$6.85M freed. One integration.

A mid-sized money transfer operator had been running the USD to NPR route for several years. By the time they spoke with Elytra, they were processing approximately $500M/year on the route, a substantial, mature operation. Compliance was in order. The correspondent relationships were established. The customer-facing product worked.

The economics, examined carefully, did not.

The operator's CFO had a number she returned to periodically, without quite knowing what to do about it: $6.85 million. That was the approximate value of pre-funded Nostro balances the company maintained across the correspondent chain to keep the USD to NPR route operational. The money was real. It appeared on the balance sheet as cash. But it was unavailable. It sat at correspondent banks, earning nothing, required as the operating condition of their core route.

She knew it was expensive. She did not have a clear line of sight to an alternative.

The standard USD to NPR correspondent chain for a mid-sized MTO looks like this:

MTO → US correspondent bank → International clearing correspondent → Nepal banking partner → Recipient

Each relationship in this chain carried its own requirements:

US correspondent: minimum Nostro balance of approximately $3.2M

International clearing correspondent: approximately $2.1M prefunded balance

Nepal banking partner: approximately $1.55M for inbound batches

Total Nostro requirement across the chain: $6.85M, held continuously.

None of this capital earned a return. The correspondents paid zero interest on operational balances. The $6.85M was, for practical purposes, a permanent working capital cost of operating the route.

The Nostro balance was the largest single hidden cost, but it was not the only one.

Transaction fees: SWIFT messaging fees, correspondent processing charges, and the Nepal banking partner's receiving fees together accounted for approximately 1.2 to 1.8% of transaction value.

FX spread: the operator converted USD to NPR at each settlement cycle. The rate applied was consistently 2 to 3% worse than mid-market. On $500M/year in volume, a 2.5% average FX spread represents $12.5M in annual FX cost.

Settlement delay: standard settlement was 3 to 4 business days end-to-end. This meant approximately $4.1 to $5.5M in working capital in transit at all times.

The operator's internal cost model had them running the route at approximately 1.8%, the figure that captured only the visible fees. The actual all-in cost was more than double that.

Nostro requirement: $0. Real-time routing against live liquidity eliminates the pre-funding requirement. There is no correspondent holding a standing balance on the operator's behalf. The $6.85M returns to the operator's balance sheet as deployable capital.

Settlement time: under 6 hours. End-to-end settlement, from sender initiation to funds available in recipient account, would complete in under six hours.

All-in cost: under 1%. Elytra settles at or near mid-market FX rates with no correspondent spread and no hidden markup. Against the operator's current 4.1% actual cost, this represents a recovery of approximately $15M per year on $500M in volume.

Integration: single API, no customer-facing change. The Elytra integration is a back-end replacement. The operator's existing tech stack, transaction management, KYC/AML, customer accounts, agent interfaces, remains unchanged.

$6.85M in Nostro balances recovered. At 5% cost of capital, that is $342,500 per year in recovered opportunity cost, before counting a single dollar of fee savings.

More significantly: every new route the operator launches on Elytra requires zero additional Nostro capital. On the correspondent chain, each new route carries its own pre-funding requirement, typically $2 to $5M in new Nostro balances for a mid-volume route. On Elytra, a new route is a routing configuration. The marginal capital cost of expansion is zero.

For an operator that has been constrained from launching new routes because of the capital requirement to pre-fund them, this changes the growth model entirely.

Beyond the financial recovery, the move from four correspondent relationships to one Elytra integration simplifies the operator's operational overhead significantly.

On the correspondent chain, the operator maintained: four bilateral correspondent relationships with individual contract terms, four separate Nostro accounts requiring periodic rebalancing, four sets of compliance requirements, and four points of operational failure.

On Elytra: one API integration, one contract, one consolidated audit trail generated automatically per transaction, and one point of contact for settlement issues.

This operator's situation is not unusual. It is the standard operating condition for any MTO running a high-volume route on the correspondent chain.

The $6.85M Nostro figure is not specific to this company. It is a function of the route volume and the correspondent chain's pre-funding requirements. An MTO at $250M/year carries approximately $3.4M in idle Nostro balances. At $1B/year, approximately $13.7M.

The FX spread is not negotiable in any meaningful way. Correspondents set rates to capture a margin on conversion. The spread is a structural feature of the correspondent chain.

The working capital in transit is not avoidable on a T+3 settlement rail. It is a direct consequence of the settlement timeline.

None of these costs are inherent to the business of moving money. They are costs imposed by the infrastructure the money moves through. The infrastructure can be changed.