Tidal power can be a business, but it is not yet a routine small-power-plant investment in the United States. A founder is usually building two businesses at once: a marine-energy technology company that must prove reliability, and a project-development company that must secure a site, permits, interconnection, an offtake agreement, construction capital, and long-term operating capability. The project can sell electricity, but electricity revenue alone rarely pays for an early demonstration device.
The opportunity is real. A National Renewable Energy Laboratory assessment published by the U.S. Department of Energy estimated the U.S. tidal technical resource at 220 terawatt-hours per year, equivalent in that study to about 5.4% of 2019 U.S. electricity generation and roughly 21 million homes. The same national marine-energy resource assessment emphasizes that technical resource is not the same as commercially recoverable output. Water depth, current speed, seabed conditions, navigation, environmental constraints, cable distance, grid capacity, and local power value determine what can actually be financed.
220 TWh/year
Estimated U.S. tidal technical resource in the DOE/NREL assessment. A developer should treat this as a market ceiling, not a sales forecast. The investable resource is the portion that survives site screening, permitting, engineering, interconnection, and price tests.
The most credible near-term revenue models are therefore broader than a standard wholesale power sale. They include grid-connected demonstration contracts, premium-priced power for remote or fuel-dependent communities, utility partnerships, research awards, equipment sales, licensing, engineering services, and eventually ownership of multi-device arrays. DOE’s Cook Inlet work illustrates why location matters: its analysis found that today’s Alaska Railbelt grid could support up to 200 MW of tidal generation and potentially more after upgrades, while the resource itself is much larger. The Cook Inlet grid study also highlights tidal predictability, which can carry value beyond a simple energy price.
How Much Capital Does a Tidal Power Project Require?
The first budgeting decision is scale. A pre-revenue technology company may spend $3M-$12M on design, tank testing, component trials, controls, intellectual property, and a small open-water prototype. A grid-connected device changes the order of magnitude because the budget must include a certified turbine, foundation or mooring, subsea export cable, installation vessels, interconnection equipment, environmental work, insurance, and contingency. A first 1 MW pilot can plausibly require $19M-$49M under the planning assumptions below. This is not a published industry average; it is a founder-side model range for an emerging U.S. project.
Technology maturity matters as much as nameplate capacity. NREL’s marine-energy risk framework recommends formal design and risk reviews, technology qualification, risk registers, and failure-mode analysis through the development cycle. Those activities add cost early, but they are cheaper than discovering an unqualified drivetrain, cable termination, blade, bearing, or foundation after deployment. The NREL risk-management framework is especially useful when building the engineering work breakdown structure.
1 MW pilot cost category
Planning range
What moves the number
Resource, bathymetry, geotechnical, and site studies
$800,000-$2.0M
Survey seasons, current measurements, seabed complexity, metocean data quality
Detailed engineering, qualification, and certification support
Prototype risk, installation delay, spares, first retrieval, cash reserve
Total modeled investment
$19.0M-$49.0M
Before financing fees and any large transmission upgrade outside project scope
Illustrative share of a $30M pilot budget
The device is the largest line, but balance-of-system and marine work can consume more than half of total capital.
Turbine and powertrain38%
Foundation and cable21%
Installation and grid17%
Engineering and permitting12%
Contingency and reserve12%
What Does Monthly Operating Cost Look Like After Deployment?
Marine energy has no fuel bill, but “no fuel” does not mean low operating cost. The expensive event is often not the failed component; it is mobilizing a vessel, waiting for a safe current and weather window, retrieving the device, moving it to port, repairing it, redeploying it, and losing generation throughout the process. A good operating budget separates predictable monthly overhead from event-driven marine interventions.
Specialist labor is another major line. For context, the Bureau of Labor Statistics reported a 2024 median annual wage of $105,670 for marine engineers and naval architects, while electrical engineers were above $100,000. The BLS marine-engineering profile helps explain why even a lean development-and-operations team can exceed $1M in annual payroll before benefits, payroll taxes, contractors, travel, and field allowances.
