Structured product distribution is a composite decision. Product eligibility (retail vs accredited, complexity band, K&E level), suitability (risk tolerance, loss capacity, investment objectives), cross-border rules (RM location vs client residence, solicited vs reverse enquiry), and concentration (single name, underlying class, aggregate SP allocation) - all mix in every offer.
Today that composite lives in a mix of code, workflows and spreadsheets. Every product launch, every new jurisdiction, every regulatory update means finding and updating the logic in multiple places.
What Knowledge does for a wealth manager
Knowledge holds the composite decision in one governed layer. Your OMS, your mobile RM app, your compliance dashboard and any AI copilot all consult the same source.
One decision, many callers. The same "can I offer this product to this client" question, asked from the OMS at trade time, from the mobile app pre-trade, from the RM copilot mid-conversation, and from the compliance dashboard for audit - one policy source answers all four.
Four seeded policies, thirteen rules.
| Policy | What it governs |
|---|---|
| Product eligibility | Retail vs highly-complex product, large notional retail approval, target-market alignment |
| Client suitability | K&E gate on complex products, risk-tolerance mismatch escalation, documented reverse-enquiry allow |
| Cross-border distribution | Solicited outreach into restricted jurisdictions blocked, booking-centre mismatch above threshold escalates |
| Portfolio concentration | Single-name post-trade above 30% escalates, above 50% blocks, aggregate SP allocation caps on conservative mandates |
Each threshold is a pattern shipped with a realistic default. The bank's compliance officer calibrates the exact value against firm policy.
The four canonical decisions
An RM copilot or an OMS asks Knowledge one of four questions.
| Question asked | Verdict Knowledge returns |
|---|---|
| Can I offer this product to this client ? | Blocks retail on highly-complex products ; requires approval on large notionals |
| Is this trade suitable for this client ? | Gates complex products against K&E levels ; escalates risk-tolerance mismatches |
| Cross-border : can I solicit this client from this location ? | Blocks solicited outreach into restricted jurisdictions ; allows documented reverse enquiries |
| Portfolio concentration : is this trade within limits ? | Escalates single-name concentration above 30% ; blocks above 50% |
Each verdict comes with the cited rule and a replayable audit key.
What the pack ships
| Component | What it is |
|---|---|
| Scope schema | The vocabulary the decision layer uses (product complexity, risk rating, client experience level, solicitation type, booking centre, RM location, post-trade exposure) |
| Four policies + thirteen rules | Realistic default thresholds, ready for the bank's compliance officer to calibrate |
| Reference integration | A working script showing an RM copilot calling /resolve for the four canonical decisions |
| Operator playbook | The runbook to install, calibrate and rehearse the pack |
What the bank owns, what Asplenz ships
Asplenz ships the ontology and the pattern rules with realistic defaults. The bank's compliance function owns the interpretation of every threshold - what does "large notional" mean at this firm, which jurisdictions are restricted, which risk-tolerance mismatch triggers escalation. The pack does not ship regulatory interpretation ; it gives the bank a working shape to calibrate.
Deployment options
The wealth pack inserts into an existing wealth stack in one of several ways.
| Insertion point | How it works |
|---|---|
| Behind the OMS (gate) | The OMS calls Knowledge before routing an order. Blocking verdicts stop bad trades pre-execution |
| Alongside a legacy engine (shadow → selective routing) | Knowledge runs in shadow, discrepancies surface for review, then transitions to primary for the SP scope only |
| Greenfield decision layer (primary) | For a new product line or new market entry - no legacy to work around |
Read how Knowledge fits your stack
What comes next
| Read next | Why |
|---|---|
| How Knowledge works | The mental model, the API contract, the audit surface |
| AI agents | How an RM copilot or trading agent calls Knowledge as a tool |
| Pilot | Run one of the four decisions in shadow for 4-8 weeks, measure decision agreement against your existing logic |