EDPMS
Why is my EDPMS utilisation short when I sell digital goods too?
A blended payout mixes goods and services revenue, so only the goods share can fill a shipping bill. The arithmetic that separates the two, and why the gap is not short realisation.
By Aaryan Kakani · · 9 min read
What does utilisation actually mean in EDPMS?
Utilisation is a running total on the remittance side of the ledger. A shipping bill enters the Export Data Processing and Monitoring System carrying the value declared to Customs. It is discharged when inward remittances your AD bank has allocated against it account for that value. The word describes what the bank has done with a credit, not a property of the bill itself.
Two consequences follow, and both of them matter here. One remittance can be split across many shipping bills, and one shipping bill can draw on many remittances. The relationship is many-to-many, which is why a payout can never be read off against a bill by eye. And the system reports the arithmetic result without reporting its cause: an under-utilised bill looks identical whether the money never arrived, arrived net of fees, arrived in the next window, or arrived alongside revenue that was never export of goods at all.
The symptom an exporter actually sees is a query from the bank saying utilisation against a shipping bill is less than its value, often as a list of bills approaching the end of their realisation period. That list is a starting point for diagnosis, not a finding against you. For the mechanics of how bills move through the system and how remittances are lodged against them, see how EDPMS reporting works .
Why does a blended payout guarantee a shortfall?
A physical export and a digital sale are handled by two systems that do not meet. The physical order produces a shipping bill and, through it, an EDPMS entry that carries a declared value waiting to be discharged. The digital order produces neither, because nothing crossed the customs frontier and there is no customs document to generate. Whatever reporting the services side does attract, it is not a shipping bill, and a shipping bill is the only thing utilisation is measured against. What the two orders share is the settlement: the platform sweeps both into the same payout window, nets its fees and remits one figure to one account.
So the credit that lands is composed of two things, and only one of them has a document that utilisation can be measured against. That produces a fork with no comfortable branch:
- Allocate the full credit against the goods shipping bills and the numbers close, but the closure is wrong. Part of that money is not proceeds of the exported goods, and the services revenue has simultaneously gone unreported through its own route.
- Allocate only the goods share, which is correct, and every shipping bill in the set draws less than its declared value. Utilisation reads short across the whole set.
The second branch is the right one, and the shortfall it produces is not a defect to be hidden. It is the arithmetic signature of a blended settlement, and it recurs every window for as long as the blend does. An exporter who does not know this is looking for a missing payment that was never missing.
Is this under-utilisation or short realisation?
These are different findings with different consequences, and the distinction is worth holding firmly because a bank query does not draw it for you. Short realisation means proceeds of the exported goods did not come in. Under-utilisation means the remittances allocated to a bill do not add up to its declared value. Every rupee of goods proceeds can have arrived and the bill can still read short.
Work down the causes in order. Each one is checkable against documents you already have, and each is a complete explanation on its own before you reach for the next:
| Cause of the gap | What settles it |
|---|---|
| Platform and gateway fees deducted before remittance | The fee lines on the settlement statement for exactly those orders |
| Currency conversion between invoice currency and the credit | The conversion rate and date applied on the bank advice |
| Refunds and chargebacks netted inside the same window | The refund lines, matched to their original order lines |
| Orders settling on the far side of the window boundary | The next window's statement, showing the same orders |
| Digital or services revenue sharing the payout | The line-item split of goods lines against services lines |
| Goods short-shipped against the declared quantity | The shipping bill amendment record for that consignment |
| Proceeds genuinely not received | Only what remains after every row above is evidenced |
The last row is the only one that is genuine short realisation, and it is a residue rather than a starting assumption. Where goods were short-shipped against the declared quantity the cure is an amendment to the bill rather than a reconciliation, and that is covered in short shipment amendments .
What arithmetic do I show the bank?
Show the composition of the credit, then the tie-out. The bank is not asking you to assert that fees and digital sales exist; it is asking you to account for a specific credit against specific bills. The working below uses invented figures to show the shape of the calculation. Substitute your own from the settlement statement.
A seller ships three consignments in a settlement window and also sells downloadable design files through the same storefront. One credit of
USD 9,240
lands. The three shipping bills declare
USD 4,000
USD 3,500
and
USD 2,900
USD 10,400
in total. The bank asks why utilisation is short by USD 1,160. All figures here are illustrative. </> } result={ <> The goods proceeds are
USD 8,164
, not USD 9,240, because USD 1,076 of the credit is services revenue that no shipping bill can absorb. Against declared value of USD 10,400 the goods stream is short by
USD 2,236
. A larger gap than the bank's USD 1,160, and the correct one. Every rupee of it is accounted for on the statement: nothing here is unrealised. </> } >
| Credit received in the account | 9,240 | |
| Less: digital order lines (gross) | (1,180) | No shipping bill exists for these |
| Add back: fees charged on digital lines | 104 | Fees follow their own revenue line |
| Services revenue inside the credit | (1,076) | Reported through its own route |
| Goods proceeds in the credit | 8,164 | The only figure a bill can draw on |
| Declared value, three shipping bills | 10,400 | 4,000 + 3,500 + 2,900 |
| Less: platform commission on goods lines | (1,456) | Evidenced per order line |
| Less: refund on one order, same window | (620) | Matched to its original line |
| Less: currency conversion difference | (160) | Rate and date on the bank advice |
| Expected goods proceeds after deductions | 8,164 | Ties to the credit exactly |
Two features of that working are the point of it. First, the services line is removed gross and then re-netted: fees charged against digital orders belong to the digital stream, so netting them into the goods deduction chain would understate goods proceeds and manufacture a shortfall that does not exist. Second, the honest gap came out larger than the one the bank raised, and that is the normal result. The bank was comparing the whole credit to the whole declared value, which understates the goods-side gap by exactly the services revenue.
