Bulk Wind Turbine Procurement: A 7-Step Checklist (Specs, OEM vs. Private Label, and the Costs Nobody Quotes)

Who this checklist is for

I've been the person who signs off on turbine procurement for a roughly 400-person renewables developer. Six years, about $240M a year across turbines, balance-of-plant, and O&M — and something like 35–40 vendor bid cycles. Maybe 40. I'd have to pull the system, but it's in that range.

This checklist is for the buyer who's sitting in front of a wind turbine catalog with 40 pages of model numbers and three shortlisted suppliers, and needs to turn that into a defensible order. It works whether you're buying 2 units or 200. It is seven steps, and step 4 is the one most teams do last when they should do it first.

One thing upfront: nobody in this industry publishes list prices. If you find a per-MW number on a blog, it's a modeled estimate from someone's cost review, not a quote. Treat it accordingly.

Step 1: Classify the asset before you talk to a single supplier

The single biggest time-waster in turbine procurement is getting six weeks into a conversation before discovering the supplier's platform isn't certified for your site.

Before you open a catalog, lock three things:

• Onshore or offshore. These are different supply chains, different installation vessels, and different design standards. Onshore designs follow IEC 61400-1; offshore designs follow IEC 61400-3. A platform certified for one is not automatically eligible for the other.
• IEC wind class. Class I, II, III, or S, plus the turbulence intensity subcategory (A, B, C). Your site's reference wind speed and turbulence profile determine which class you need. Overspecifying costs you capex; underspecifying costs you a type certificate.
• Site constraints. Transport route limits (blade length vs. road curves), hub height options, grid connection capacity, and noise limits near dwellings.

Do this first and your shortlist drops from 40 pages to 4. That's not a small saving — it's two months of engineering time you don't spend evaluating products you can't buy.

Step 2: Decode the model numbers so your spec sheet is actually comparable

Wind turbine naming looks arbitrary until you know the convention. The common pattern is platform + rated power + rotor diameter. An SG 3.4-132, for example, is a 3.4 MW machine with a 132-meter rotor.

That matters because the number after the dash drives almost everything you care about: swept area, capacity factor, transport logistics, crane class, and foundation loads. Two turbines with the same MW rating and different rotor diameters are not the same product.

When you request specifications, ask for these fields in a spreadsheet, not a PDF brochure. Force the same format from every supplier:

• Rated power (MW) and IEC wind class
• Rotor diameter and swept area (m²)
• Hub height options
• Cut-in, rated, and cut-out wind speeds
• Power curve, measured per IEC 61400-12-1
• Sound power level (L_WA)
• Type certificate holder and certificate number
• Drivetrain architecture (geared / direct drive / medium speed)

Here's the trap: the brochure power curve and the site-specific warranted power curve are two different documents. The brochure is generic. The warranted curve is tied to your air density, your turbulence, and your contractual remedies. Make sure the second one is what ends up in the appendix.

Step 3: Get the certified document pack, not the marketing pack

For any utility-scale order, request the certification and performance documentation directly:

• Type certificate to IEC 61400-1 (onshore) or IEC 61400-3 (offshore)
• Power performance measurement report (IEC 61400-12-1)
• Load and safety factor documentation for your site class
• Grid code compliance evidence for your interconnection jurisdiction
• SCADA and communications interface documentation (IEC 61400-25 is the usual reference)

If a supplier can't produce these on request, that's your answer — and you've saved yourself a year.

Step 4: Negotiate the 20-year terms before you negotiate the turbine price

This is the step I've watched nearly every team skip, including mine on our first two orders.

The turbine price is maybe 40–50% of your installed cost. The service agreement, spare parts, and availability regime are what decide whether the asset is profitable in year 14. Buyers usually close the hardware deal first and then go ask for a service contract, at which point they have almost no leverage left.

Do it in this order instead:

1. Define availability. Availability guarantees mean nothing unless the definition matches IEC 61400-26-1. Ask: is it time-based or production-based? Is grid unavailability excluded? Is curtailment excluded? Get the formula in writing.
2. Lock the service scope and escalation. Full-scope or partial? Who supplies the crane? What's the response-time commitment, and is it measured from alarm or from dispatch?
3. Price the spare parts you'll actually need. Ask for a 10-year indicative parts list with pricing index — not a single lump sum.
4. Get an obsolescence clause. What happens when the control system goes end-of-life in year 12? Who pays for the retrofit?
5. Confirm data ownership. Who owns the SCADA data and the operating history? If you ever repower or sell, that data is an asset.