1 MW pilot monthly expense
Planning range
Cash-flow note
Core engineering, operations, data, and project team
$80,000-$170,000
Mix of employees and specialist contractors; excludes major redesign campaigns
Vessel, ROV, diver, port, and inspection reserve
$40,000-$120,000
Reserve monthly even when the vessel is not used; intervention cash is lumpy
Planned maintenance, corrective maintenance, and spares
$40,000-$110,000
Includes consumables, seals, bearings, connectors, electronics, workshop work
Site-specific and often understated during development
Insurance and recurring compliance
$15,000-$45,000
Premiums may rise after a claim, retrieval, or extended outage
Administration, legal, accounting, community relations
$10,000-$30,000
Includes reporting to funders, agencies, lenders, and partners
Total modeled monthly cash operating cost
$215,000-$575,000
$2.58M-$6.90M annually before debt service and major life-extension work
12-24 monthsSuggested liquidity runway
A pilot should not depend on next month’s power sale to fund a retrieval campaign or environmental requirement.
1 major retrievalStress-test assumption
Model at least one unplanned retrieval with vessel, port, repair, redeployment, and lost-revenue costs.
10%-20%Operating contingency
Use a separate contingency on top of routine O&M until failure history and intervention timing are credible.
What this estimate hides is timing. Payroll and monitoring are smooth; marine work is not. The project may show a positive operating margin for five months and then spend several hundred thousand dollars in a single intervention month. That is why the cash model should use monthly or weekly timing, not a simple annual average.
How Does a Tidal Project Earn Revenue?
Electricity is the core product, measured in megawatt-hours delivered at the meter. But a first commercial project should avoid assuming that a standard wholesale price will cover its cost. The U.S. Energy Information Administration’s July 2026 outlook forecast national wholesale electricity prices around $45/MWh for the period discussed, while retail and remote-system avoided costs can be much higher. The EIA wholesale-market outlook is a useful reminder that a tidal project competing only with commodity grid energy starts from a difficult position.
For planning, separate energy value from project-support revenue. Energy value includes a PPA, utility tariff, direct sale, or avoided diesel generation. Project-support revenue may include research contracts, milestone grants, demonstration payments, engineering services, technology licensing, data services, capacity or resilience payments, and environmental attributes where available. Some of these are temporary. The financial model must show when each stream expires.
Revenue stream
Pricing unit
Bankability
Main risk
Long-term electricity offtake
$/MWh delivered
High when fixed or floor-priced with a creditworthy buyer
Price too low to support early-stage LCOE
Remote-community or industrial avoided-cost sale
$/MWh displaced or service fee
Potentially strong if tied to a durable fuel-displacement contract
Small load, seasonal mismatch, customer credit
Utility demonstration or capacity support
Annual availability payment
Moderate when contractually committed
Ends after demonstration period
Federal or state R&D award
Milestone reimbursement or cost share
Useful for de-risking, not recurring operating revenue
Cost-share obligations and reimbursement timing
Technology sale, lease, or license
Per device, royalty, or service contract
Scalable after performance is proven
Warranty exposure and long sales cycle
Data, engineering, and marine services
Project fee or monthly retainer
Helpful bridge revenue for a development company
Distracts team from core technology and asset pipeline
Example: 10 MW × 8,760 × 40% resource factor × 88% availability × 94% electrical yield = about 29,000 MWh sold per year.
Commodity grid sale$45-$80/MWh
Illustrative low-value case. Usually insufficient for a first array without very large grants or dramatic cost reduction.
Premium utility PPA$140-$220/MWh
Model assumption for predictable renewable power, demonstration value, or a constrained high-cost system.
Remote or bundled value$220-$400/MWh
Possible only where displaced fuel, resilience, capacity, logistics, and service value are contractually captured.
The cleanest revenue strategy is a long-term contract with a creditworthy utility, community, industrial load, or public entity. The second-best strategy is a transparent stack where each non-energy payment is documented, time-limited, and matched to the cost it supports. Do not fill a permanent operating deficit with a one-time grant in the model.
Capacity Factor, Availability, and Marine Access Drive the Economics
Tides are predictable, but project output is not automatically predictable. The tide tells you when kinetic energy is available. It does not guarantee that the turbine, drivetrain, generator, power electronics, cable, connector, SCADA system, or grid interconnection will be available to convert and sell it. Financial performance therefore depends on three linked quantities: resource, machine availability, and electrical losses.
DOE’s Marine Energy Program supports foundational research, testing, and demonstrations because the sector still needs cost reduction and performance validation. Its marine-energy program overview also points to remote communities and ocean-based industries as markets where the value proposition may be stronger than commodity electricity alone.
