Trust, Reimagined
Interoperability Is Not Trust
The big questionWhen no single platform owns the transaction, how does trust travel?
If nothing opened, use your browser’s print command (Ctrl or ⌘ + P). The teardown is laid out for it.
The three-minute version
Open networks let any buyer app, seller app and service provider interact through shared standards. Trust is a different question: can they interact safely, accountably and fairly, and what happens when something goes wrong? This series imagines trust that belongs to the transaction rather than to a platform, so that it can travel with the people involved. It builds on the complaint, payment and dispute foundations India already has. Eight shifts, each with its sources and the result that would end it.
Do protected payments and early exception alerts reduce formal disputes in one category, without making it harder for small sellers to join?
Where things stand.
The question
Platforms made digital commerce trustworthy for millions of people. They brought discovery, identity, payment, fulfilment, support, disputes and reputation together, so when something went wrong, a buyer knew exactly where to go. That simplicity is one of platform commerce’s great strengths.
Open networks such as ONDC separate those functions on purpose. A buyer may discover through one app, a seller may sell through another, and payment and logistics may come from others still. That is what makes the network open. It also raises a new question: who is responsible for trust?
It is a question with real traffic behind it. Between 25 April and 26 December 2025, India’s National Consumer Helpline handled 67,265 refund-related complaints, and nearly 40,000 of them were about e-commerce[8].
When trust belongs to the platform, leaving the platform means leaving the trust. When trust belongs to the transaction, it can travel.
The ideal is not “I trust this platform” but “I can trust the conditions under which this transaction takes place.” This series does not argue that platforms handle trust badly. It asks what would have to exist if the platform were no longer the one place where trust lived.
Interoperability is not trust
Interoperability asks whether these participants can interact. Trust asks whether they can interact safely, accountably and fairly, and what happens when something goes wrong. The first is a matter of standards. The second needs something more, and around the rails of an open network six things still have to exist: protection, evidence, resolution, reputation, capability and learning.
None of them has to start from nothing. India has already begun to build several, and the eight shifts that follow are an attempt to join them up, not to replace them.
What India has already built
| Where | What already exists |
|---|---|
| Complaints to an app | Since July 2020, India’s e-commerce rules have required a grievance officer, acknowledgement of a complaint within 48 hours and redress within one month[1]. |
| Escalation on the network | ONDC’s Issue and Grievance Management framework lets an unresolved complaint go from the buyer app to ONDC. ONDC lists empanelled online dispute-resolution providers for conciliation, mediation or arbitration, and says an automated process is being added to the framework[2][3]. |
| Payment disputes | A UPI complaint moves from the app to the payment service provider’s bank, then the customer’s bank, then NPCI, and then the Ombudsman[4]. |
| Money in transit | A payment aggregator keeps merchants’ money in a separate escrow account with a scheduled commercial bank, and must run a mechanism for payment disputes[5]. |
| The consumer’s own route | The Consumer Protection Act, 2019 applies in addition to other law, and a complaint to a consumer commission can be filed within two years[6]. |
| Trust as a design problem | ONDC’s 2022 consultation paper treated discovery, payments, grievances, scoring and enforcement as separate trust problems, and put forward proposals for each[7]. |
Scope. The consultation paper’s ideas were proposals put out for comment. We have not checked each one against the network’s current policy, so we treat them as foundations to test, not as settled facts.
Eight shifts, each one testable.
Each shift moves from today’s model to a possible alternative. Open one to see how it would work, the sources behind it, what could go wrong and the result that would end it.
- 01The best participant doesn’t have to be the most visible one.Visibility→Trusted fitRead →
- 02Money shouldn’t become unprotected the moment it moves.Pay, then recover→Protect, then releaseLooking for a pilot partnerRead →
- 03Evidence should be a by-product of commerce.Evidence after the fact→Evidence as a by-productLooking for a pilot partnerRead →
- 04Not every problem needs a dispute.After the complaint→Before the disputeLooking for a pilot partnerRead →
- 05A complaint is not a finding.Allegation→Fair processRead →
- 06A failure should tell us what needs to improve.Who’s wrong→Why it failedRead →
- 07Evidence of what you’ve done, including how you’ve grown.Star rating→Reputation as proofRead →
- 08The purpose is to make fewer disputes necessary.Case by case→Network learningRead →
The smallest real test.
The set-up
One category on one network, with the apps, a regulated payment provider and an online dispute-resolution provider already working in it. Start with protected payments and early exception alerts, then add each layer in turn. Every kill threshold is written down before launch.
What each test would show
| Shift | What is tested | The question it answers |
|---|---|---|
| Shift 1 | Trust-aware discovery | Do buyers use protection as a search condition, and do newcomers still get found? |
| Shift 2 | Protected payments | Will small sellers accept protected terms, and how long do they wait for release? |
| Shift 3 | Evidence from the record | Can most disputes be resolved from the record alone? |
| Shift 4 | Early exception alerts | Do alerts end problems before they become disputes, without feeling like pressure? |
| Shift 5 | Fair resolution | Is a record-based process faster and fairer than today’s, as both sides rate it? |
| Shift 6 | Capability support | Do struggling sellers take support up, and does the improvement last? |
| Shift 7 | Proof of record | Do buyers use proof-based profiles, and do recovered sellers regain business? |
| Shift 8 | Network learning | Do the top dispute patterns fall after a fix, and stay down? |
Behind the teardown.
The rules it has to fit, how we would bring it to life, what it would take and how to read it.
