Latest / The Cloud Business Podcast with Fexingo: AWS, Azure, GCP, and Enterprise Infrastructure / The Cloud Broker Role That Enterprises Need Now
Transcript
- Lucas: So there's this role emerging in enterprise IT that didn't really exist five years ago — the cloud broker. Not a person who brokers cloud capacity like a commodities trader, but a dedicated intermediary — sometimes internal, sometimes a third-party platform — that helps companies navigate the mess of multi-cloud pricing, performance, and contract terms. Luna: I've seen that title pop up on LinkedIn more and more. Cloud broker, cloud procurement manager — it sounds like a real shift from the days when IT just picked one provider and stuck with it. Lucas: Exactly. And the catalyst is obvious: enterprise cloud spend has ballooned. Gartner projected global cloud spending to hit $675 billion in 2025, and with that kind of money on the table, companies realised they need someone whose job is to optimise across AWS, Azure, and Google Cloud — not just manage a single account. Luna: Right. So the broker role is partly about cost, but partly about leverage — especially when renegotiating contracts. Lucas: That's a huge part of it. Take a company spending $50 million a year across three clouds. Each provider gives them a discount tier based on committed spend, but those commitments are siloed. A broker can say, 'We'll shift 20% of our Azure workload to AWS if you don't match the price.' That kind of cross-cloud leverage didn't exist when everyone was single-cloud. Luna: And the brokers have data to back that up. They see the market — they know what Google is offering new customers, what Azure's enterprise agreements actually look like after negotiation. Lucas: Right. The transparency shift is real. Companies like CloudHealth, which VMware acquired, and Spot by NetApp started as cost-monitoring dashboards. Now they're evolving into advisory platforms that benchmark your spend against anonymised peer data. They can tell a CFO: 'Your AWS bill is 18% higher than comparable companies in your industry.' That's powerful. Luna: And it changes the conversation from 'we need to cut costs' to 'we need a better strategy.' Lucas: Exactly. But I think the most interesting angle is how cloud brokers are reshaping the procurement cycle itself. Historically, an enterprise would sign a three-year commitment with AWS, get a discount, and then forget about it. Now, brokers are recommending shorter commitment terms — one-year, even month to month for certain workloads — because the pricing landscape is shifting so fast. Luna: Especially with AI workloads. GPU pricing has been volatile — you don't want to lock in a high rate for NVIDIA instances only to see prices drop when new capacity comes online. Lucas: Exactly the case. And that's where brokers earn their keep. They model different scenarios: what if we commitment 50% of our GPU demand but keep 50% flexible? What if we use spot instances for training and reserved for inference? They're doing the math that most IT teams don't have the bandwidth for. Luna: So is this role mostly about cost, or does it extend into technical architecture decisions? Lucas: It's getting more technical. The best brokers are now involved in workload placement — not just which cloud, but which region, which instance type, which storage tier. They're using tools that monitor performance and cost in real time and automatically shift workloads. That's where the value really compounds. Luna: And that raises a question: if the broker is automated — a platform that moves workloads — do you even need a human in that role? Lucas: I think you still do, because the strategic decisions — which cloud to bet on for the next three years, how to structure an enterprise agreement — those are judgment calls. The platform gives you data, but the broker interprets it in context of your business. Plus, the human relationship with the cloud providers matters. If you have a dedicated broker who knows the sales rep, they can often get better terms. Luna: So it's a hybrid role: data-driven but relationship-heavy. That's interesting. Lucas: And it's growing fast. Gartner predicted that by 2028, 60% of large enterprises will use a cloud broker — either internal or external. That's up from maybe 20% today. The market for cloud brokerage services is estimated at around $4 billion now, and it's growing at over 25% a year. Luna: Let's talk about the internal vs. external question. Some companies build their own broker team. Some hire a consultancy. Which works better? Lucas: It depends on scale. If you're spending over $100 million annually on cloud, you probably want an internal team — they know your architecture, your compliance needs, your culture. But for companies in the $10 to $50 million range, an external broker often makes more sense because they bring aggregated data from many clients. Luna: And smaller companies probably can't afford a dedicated broker salary anyway. Lucas: Right. A good cloud broker with experience can command $150,000 to $200,000 a year. But a third-party service might charge 5-10% of the savings they generate. If they cut your bill by 20%, that's a no-brainer. Luna: I want to push back a little on the savings claim. Are these savings real, or are they just renegotiating discounts that were already available? Lucas: It's a fair question. The savings are real in many cases because most enterprises are leaving money on the table — they're not using reserved instances efficiently, they're not taking advantage of committed use discounts, they're not auditing their bills for errors. A broker finds those leaks. I've seen cases where a broker saved a company 30% in the first year, and then another 10-15% in subsequent years through ongoing optimisation. Luna: But there's also the risk that brokers push customers toward their own affiliated services. Some brokers are owned by cloud providers or have kickback arrangements. Lucas: That's a real concern. Transparency is key. A good broker will disclose any relationships. The best ones are independent and charge a flat fee or a percentage of savings — never a commission from a cloud provider. You want alignment of incentives. Luna: So how do you vet a broker? What questions should an enterprise ask? Lucas: First, ask for references from clients with similar spend and workload mix. Second, ask how they handle data sovereignty and compliance — if you're in healthcare or finance, that's critical. Third, ask about their technology stack — do they use their own tools or third-party platforms? Fourth, demand a clear fee structure. And fifth, ask for a sample audit of your current bill — a good broker can find savings in the first hour. Luna: That's practical. I also wonder about the future of this role. As ai driven optimisation tools get better, will the human broker become obsolete? Lucas: I think the tools will get smarter, but the human element — negotiation, relationship management, strategic trade-offs — will remain. Cloud is too complex and too dynamic for a fully automated approach, especially at the enterprise level. But the role will evolve. The broker of 2030 might be more of a strategist and less of a tactician. Luna: So if you're an IT professional looking to move into this space, what skills do you need? Lucas: You need a mix: deep knowledge of at least two major cloud providers, financial analysis skills — understanding unit economics, amortisation, discount structures — and strong communication. You're often the translator between finance and engineering. Plus, you need to stay current because pricing changes constantly. AWS alone has over 200 price changes a year. Luna: And a bit of sales acumen, because you're convincing teams to change how they work. Lucas: Absolutely. It's a fascinating career path. And if these conversations are useful for what you're building or running, listener support is what keeps this podcast ad-free and independent. You can find us at buy me a coffee dot com slash fexingo. Just a simple way to say the show matters to you. We appreciate it. Luna: Yeah, it really helps us keep doing deep dives like this. So back to the broker role — one trend I'm watching is the rise of 'cloud brokerage as a service' startups. Companies like CloudBolt and Densify are building platforms that combine FinOps with automated workload management. Lucas: And we're also seeing the hyperscalers respond. AWS launched a 'cloud advisory' service last year, Azure has its 'landing zone' consulting — they're essentially trying to become your broker. But there's an inherent conflict of interest: they want you to spend more on their cloud. An independent broker doesn't have that bias. Luna: So the value of independence is central. And that's probably why the role is growing — because enterprises need someone who's purely on their side. Lucas: Exactly. Cloud is too big, too complex, and too expensive to manage without dedicated expertise. The broker role is going from nice to have to necessity. I'd bet that in five years, most companies spending over $20 million on cloud will have someone with 'broker' in their title. Luna: And if they don't, they'll be paying more than they should. Lucas: That's the bottom line. Thanks for listening to The Cloud Business Podcast with Fexingo. We'll be back next week with another deep dive.