A seller packs orders late at night, sees roughly ₹25,000 in monthly sales, and feels relieved: people are buying.
Then a return comes back. Fees and packaging have already been paid. The sales number remains, but the cash may not. The useful question is: “Did this order earn anything?”
The sales number is only the first number
A customer payment is only the starting number. Product, packing, fees, refunds and owner time sit between that payment and actual contribution.
Those figures may live in different reports. Without matching them order by order, revenue can grow while cash disappears.
The usual moment this begins is before repricing or restocking after returns and settlement deductions. What follows is scaling orders that lose money and tying up cash without noticing.
Follow one ₹1,000 order
Imagine a ₹1,000 order. If product, packing, fees and a return total ₹950, almost nothing remains before the seller’s time. This is an example, not the seller’s actual margin.
The source lacks the order records needed for a real calculation.
What is supported by evidence—and what is still only a hypothesis?
What the evidence says
Episode 8 reports about ₹25,000 monthly sales, ₹250 product cost, ₹50 packaging and returns falling from 30% to 10%. These claims are unverified.
Independent seller accounts repeat concerns about returns, fees and labour.
Amazon documents multiple fee categories; Cointab and EasyEcom show reconciliation is an existing task.
What we know
- sales reported; profit unknown
- ₹25K / month Reported in source
- Evidence level
- E1 Repeated signal
- Field interviews
- Not completed Proposed in the next test
Counterargument: Returns reportedly improved. Existing reports, accounting support or software may already be adequate.
Where the order money goes
Follow one settled order from customer payment to cash left after its costs. The model prices an audit; it does not estimate the seller’s margin.
₹250 product cost plus ₹50 packaging were reported. Selling price, fees, return costs and owner time remain unknown.
Who might pay: The seller would decide whether to pay for an audit. Willingness to pay is unknown.
| Price to test | ₹2,000 |
|---|---|
| Direct cash cost | − ₹200 |
| Contribution before owner time | ₹1,800 |
| Owner time | 8 hours × ₹150 |
| Contribution after owner time | ₹600 |
These numbers are not a forecast. They make the hypothesis measurable and keep weak economics visible.
How sellers try to find the answer now
Reports, spreadsheets, bookkeepers and reconciliation tools answer parts of the question. We do not know whether they already give this seller a reliable order-level answer.
Why the true margin stays hidden
The owner is busy sourcing, packing and handling returns. Checking every deduction is easy to postpone until cash becomes tight.
Order, return, refund, fee and settlement records need matching. Purchase invoices and packaging costs may sit outside the marketplace dashboard. Growth taking priority over full costing is a behavioural hypothesis.
What could make a test easier: Exportable transaction records make a manual audit possible.
A useful answer may be smaller than a dashboard
The gap may be a one-time audit, not another dashboard: sample settled orders and find what the current process misses.
It matters only if the finding changes a decision and is worth more than the audit.
One-time margin audit
Match 50 settled orders to purchases, fees, returns and cash receipts.
Weakness: Messy records can take too long.
Improve the current process
Fix a cost sheet, accounting workflow or existing tool.
Weakness: If this already works, a separate service is unnecessary.
Practical boundary: Seller consent and anonymized exports. Keep tax or assurance work within qualified scope. Yes, if the operator is reconciliation-competent.
How we’d test this for ₹2,500
Start with evidence, not a product. The experiment should answer one decision before any larger commitment.
- Participants
- 5 sellers · 50 settled orders each
- Time
- 7 days
- Research budget
- ~₹2,500
- Owner time
- 18 hours
Does the seller’s current process miss an actionable margin or settlement issue?
- Get consent and define a usable sample.
- Match costs, fees, returns, refunds and cash receipts.
- Ask sellers to verify each exception against current reports.
Continue if
- 3 of 5 provide usable records.
- 2 confirm an actionable gap missed by their process.
- A 50-order audit fits within 8 owner hours.
Stop if
- Existing reports already answer the question.
- Records cannot be verified.
- Delivery takes too long for an acceptable price.
Will you test this?
These are our proposed decision rules, not industry benchmarks.
Why a margin audit may be a bad business
This can be real and still make a poor service. Messy exports and edge cases may consume hours, while existing tools may be good enough.
- Incomplete records, changing platform rules, limited budgets and labour-heavy exceptions.
If nobody will pay for this, it is not a business—just a real problem.
Evidence snapshot
Opportunity snapshot
A public view of what is known before a solution is proposed. It is not a market-size or margin claim.
- Pain / severity
- High consequence reported; prevalence is not established.
- Who pays
- The seller would decide whether to pay for an audit. Willingness to pay is unknown.
- Current alternatives
- Marketplace reports, spreadsheets, bookkeepers and reconciliation software. Whether the seller already has an adequate answer is unknown.
- Test cost
- Proposed 7-day research budget: ₹2,500.
- Main risk
- Incomplete records, changing platform rules, limited budgets and labour-heavy exceptions.
- Evidence
- E1 · Repeated signal.
What would change our mind? Returns reportedly improved. Existing reports, accounting support or software may already be adequate. We would also need: Will sellers pay for findings their current process misses?
Our current view
Worth investigating. Too early to build software.
The money can be measured and the first test is cheap. What remains unknown is whether small sellers will pay separately for findings their current process misses.
Next move: Audit real orders.
Research notes
- Problem score
- 54 / 100
- Opportunity score
- 47–67 / 100
- Biggest unknown
- Will sellers pay for findings their current process misses?
Scores are internal research judgments, not probabilities of success.
Sources
Sources support the specific claims described here. A reported account or an operator’s existence does not validate a market-wide opportunity.
Business Mastery · Episode 8
Source conversation
Participant or facilitator account. Figures have not been independently verified.
Open source ↗Amazon India selling fees
Platform source
Explains charge categories; does not establish this seller’s losses.
Open source ↗Seller order economics discussion
Independent account
Anonymous, unverified account. Supports a repeated signal, not prevalence.
Open source ↗Seller exit and returns discussion
Independent account
Self-reported experience; not a population estimate.
Open source ↗Cointab ecommerce reconciliation
Existing operator
Shows that matching orders, returns, fees and settlements is an existing task.
Open source ↗EasyEcom payment reconciliation
Existing operator
A process alternative; marketing claims are not audited outcomes.
Open source ↗
Reality check
Does this happen in real life?
One anonymous signal helps us decide what to investigate next.
Continue exploring