35%-45%Hydrodynamic capacity-factor assumption
Use site measurements and a validated power curve. A 5-point miss changes annual energy by roughly 438 MWh for every installed MW before other losses.
85%-92%Pilot availability target
A planning target, not an industry guarantee. Access delays and retrievals can pull an early device below 80%.
92%-97%Electrical yield assumption
Covers array, transformer, cable, conversion, and curtailment losses. Longer cables and weak grids push the result down.
At 10 MW, 40% resource factor, 94% electrical yield, and $180/MWh, one percentage point of availability is worth about $59,000 of annual revenue.
Marine access also creates an inventory decision. Keeping a complete spare drivetrain may tie up several million dollars, but waiting six months for a custom part can destroy a year’s revenue and breach an offtake obligation. The model should compare spare capital against the expected value of avoided downtime, using failure probability, lead time, vessel timing, and revenue per day.
Where Is Break-Even, and What Can the Owner Earn?
Break-even is not the point where electricity revenue covers routine maintenance. It is the point where total recurring revenue covers fixed operating costs, variable operating cost, site and grid charges, insurance, compliance, and the recurring reserve needed to keep the asset operating. A second test then asks whether cash after operations can also cover debt service, taxes, maintenance capital, and distributions.
Operating break-evenBreak-even MWh = annual fixed cash costs ÷ (realized revenue per MWh − variable cost per MWh)
If fixed cash costs are $3.5M, realized revenue is $190/MWh, and variable O&M is $40/MWh, break-even is about 23,333 MWh. A 10 MW array must therefore achieve about 26.6% net capacity factor just to cover operating costs.
Owner earnings come later. For a venture-backed technology company, the founder’s early income is normally a budgeted salary, not a draw from project profits. For an operating asset owned through a project company, the sponsor can distribute cash only after meeting lender covenants, tax obligations, working-capital needs, and reserve requirements. Revenue and accounting profit are not owner income.
10 MW operating scenario
Conservative
Base
Upside
Net electricity sold
25,000 MWh
31,000 MWh
37,000 MWh
Blended realized revenue
$130/MWh
$185/MWh
$260/MWh
Annual operating revenue
$3.25M
$5.74M
$9.62M
Cash operating cost
$5.00M
$4.80M
$4.30M
Operating cash before financing
-$1.75M
$940,000
$5.32M
Debt service, tax provision, and maintenance reserve
$2.50M
$3.00M
$2.80M
Potential sponsor cash before owner-level tax
$0
$0
$2.52M
The base case produces positive operating cash but no distributable owner earnings because financing and reserves absorb it. The upside case works because it combines stronger output, a premium realized price, and lower intervention cost. That is the central investment lesson: a tidal project usually needs both technical improvement and a high-value market. One alone may not be enough.
Which KPIs Should a Tidal Power Developer Track?
A tidal project needs engineering KPIs and finance KPIs in the same dashboard. Availability without cost can be misleading; a turbine may be available because the team spends too much on vessels and spares. Low operating cost can also be misleading if the device is offline. The useful metrics connect output, access, reliability, price, cash, and environmental obligations.
Exact commercial benchmarks remain thin because U.S. grid-scale deployments are limited, so the ranges below are planning targets rather than industry averages. They should be replaced with site data, supplier warranties, test results, and contract requirements as the project matures. NREL’s risk framework specifically treats risk registers and failure-mode, effects, and criticality analysis as living management tools, which makes them natural inputs to KPI review.
KPI
Formula
Planning interpretation
Decision it drives
Net capacity factor
Net MWh ÷ (nameplate MW × 8,760)
Model 25%-40% after all losses; investigate sustained results below the bank case
Revenue forecast, site quality, PPA coverage
Mechanical availability
Available generating hours ÷ scheduled generating hours
Pilot target 85%-92%; warning below 80% after ramp-up
Aim for 90%-105% over rolling 12 months after normalization
Resource model, power curve, lender confidence
Cash O&M per MWh
Cash operating cost ÷ net MWh sold
Must fall below realized revenue per MWh with room for debt and reserves
Commercial viability and cost-reduction roadmap
Revenue per MWh
Total recurring operating revenue ÷ net MWh
Separate temporary grants from recurring contracted value
Market selection and contract structure
Mean time between retrievals
Operating months ÷ unplanned retrievals
Directional target above 12 months for a pilot, rising with maturity
Design qualification and vessel budget
Intervention cost per event
Vessel + port + labor + repair + redeployment + lost margin
Track planned and unplanned events separately
Local vessel strategy, modularity, insurance
Debt-service coverage ratio
Cash flow available for debt service ÷ scheduled debt service
Model at least 1.30x in the base case; lender requirement controls
Debt size, reserve account, distribution lock-up
Cash runway
Unrestricted cash ÷ average monthly net burn
Keep 12-24 months during demonstration and permitting
Fundraising date, hiring pace, deployment timing
Reforecast energy monthly using measured current and downtime.
