Inside a 14-Section Merchant Onboarding Playbook
Abstract
A playbook is not a policy document and not a form. It is the bridge between what a payment provider's rules require and what a merchant can actually prepare before applying. This note describes the structure of a 14-section, compliance-first onboarding playbook prepared under the Zarelva merchant-risk advisory — the five assessment layers, the evidence pack, the risk-triage table, the readiness checklists, and the decisions that shaped each. Client details, provider thresholds and rule identifiers are deliberately excluded.
1. The problem
Merchants rarely fail onboarding because their business is unacceptable. They fail because the submission cannot answer the reviewer's questions in the order the reviewer asks them. The merchant side of that problem is under-served: providers publish rules, not preparation guides, and what circulates in merchant groups is folklore about which documents "work".
2. Why current practice falls short
Three patterns recur among merchants preparing their own submissions.
Document dumping. Forty files, no cover note, no index. The reviewer must reconstruct the business, and reconstructed submissions get returned.
Copying a competitor's pack. The documents that satisfied one provider for one business model say little about another; category and jurisdiction change the evidence required.
Treating approval as the finish line. Most post-approval problems — frozen payouts, descriptor flags, category reclassification — trace back to a mismatch between what was declared and what the account actually did in its first ninety days.
3. The proposed approach
Fourteen sections, in four parts. The order is the deliverable: it is the order a reviewer works through.
Part one — orientation. What the business does, expected volumes and ticket sizes, and a one-page summary of the three characteristics that usually raise questions, each with its answer. Stating the awkward facts first is what buys credibility for the rest.
Part two — the five assessment layers.
| Layer | The question it settles | Evidence it needs |
|---|---|---|
| Identity and KYB | who contracts, and who controls it | registration, UBO chain, director identity, signing authority |
| Business model and storefront | what is sold, to whom, at what price | live URLs, catalogue, pricing, shipping, returns, reachable support |
| Category and product risk | is anything restricted or regulated | product list mapped to category rules, licences where applicable |
| Early trading behaviour | does traffic match the declared model | first-90-day plan, ticket size, refund and chargeback expectations |
| Payout exposure and linkage | where the money lands, and who else lands there | settlement proof, entity-to-beneficiary linkage, shared-service disclosure |
Part three — packaging. A fixed evidence-pack template: cover note, issue summary, identity and business evidence, product and website evidence, order and fulfilment records, reconciliation, and a document index with capture dates. Fixed structure is what makes a pack reviewable in one pass.
Part four — operating after approval. A risk-triage table for the five review requests that arrive most often (identity and bank verification, website and policy review, product and category concerns, trading and payout reviews, account-relationship queries), each with the response shape and the evidence that closes it.
4. How it works in practice
The triage table is the part practitioners use weekly. A worked row:
Request: "Please explain the increase in monthly volume." What the reviewer is testing: whether the growth is organic or manufactured, and whether the fulfilment capacity exists. Response shape: volume by channel for three months, order-to-delivery time, refund rate, inventory or supplier evidence, and one sentence on what changed operationally. What closes it: numbers that reconcile with the bank statements already on file.
Four anonymised cases illustrate the judgement calls: an incomplete storefront, a genuine name variation between entity and bank account, growth that exceeded fulfilment capacity, and multiple accounts legitimately sharing a service provider. Each case shows the reasoning, not just the outcome — the reasoning is what a merchant can transfer to their own situation.
5. Trade-offs and failure modes
- It is merchant-side, so it can be gamed. The same structure that helps an honest merchant helps a dishonest one present better. That is why it is published as preparation guidance, not as a scoring rubric, and why it deliberately omits thresholds and rule identifiers.
- It cannot create evidence that does not exist. A merchant with no trading history cannot produce trading history; the pack has to say so honestly and offer a monitoring plan instead.
- Proportionality is a judgement. A five-layer pack is overkill for a low-risk micro-merchant; applying it uniformly is how onboarding becomes a bottleneck.
- Jurisdiction drift. Category rules and document expectations change; a playbook without a review date quietly becomes wrong.
6. What to measure
- First-pass approval rate for merchants who used the playbook versus those who did not.
- Round trips per application before decision.
- Days from submission to first settled transaction.
- Post-approval mismatch rate — declared versus observed descriptor, category and volume.
- Chargeback ratio at ninety days, which is the honest test of whether the pack predicted anything.
7. Conclusion
Preparation is not paperwork; it is answering the reviewer's questions before they are asked, in their order, with evidence that reconciles. A merchant who can do that is not asking for a favour at onboarding — they are handing over a decision that can be made in one pass.
Disclaimer. The views here are the author's own and describe a method, not a client engagement. Any figures or cases are synthetic or drawn from public sources. Provider thresholds, rule identifiers and client-identifying details are deliberately excluded.