Why IC memos for BESS break the IE-report model
Independent engineer's reports were built for project finance on solar and wind. They were not built for batteries, and they were not built for the screening stage. Here is the structural gap.
Ask any infrastructure analyst in Sydney what the most painful part of their week is, and the answer is the same: reading an independent engineer's report for a deal that will never make it to investment committee. Six weeks of bespoke engineering work, between A$150k and A$500k of client money, and at the end of it the analyst still has to rebuild the financial model from scratch to answer the question their IC actually cares about.
The IE-report model worked for solar and wind because those are power products. The technical answer (resource estimate, capacity factor, AEP with uncertainty bands) maps cleanly onto the financial question. A bankable IE report plus a financial model is, broadly, an IC memo.
Batteries are revenue products, not power products
A BESS does not have an AEP in any meaningful sense. It has a revenue distribution across energy arbitrage, FCAS products, capacity payments, and (in some markets) ancillary services, that depends on a price curve and a dispatch decision. Ship a bankable P50 revenue number and you have told the IC something they already suspected and cannot use.
What the IC actually needs is the cone: P10, P50, P90 by revenue stream, the sensitivities that widen or tighten it, and the handful of red flags on the grid side that could force the cone open wider than the model suggests. None of that fits the IE-report template.
The screening-stage gap
Even if the IE-report model worked for batteries, it would not solve the screening problem. Investors screen roughly 50 projects to approve 5. The IE reports only get commissioned on the 5. So 90% of the diligence burden at screening stage has no productised tool behind it at all.
Analysts fill the gap by hand. Excel packs, hand-built red-flag schedules, informal calls with ex-developers to sanity-check the connection register. It is the most expensive way to run a pipeline.
What a screening-first product looks like
If you design from first principles for the investor workflow rather than for project finance, three things shift:
- 1.The unit of work becomes the memo, not the report. Technical, revenue, and counterparty read-across live in one document, shaped for the IC question.
- 2.The output is a distribution and a recommendation with uncertainty bands, not a point estimate presented as fact.
- 3.The cost structure flips. Subscription for pipeline screening (where the quantum of work per project is small and highly repeatable), per-deal fees for the full IC memo. Investors stop paying IE rates for projects that never clear screening.
Agentic coding has collapsed the cost of getting there from roughly US$20M to roughly US$2M. Whoever productises this layer first wins a ten-year window.