Price every intervention with lost generation included.
Separate recurring revenue from grant milestones.
Track environmental monitoring against permit commitments.
Update risk priority after each test, failure, and retrieval.
Lock distributions when reserve or DSCR tests fail.
The one-line rule is simple: every KPI should change a decision. If a metric does not change engineering priority, contract strategy, staffing, maintenance, cash reserve, or funding timing, it does not belong on the main dashboard.
Permitting, Environmental Monitoring, and Technology Risk
Permitting is not a single application. Depending on location and project design, a U.S. tidal developer may face FERC licensing, U.S. Army Corps authorization, state coastal and water-quality approvals, Endangered Species Act consultation, marine-mammal review, navigation coordination, fisheries engagement, tribal consultation, cultural-resource work, and local land-use or cable-landing approvals. A project on the federal Outer Continental Shelf can also involve BOEM and BSEE.
FERC says its hydrokinetic pilot process can complete licensing in as few as six months for eligible small, removable projects that avoid sensitive locations, but that target does not mean the whole development process takes six months. Pre-filing studies, stakeholder work, other permits, engineering, procurement, and interconnection can take years. The FERC pilot licensing page should be treated as one path inside a broader schedule.
For federal offshore waters, BOEM explains that grid-connected marine hydrokinetic jurisdiction is shared: BOEM can issue leases, easements, and rights-of-way, while FERC has authority over construction and operation. The BOEM jurisdiction summary is a useful starting point. NOAA Fisheries also notes that federal agencies must consult when an action may affect ESA-listed marine species or critical habitat; its ESA consultation guidance explains the basic process.
Risk
Financial effect
Early control
Model treatment
Collision, displacement, or protected-species concern
Local vessel qualification, framework contract, multiple access methods
30-90 day schedule and revenue delay
Interconnection upgrade
Unexpected capital call and later commercial operation date
Start utility study before final site selection
$1M-$10M separate sensitivity
Supplier failure or custom-part lead time
Redesign, warranty gap, idle device
Dual-source critical parts and retain manufacturing documentation
6-12 months additional runway
PPA or grant milestone delay
Working-capital shortfall even if project remains viable
Milestone acceptance criteria and bridge facility
90-180 day receivable delay
Sandia National Laboratories examined permitting and compliance costs across 19 U.S. marine-energy projects and found that these costs are difficult to generalize because technology, capacity, location, study scope, and project stage differ. That is why the Sandia permitting-cost research is more useful as a warning against false precision than as a single percentage allowance.
How Should the Project Be Funded and Staged?
A first tidal project is normally financed in layers because each source accepts a different risk. Founder capital and seed equity fund the concept and team. Grants, prizes, research contracts, and strategic partners fund testing and demonstration. Growth equity funds manufacturing capability and the development pipeline. Project equity enters once the site, resource, permits, offtake, technology, and installation plan are credible. Construction debt is usually the last major layer, not the first.
DOE maintains water-power funding opportunities and test-support programs, but availability, deadlines, cost share, and eligible applicants change. The DOE water-power funding page should be checked directly rather than assuming a grant will exist when the model needs it.
0-18 monthsConcept and subsystem proof: spend $1M-$4M on design, simulation, tank work, controls, intellectual property, and initial commercial validation.
12-36 monthsOpen-water prototype: spend $3M-$12M on fabrication, deployment, instrumentation, monitoring, retrieval, and redesign. Revenue is mainly grants and contracts.
24-60 monthsGrid-connected pilot: spend $19M-$49M for a modeled 1 MW project. Secure site control, permits, interconnection, offtake, insurance, and a funded operating runway.
48-96 monthsEarly array: raise project equity, strategic capital, tax-credit capital where eligible, and limited debt after pilot performance is independently verified.
After proofReplication: standardize the device, contracts, installation method, monitoring plan, spare package, and financing structure so each new site is not a first-of-a-kind project.
