing SaaS companies target gross margins of 75% or higher, with leaders approaching 80–90%. The gap between the median and the best performers is often explained not by product differences but by cost discipline, including cloud financial management. A commitment strategy is not just an infrastructure decision. It is a gross margin decision.
Talk to an Archera expert about protecting your margins during your next growth phase. Book a demo →