Where you place fifty wind turbines on a mountainside is worth millions. Move them a hundred metres and the wake losses change, the access roads get longer, the cable runs get more expensive, and the project’s return moves by a fraction of a percent that decides whether a bank funds it. Traverse Technologies built its business on brute-forcing that problem: generate millions of possible layouts, cost each one properly, and hand the developer the configuration with the highest internal rate of return. The engineering premise is sound and unusually specific. The problem is that the company’s shop window looks like it was abandoned five years ago.
What Traverse Actually Does To A Wind Site
The service covers four stages, from first look to financial close. Prospect assesses preliminary feasibility across large areas. Measure plans and budgets a measurement campaign. Monitor checks the quality of those measurements as they come in. Predict optimises the layout and produces the energy, cost and IRR assessment.
The method behind all four is permutation at scale. Traverse states it generates hundreds of thousands of energy layouts and millions of road and inter-array cable permutations, then selects the configurations with the highest returns and lowest risks. One of its own blog posts frames the cabling problem as choosing between ten to the power of 115 possible electrical connections for fifty turbines, which is a memorable way of showing why humans cannot solve it by hand.
That is a real application of computation to an engineering problem. It is not a chatbot, an assistant or a generative model. It is optimisation, which is where AI has quietly delivered its most reliable industrial value.
Why Optimising For IRR Changes The Answer
Most early-stage wind assessment optimises for energy yield. More megawatt hours looks like a better project. Traverse argues that is the wrong target, because the layout producing the most energy often needs longer access roads through difficult terrain and more expensive cable runs, which eats the extra revenue.
Their approach performs civil and electrical balance-of-plant cost estimation at the earliest development stage, at a level of detail the company compares to what an EPC contractor would produce for a quote. Energy and cost are then traded against each other to find the layout with the best return rather than the biggest output.
The headline claim is that this raises equity IRR on greenfield projects by 0.5%. On a project costing hundreds of millions, that is a large number. No methodology, sample size or independent verification is published to support it.
Check whether this company is still trading before you contact it
- The blog stops in September 2021. The three most recent posts are all from August and September of that year.
- The copyright line reads 2023, which is itself three years behind the current date.
- Placeholder text is still live. The track record section and all four capacity figures sit above Lorem Ipsum dummy copy.
- The testimonials belong to another product. Five identical quotes on the homepage describe hearing back about an internship applied for through a job finder, which has nothing to do with wind farms.
- The capacity claims disagree. The page headline says 7+ GW, the meta description says 10GW, and the individual figures add up to 7,850 MW.
- None of this proves the company has closed. Small engineering firms often let marketing sites go stale while the consulting work continues. But a developer about to share site data should confirm the company is active before starting a conversation.
The Commercial Model Is The Most Interesting Part
Traverse makes a promise most engineering consultancies avoid: no hourly billing, no scope haggling, unlimited revisions and no variation orders. Whenever a development issue appears or new data arrives, revisions are included.
Anyone who has managed a consultancy contract will recognise why that matters. Variation orders are where budgets die, and the fear of triggering one stops developers asking for the extra scenario that might have improved the project. Removing that friction is a genuine commercial differentiator.
The catch is the absence of any number attached to it. Without a fee range, a developer cannot compare Traverse against a traditional owner’s engineer, and cannot tell whether the unlimited revisions are priced into a premium that outweighs the benefit.
The Free Tools Are The Only Way In
Two pieces of free software exist. WindDesk Discover runs in the browser and lets you prospect for wind farm sites without a consulting engagement, and the company announced global availability for it in 2021. WindDesk Monitor covers measurement campaign quality.
For anyone curious about the platform, Discover is the sensible first step. It is the only part of the offering you can evaluate without a sales conversation, and it will tell you quickly whether the underlying data and interface meet your standards.
| Best for | Main use | Pricing model | Platforms | Key limitation |
|---|---|---|---|---|
| Renewable energy developers assessing greenfield wind sites | Layout, road and cable optimisation aimed directly at IRR and LCOE | Quote only. No published pricing. Two free browser tools. Checked 21 Sep 2026. | Consulting engagement, plus WindDesk Discover in the browser | Site unchanged since 2021, with placeholder text still live |
Is Traverse AI Worth Contacting?
The thinking behind this company is better than the website representing it. Optimising directly for IRR instead of energy yield is the correct instinct, the permutation approach is a legitimate use of computation, and the no variation orders model addresses a real commercial pain in engineering consulting. The blog posts from 2021 are genuinely good technical writing.
Everything a buyer would use to build confidence, though, is either missing or broken. There is no pricing, no post-2021 activity, no verification of the IRR claim, and a homepage carrying dummy text and another product’s testimonials. For a business asking developers to trust it with hundred-million-dollar decisions, that presentation is a problem regardless of how good the engineering is.
Our position: try WindDesk Discover, because it costs nothing and shows you the data quality. Before any commercial conversation, confirm through LinkedIn or direct email that the company is still operating and ask directly for the evidence behind the 0.5% IRR figure.
What Traverse Should Fix First
- Remove the Lorem Ipsum placeholder text from the live track record section.
- Replace the job finder testimonials, which belong to a completely different product.
- Reconcile the 7+ GW and 10GW capacity claims into one accurate figure.
- Update the copyright year and publish something, anything, to show the business is active.
- Publish the methodology behind the 0.5% IRR improvement claim.
- Give an indicative fee range so developers can shortlist without a sales call.