Tax credits can materially change the capital stack, but eligibility and structure need specialist review. The IRS describes the Section 48E Clean Electricity Investment Credit as a technology-neutral credit for qualifying clean-electricity facilities placed in service after 2024. Its Section 48E guidance should be modeled alongside the alternative Section 45Y production credit, not automatically added to it. The IRS Section 45Y page explains the production-credit framework. Prevailing-wage, apprenticeship, domestic-content, ownership, foreign-entity, beginning-of-construction, and emissions-rate rules can affect value.
Fund development through the next measurable de-risking milestone.
Match grant reimbursements with bridge cash and cost-share equity.
Keep technology-company liabilities separate from project-company assets.
Use contingent capital for installation and interconnection overruns.
Do not size debt from an upside energy case.
Document who owns data, patents, warranties, and replacement obligations.
The funding plan should answer one uncomfortable question: who writes the check if commissioning is six months late and the project needs another retrieval before it reaches stable operation? If the answer is unclear, the project is undercapitalized.
What Payback Period Is Realistic?
Payback is difficult for a first tidal array because the initial investment is high, output takes time to stabilize, and a large share of project value may come from grants or tax attributes rather than recurring operating cash. A simple capex-divided-by-EBITDA calculation can therefore look attractive while the sponsor still waits years for actual cash distributions.
Sponsor payback periodPayback period = sponsor cash invested ÷ annual free cash flow available to repay sponsor capital
Use cash after operating cost, debt service, taxes, maintenance capex, reserve funding, and working-capital changes. Exclude noncash depreciation and one-time grant income from annual recurring cash flow.
ConservativeNo payback
$70M sponsor capital, $130/MWh realized revenue, weak availability, and high intervention cost leave recurring cash negative. The project needs restructuring or additional support.
Base18-25 years
$45M sponsor capital after grants and credits, roughly $2.0M-$2.5M annual free cash after stabilization, plus a three-to-five-year development and ramp period.
Upside9-14 years
$30M sponsor capital, premium offtake, strong availability, low-cost local access, and $2.5M-$3.5M annual free cash after reserves.
These are scenario assumptions, not market averages. The clock also needs a clear start date. A 12-year operating payback after commercial operation can become a 17-year sponsor payback when five years of site work, permitting, engineering, and construction are included. Conversely, grants and tax-credit value can reduce sponsor capital enough to shorten payback even when the project’s unsubsidized levelized cost remains high.
Payback improves fastest when the project reduces sponsor cash invested and increases durable cash per MWh. Cutting a turbine’s factory cost by 10% helps, but securing a higher-value buyer, using a local installation method, avoiding one major retrieval, and increasing availability by five points may have a larger combined effect.
How the Financial Model Connects the Whole Project
A useful tidal-power financial model is not a single revenue sheet. It is a chain of engineering, commercial, permitting, financing, and cash assumptions. Site data drive the power curve. The power curve, availability, and electrical losses drive net MWh. Net MWh and contract prices drive revenue. Reliability and access strategy drive O&M. Capital cost and funding sources drive debt, tax-credit value, depreciation, and sponsor equity. All of those lines determine owner cash and payback.
Grant timing, equity, debt, credit eligibility, fees
Sponsor capital, debt service, tax cash
Tax credit delayed or 20% lower
Cash and distributions
Reserve rules, working capital, maintenance capex, DSCR lock-up
Owner cash, runway, payback, equity return
Six-month grant receivable and distribution lock-up
Working capital deserves its own schedule. Grants may reimburse approved costs after payment. Utilities may pay 30-60 days after metering. Vessel contractors may require deposits. Custom suppliers may demand progress payments months before delivery. Insurance can be annual in advance. A project can therefore report positive annual EBITDA and still run out of cash before the next reimbursement or power invoice arrives.
Minimum funding needFunding need = development spend + construction capex + financing fees + peak working-capital deficit + required reserves − committed grants − tax-credit proceeds − available debt
Use committed amounts only. Keep unawarded grants, unexecuted PPAs, and unqualified tax benefits in a separate upside case.
Founders often use a financial model, business plan, and investor materials to keep this chain consistent. The discipline matters more than the format: every technical claim should change energy, cost, timing, or risk; every financing assumption should change cash, ownership, debt service, or payback. When those links are explicit, the model becomes a decision tool rather than a fundraising spreadsheet.
Choosing a selection results in a full page refresh.