Which documents evidence the split?
The composition of the credit is the thing in dispute, so the evidence has to be at line-item level. A summary of the payout restates the number you are trying to explain. Assemble these for each window:
- The settlement statement export for the payout, at order-line level, showing each order, its character (physical or digital), its gross value and the fees charged against it.
- The shipping bills for the physical order lines in that window, with their declared values, so the goods set is defined rather than asserted.
- The refund and chargeback lines falling inside the window, each matched to the original order line it reverses.
- The bank advice for the credit, carrying the conversion rate and the value date applied.
- The tie-out sheet itself. The arithmetic in section 4, run on your figures, ending in a residue you can name.
Where the remittance arrives through a payment provider rather than direct from a buyer, the provider's own advice document is part of this set but does not replace your AD bank's lodgement of the remittance. That distinction, and which providers issue what, is set out in FIRA by payment provider and, for settlement-level e-BRC mechanics, in e-BRC for e-commerce settlements .
What do I do once the streams are separated?
The deliverable is a tie-out sheet ending in a named residue, not an assertion that fees and digital sales exist. A residue you can name is a manageable conversation; an unexplained gap against a realisation deadline is not. </> } />
The last step is the one this guide cannot take for you, and it is deliberately left open. How a bank codes and evidences the non-goods portion of a blended remittance, and whether it will split one credit across two purpose codes or ask you to take separate payout streams, is a matter of that bank's practice. Start from purpose codes for marketplace receipts and RBI purpose codes ; where a code has already been applied and is wrong, the correction route is in changing a purpose code .
If bills are already close to the end of their realisation period while the reconciliation is in progress, say so to the bank in writing rather than letting the deadline pass in silence. the pending-realisation letter is the format for that.
Which figures does this guide deliberately not state?
This guide states no rate, threshold, deadline or purpose code of its own, and that is a design decision rather than an omission. The reconciliation above is arithmetic over your own documents: it holds whatever the current regulatory figures are, and it would still hold if they changed tomorrow. Figures that go stale silently are the failure mode this page is built to avoid.
So where a number is needed, take it from the instrument or the page that carries it with its citation:
| Figure you may need | Where to take it from | |||||
|---|---|---|---|---|---|---|
| The realisation period for a shipping bill, and when its clock starts | /guides/nine-month-clock-starts-when | text-teal-700 hover:text-teal-900 underline underline-offset-2 | /guides/fema-repatriation-ecommerce | text-teal-700 hover:text-teal-900 underline underline-offset-2 | ||
| Reduction in invoice value, write-off limits and small-value closure | /guides/one-payout-many-shipping-bills | text-teal-700 hover:text-teal-900 underline underline-offset-2 | /guides/edpms-under-10-lakh | text-teal-700 hover:text-teal-900 underline underline-offset-2 | ||
| The purpose code for a given receipt, and how to correct one | /guides/rbi-purpose-codes | text-teal-700 hover:text-teal-900 underline underline-offset-2 | /guides/purpose-code-change | text-teal-700 hover:text-teal-900 underline underline-offset-2 | ||
| Whether the digital half is a zero-rated export of services | /guides/gst-export-services | text-teal-700 hover:text-teal-900 underline underline-offset-2 | /guides/marketplace-payout-gst-split | text-teal-700 hover:text-teal-900 underline underline-offset-2 | ||
| Your platform's own fee schedule for goods and digital lines | The settlement statement itself. Published fee tables are indicative; the statement is what the bank will accept. |
Frequently asked questions
What does utilisation mean in EDPMS?
It is the amount of inward remittance your AD bank has allocated against a particular shipping bill. The bill enters the system with the value declared to Customs and is discharged when allocated remittances account for that value. Because one remittance can be split across many bills and one bill can draw on many remittances, utilisation is a running total on the remittance side rather than a property of the bill.
Why does selling digital goods make my utilisation come up short?
Because a digital sale never generates a shipping bill, and a shipping bill is the only thing utilisation can be measured against. When one payout settles physical and digital orders together, only the physical portion has a customs document behind it. Allocated correctly, the goods share is smaller than the credit, so every bill in the set draws less than its declared value. The shortfall is a consequence of the blend.
Is an under-utilised shipping bill the same as short realisation?
No. Short realisation means goods proceeds did not arrive. Under-utilisation means allocated remittances do not add up to the declared value, which happens routinely while every rupee has in fact been received. Through fees, conversion, refunds netted inside the window, orders settling either side of the boundary, and services revenue sharing the payout. Only the residue after all of those are evidenced is a candidate for short realisation.
Can I allocate digital or services revenue against a shipping bill?
No. A shipping bill covers goods that crossed the customs frontier on a stated date. Closing it with services revenue misstates the goods position and the services position at once. Separate the two on the settlement statement and report each through its own route. How your AD bank wants the non-goods portion coded is a question for the bank, and bank practice differs.
What evidence does an AD bank need for a blended settlement?
The line-item settlement statement for exactly the orders in the payout, showing which lines are goods and which are digital, the fees against each, and any refunds netted inside the window. Plus the shipping bills for the goods lines and the tie-out arithmetic. A payout summary does not work, because the composition of the credit is the thing in dispute rather than its total.
How do I stop this recurring every settlement cycle?
Separate the streams before the money moves. Where the platform allows separate payouts for physical and digital sales, each credit then has a single character and needs no apportionment. Where it does not, keep the line-item settlement export for every window as a standing record and reconcile while the orders are still identifiable. Doing it monthly on fresh data is a different job from doing it a year later against a caution-listing threat.
Update history
- First published.