The architecture, layer by layer
| Layer | What it does |
|---|---|
| Discovery | Find participants by capability, evidence, reputation and terms |
| Payment protection | Hold funds under agreed fulfilment conditions |
| Commitments | Record exactly what was agreed |
| Evidence | Build a verifiable record of what happened |
| Exceptions | Catch deviations early and help correct them |
| Resolution | Negotiation, mediation and fair decisions when needed |
| Capability and reputation | Help participants improve, and record verified performance and recovery |
| Network learning | Turn recurring failures into better standards and tools |
Three things are protected: money, truth and reputation. Four things are kept apart: a complaint is not a dispute, a dispute is not a finding of fault, and a fault is not a permanent reputation.
One principle throughout
AI helps the network understand what happened. People and fair process decide what should happen.
In this design, software may assemble evidence, notice when a transaction drifts from its commitments, explain a disagreement and suggest ways to settle it. It does not accuse, decide or punish. Every shift that uses it keeps to that line.
Where Trust sits in the stack
Trust is one layer of a larger open-network stack, and it leans on the others.
- Discovery asks how everyone becomes findable. Shift 1 here asks how the conditions of a transaction become part of what is found.
- Hyperlocal Commerce meets the cold-start problem when a new provider has no history. Proof built from real transactions is how it ends.
- Human Capital asks how proof of capability travels with a person. Shift 7 asks the same of a seller’s record.
- Trade Finance reads the same transaction evidence as the basis for credit.
How we’d bring it to life
Insight Visualization. Agree what must be true: that protected payments, structured evidence and early exception handling measurably reduce disputes and increase trust between strangers.
Situational Awareness. Map how disputes arise and are resolved today in one network category, with buyers, sellers, apps, payment providers and dispute-resolution providers in the room.
Evolution Cycles. Add each layer in turn, starting with protected payments and exception alerts.
Grow what works, category by category, under shared governance.
What it would take
Business Building, Product & Technology Engineering, Systems Architecture, Data & AI, Business Model Innovation and Operating Model Design, with payment regulators, dispute-resolution providers and network governance involved from day one.
How to read this teardown
This is a Level 1 · Concept teardown: the idea, how it would work and how we would test it. Every figure is sourced and every policy claim is cited, and the sources are listed at the foot of each page. Examples marked illustrative are constructed to explain an idea, and the numbers inside them are not data. No separate evidence review has yet been commissioned for this series, so it does not carry the claim-by-claim evidence ratings that Level 2 teardowns do.
The people in the examples are invented: Kavita, who runs a small packaging unit in Pune, and Imran, who runs a six-person fabrication workshop in Rajkot.
What the evidence supports.
Every claim sits in one of four tiers. Sources carry the status of our last check.
How strong it is
- Established
- India already runs layered complaint handling: e-commerce rules, ONDC’s grievance framework with dispute-resolution providers, a UPI dispute route and RBI rules for escrow. Payment escrow has been run at scale elsewhere.
- Evidence-supported
- Star ratings can compress into a narrow band near the top and under-report bad experiences (eBay data, 2011). Software alone has resolved most disputes in a large marketplace (eBay, self-reported, 2010).
- Hypothesis
- That a transaction record shared across independent apps can be assembled cheaply and trusted; that early alerts cut formal disputes; that support beats removal for sellers who want to improve; that portable, proof-based reputation changes what buyers choose.
- Unknown
- Whether independent apps and providers will adopt common commitment and evidence formats; whether conditional release can be offered across payment aggregators; what it costs to keep a participant-held record.
Before a pilot we would confirm
- Primary text of the Consumer Protection (E-Commerce) Rules, 2020 not fetched (legal summary used).
- ONDC IGM specification documents (Explainer, User Manual, Business View) not read; only the overview pages.
- Whether ONDC has adopted, changed or dropped the 2022 consultation proposals (scoring agencies, settlement triggers, ODR tiers).
Notes on the evidence
ONDC’s consultation paper is from September 2022 and sets out proposals for comment. We have not checked each proposal against the network’s current policy.
Holding money and releasing it on conditions sits inside RBI’s rules for payment aggregators. Whether release tied to delivery or milestones can be offered across providers is to be confirmed with regulated providers before any pilot.
The eBay dispute figures date from 2010 and were given by the person then responsible for its online dispute resolution. They show what software can do in principle, not what this network would achieve.
Sources
- [1]Lakshmikumaran & Sridharan, Consumer Protection (E-Commerce) Rules, 2020 notified (2020)Checked via a legal summary; the Rules’ own text not read
- [2]ONDC, Enabling fair commerce: ONDC’s Issue and Grievance Management goes live; Issue and Grievance Management (IGM) resourcesChecked; the linked specification documents not read
- [3]ONDC, ComplaintsChecked
- [4]NPCI, UPI dispute redressal mechanismChecked
- [5]Reserve Bank of India, Regulation of Payment Aggregators Directions, 2025 (15 September 2025)Checked
- [6]Consumer Protection Act, 2019: section 69 (limitation) and section 100 (not in derogation of other law)Checked
- [7]ONDC, Building Trust in the ONDC Network: Consultation Paper (29 September 2022)Checked; proposals for comment, not adopted policy
- [8]Storyboard18, Consumer Affairs dept facilitates Rs 45 cr refunds in 8 months as e-commerce tops complaintsChecked; press report of Department of Consumer Affairs figures
Spotted a source we missed or a rule we misread? Tell us what we got wrong.
Looking for a pilot partner
We would most like to test shifts 2, 3 and 4 with a real partner. If one of them fits a pilot you could host, fund or take part in, we would like to hear from you.
Closed platforms create trust by controlling the transaction. Open networks can create trust by making the transaction accountable.