We ran this backwards on an early order. Availability was defined as “contractual availability” with no formula attached. When we had a dispute in year 3, we discovered the vendor's definition excluded scheduled maintenance windows entirely — which, honestly, made the guarantee close to meaningless.

Step 5: OEM vs. private label — know what actually changes

“Private label” or white-label supply in wind usually means one of two things: a licensed manufacturing arrangement where a regional partner builds and brands an existing platform, or a brokerage/repackaging arrangement where the seller is not the original equipment manufacturer.

The hardware may be identical. Six other things usually are not:

• Who holds the type certificate. If it's not your counterparty, your compliance path runs through a third party.
• Who backs the warranty. A warranty from a 12-person trading company is not the same instrument as one from a balance sheet with a global service footprint — regardless of who built the nacelle.
• Who services it in year 8. Parts availability, firmware updates, and technical support scale with installed base.
• Software and SCADA access. Some white-label routes restrict your ability to integrate third-party condition monitoring.
• Upgrade path. Can the platform be repowered or upgraded with the OEM's own kits later?
• Residual value. At asset sale, buyers ask who the OEM is. A non-OEM pedigree is a discount in some markets and a non-issue in others — know which market you're in.

I'm not telling you one route is wrong. For small fleets under 20 MW with short hold periods, the private-label route can be genuinely cheaper on capex. For a 200 MW portfolio you intend to hold 25 years, the service and parts infrastructure question usually settles it. Just make the decision deliberately rather than discovering it in year 6.

Step 6: Build the TCO model before you open the bids

Build the model first. Every bid then lands in the same structure, and you stop comparing apples to ex-works oranges.

Include:

• Delivered turbine cost — and specify the Incoterm. Under Incoterms 2020 (ICC), Ex Works, FCA, and DAP put very different logistics and risk obligations on you. Comparing an Ex Works quote to a DAP quote is not a comparison.
• Foundation, BoP, and installation. Crane class, crane availability, and ground conditions drive this more than turbine MW does.
• Grid connection and any network upgrade costs.
• Opex over 20–25 years — service, insurance, land lease, and expected major component replacements.
• Production risk. Net AEP with an uncertainty band, not a single P50 number. A cheaper turbine with 2% lower net AEP can be the more expensive asset.
• Indexation and FX. Steel, copper, and freight clauses. If a supplier insists on an unrestricted steel index, you've bought a commodity position you didn't want.

We once saved about $85,000 by taking a non-OEM gearbox overhaul quote instead of the OEM route.

The overhaul failed in month 4. Replacement power, crane mobilization, and the redo came to roughly $280,000 — call it $300,000 with the crane overrun. Net loss on that “saving”: somewhere near $200,000, plus a very uncomfortable conversation with our asset manager.

Step 7: Verify the counterparty, the references, and the exit

Three items before signature:

1. Check the legal entity. Corporate rebranding and restructuring happen constantly in this sector. The name on the brochure and the name on your warranty may not be the same entity. Confirm who is actually signing, who is guaranteeing performance, and whether a parent guarantee is available.
2. Call references with the same site class. Not the same country — the same IEC class and similar turbulence. A turbine that performs on a Class III coastal site tells you nothing about a Class I inland ridge.
3. Read the exit clauses. Assignment rights if you sell the project. Decommissioning responsibility. Repowering rights. Removal obligations at end of life.

Common mistakes that cost real money

• Comparing Ex Works to delivered pricing. This alone can swing a bid by 8–12%.
• Accepting “standard warranty” without a schedule. Standard to whom? Ask for the excluded-components list. Every vendor has one.
• Assuming an availability guarantee is a production guarantee. It isn't. You can hit 97% availability and miss your AEP target if the wind isn't there — and that's your risk, not theirs.
• Letting the spec sheet drift. The most frustrating part of this job: you send the same specification table to three suppliers and get three different interpretations of the same sentence. Version-control your spec, and require written confirmation against the version number, not the date.
• Skipping the crane strategy. Crane availability is a schedule risk, not a cost line. If your installation window is 14 weeks, book the crane before you book the turbines.

Where to verify the technical references: IEC 61400 series (iec.ch) for design, power performance, communications, and availability definitions; the ICC Incoterms 2020 publication for delivery terms; and for US market cost context, Lawrence Berkeley National Laboratory's annual Land-Based Wind Market Report and NREL's Cost of Wind Energy Review. Numbers in those reports are national averages from survey and modeling work — useful for sanity-checking your model, not for quoting your project.