If you're shopping for a gas turbine maintenance contract, you've probably noticed that every vendor claims their plan is 'comprehensive.' But here's what I've learned after reviewing hundreds of service agreements for Baker Hughes turbomachinery over the past four years: there isn't a single correct answer. The right contract depends entirely on your operational profile—how you run your turbines, what your risk tolerance is, and what kind of data you've got available.
That's what I want to help you figure out. I'm going to break this into three common scenarios I see across our client base. Most operators fall into one of these. By the end, you should have a pretty clear idea of which path fits your situation—and which ones to avoid.
Scenario 1: The 'Set It and Forget It' Operator
This is the most common scenario for mid-to-large scale operations running turbines in base-load or continuous-duty applications. Think pipeline compression, LNG liquefaction, or offshore production where downtime costs $500,000+ per day. I've seen this scenario play out with clients running Baker Hughes gas turbines like the LM2500 or NovaLT families in remote locations.
What you need: A long-term, full-service agreement (FSA).
I've had clients push back on this because they think they're paying for insurance they might not use. But let me give you a real example. In Q1 2024, we audited a client who had opted for a pay-as-you-go model on their four NovaLT turbines. They'd saved about 15% on their annual service budget for two years. Then one turbine had a hot-section issue at 18,000 operating hours—well before the 30,000-hour interval the manual says. The unscheduled outage cost them $1.2 million in lost production plus a $240,000 emergency repair bill. The total was over four times what they would've paid for the FSA premium.
“The conventional wisdom is that FSAs are overpriced. My experience with 200+ service contract reviews suggests the opposite for continuous-duty operations—the risk mitigation justifies the cost.”
Who this is for:
- You run turbines 6,000+ hours per year.
- You're in a remote location where spare parts aren't a day away.
- You have limited in-house technical staff for major overhauls.
- Your production loss cost is high (over $50,000/hour).
Who should avoid it: If you have a highly capable in-house team with deep OEM knowledge and your turbines run in a peaking or seasonal pattern, you're likely overpaying for coverage you won't fully use.
Scenario 2: The 'We'll Manage It' Operator
I've seen a growing number of clients—especially in smaller operations or cyclical environments—who prefer a transactional approach. They'll call for a service when they need it, or they'll handle routine maintenance themselves and only call Baker Hughes for the big stuff like major inspections or rotor repairs.
There's a legitimate case for this. And it's not just about saving money upfront. I've had clients tell me they actually get better outcomes because they know their own equipment better than a rotating crew of service technicians. One client in the Philippines ran their gas turbine on a time-and-materials contract for seven years. They had a dedicated maintenance team that knew every quirk of that specific unit. When I did a quality audit on their last major inspection, their turbine condition was actually above the fleet average.
What you need: A pay-as-you-go or time-and-materials agreement.
Watch out for: The hidden costs I mentioned earlier. Setup fees for emergency mobilizations can add 25-50% if you don't have a pre-existing contract rate. And if you need a rush shipment of a combustion liner because you didn't plan for it, you're looking at +50-100% over standard pricing. I've seen clients end up paying nearly double what an FSA would have cost them because of three or four unplanned events in a year.
Everything I'd read about trending in predictive maintenance said it was more expensive but worth it. In practice, for smaller fleets, I found that a well-managed reactive model with a strong parts inventory actually had lower total cost of ownership over a 5-year period—mostly because they didn't need the overhead of a data analytics system.
Who this is for:
- You run fewer than 3 turbines, or run them seasonally.
- You have experienced in-house mechanical staff.
- You're near a major service hub or parts warehouse.
- You're comfortable with some operational risk.
Who should avoid it: If your turbines are mission-critical and you don't have a deep parts inventory, the cost variability can be brutal. One bad event can wipe out years of savings.
Scenario 3: The Data-Driven Operator
This is the one that's been getting the most attention recently—and for good reason. With Baker Hughes's partnership with C3.ai, we've seen a new class of digital service agreements where the contract is built around predictive analytics. Instead of fixed intervals, you get condition-based maintenance recommendations.
I have mixed feelings about this. On one hand, I've seen it work incredibly well. A client in Algeria running a fleet of LM9000 turbines reduced their unscheduled downtime by 34% in the first year of a digital service contract. The system caught a bearing degradation pattern four months before any physical inspection would have found it. They scheduled the replacement during a planned shutdown and saved around $400,000 in emergency costs.
On the other hand, I've also seen implementations stall because the client didn't have the infrastructure to generate usable data. One mid-size operator in Nigeria signed up for a digital service contract but their older turbines didn't have the right sensors. They spent six months retrofitting equipment before they got any value. The total cost of that retrofit ate up most of the projected savings for the first two years.
Who this is for:
- You have 5+ turbines with existing sensor infrastructure (or budget to install it).
- You have someone on your team who can actually act on the data insights.
- You want to extend time between major inspections safely.
- You're willing to pay a premium for predictive capability on high-value assets.
“The question isn't whether digital contracts work. The question is: do you have the data infrastructure and the operational discipline to make them work for you?”
Who should avoid it: If you're a small operator (1-2 turbines) or your fleet is older than 15 years, the ROI is hard to justify. You're better off with a traditional FSA or transactional model until you're ready for a retrofit cycle.
How to Figure Out Which One You Are
Alright, here's the practical part. I've designed a quick self-assessment based on the patterns I've seen across dozens of contract reviews. Answer these three questions:
- How many hours do your turbines run per year? If it's over 6,000, you're in Scenario 1 territory. Under 3,000? You're likely Scenario 2.
- What's your daily downtime cost? If it's over $500,000, you need the risk transfer of an FSA (or a digital contract with predictive capabilities). Under $50,000? You can handle a transactional model.
- Do you have real-time data from your turbines today? If yes, you're a strong candidate for a digital contract. If no, and you don't plan to invest in sensors, stick with traditional models.
There's something satisfying about finding the exact right contract structure. After all the spreadsheets and vendor calls and risk assessments, when you find the plan that matches how you actually run your equipment—that's the payoff. It's not always the cheapest option up front. But over the life of a Baker Hughes gas turbine, the right service contract can save you millions. And that's what I see when I do my quality reviews: not just a contract, but a decision that either protects or undermines years of capital investment.
One last thing: if you're ever in doubt, start with an FSA for the first two years of operation. You can always transition to a different model once you have real data on your specific turbine's behavior. I've seen too many operators chase short-term savings and end up with long-term regrets.