Latin American Pricing Complexity #256
martinferreira90
started this conversation in
Ideas
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Latin American Pricing Complexity — The "Displayed Price Is Never the Real Price" Problem
Context
This discussion is motivated by real-world experience building a UCP Proof of Concept and encountering a fundamental disconnect between how UCP represents prices today and how commerce actually works in Latin America. The issues described here affect Argentina, Brazil, Mexico, Colombia, Chile, and most LATAM markets.
The Problem
When a buyer in Latin America (LATAM) searches for a product — say, "zapatillas deportivas" (sneakers) — the listed price is almost never the price they will actually pay. Unlike markets where the sticker price is the final price, in LATAM the buyer's actual cost depends on a combination of factors that are only revealed deep into the checkout process:
The result: the same ARS 150,000 sneaker has wildly different effective costs depending on who the buyer is and how they pay.
A concrete example
A buyer searches "Nike Air Max" on a marketplace. The listing shows ARS 150,000. But the actual options are:
The "real" price ranges from ARS 127,500 to ARS 283,500 — nearly a 2.2x difference — for the same product. And the buyer doesn't see this until they're already in checkout, selecting payment methods.
Why this matters for agentic commerce
An AI agent helping a LATAM buyer should be able to say:
Instead of:
This is the difference between a useful agent and a glorified search engine. In LATAM, the pricing IS the value proposition — installments are not a financing afterthought, they're the main motivation for people to choose what to buy. That's why we think being able to show the effective cost at the search phase is vital for this market.
What makes LATAM different
1. Installments are not a financing product — they're a pricing mechanism
In the US/EU, installments (BNPL) are a separate financial product layered on top of a fixed price. In LATAM, the merchant subsidizes interest-free installments as a sales tool. The merchant pays the bank a fee (e.g., 8-15% of the sale) so that the buyer can pay in 12 monthly installments with zero interest. This is baked into the product's economics — it's not optional, it's the default way people shop.
Key differences:
2. Bank promotions are ubiquitous and time-sensitive
In LATAM, banks run aggressive promotions with merchants:
These are not merchant discounts — they're trilateral agreements between the bank, the payment processor, and the merchant. The bank subsidizes part of the discount, the processor may waive fees, and the merchant absorbs the rest.
Digital wallets like MODO add yet another layer: MODO acts as an aggregator that surfaces bank-specific promotions through a unified payment experience. A buyer paying through MODO with their Visa Banco Nacion card on a Thursday gets 15% cashback — but that same MODO wallet with a Mastercard HSBC card on a Monday gets nothing. The benefit depends on the intersection of wallet + card issuer + card brand + day of the week, a combinatorial explosion that no current UCP schema can represent.
3. Loyalty tiers span multiple actors
Unlike a simple "Gold member 20% off" from one merchant, LATAM loyalty tiers create cross-cutting benefits:
The benefit isn't from the merchant — it's from the platform, the bank, or the super-app, and it applies across merchants.
Gap Analysis: Where UCP Falls Short Today
Gap 1: No representation of installment plans at catalog/search time
UCP's catalog schemas (
price,list_priceon variants) represent a single price point. There's no way to express:The
payment_handlerschema defines available instruments but has no concept of installment plans or payment-method-specific pricing.Gap 2: Benefits can come from any actor — or combinations of them
UCP's
discountextension supports merchant-side discounts (codes and automatic rules) at checkout time, but has no model for benefits like cashback, loyalty points, or zero-interest installments. In LATAM, these benefits don't come from a single actor — they can originate from any participant, or from agreements between multiple participants:The protocol needs to support benefits from any single actor and from multi-party agreements, with clear attribution of who funds each part.
Gap 3: No concept of payment-method-specific pricing
UCP currently assumes the price is independent of the payment method. In LATAM:
The price literally changes based on what you're paying with, and this happens at the merchant level, not the CP level.
Gap 4: No temporal/conditional promotion model for bank promotions
Bank promotions like "20% off on Tuesdays with Banco Galicia" need:
UCP currently has no promotion or temporal conditions model at all — the
discountextension only handles codes and automatic discounts at checkout time, with no temporal, bank-specific, or multi-party promotion support.Gap 5: Cross-merchant loyalty tiers
UCP has no loyalty tier model today — it's listed as a future initiative in the roadmap ("Loyalty & Member benefits"). When designed, it will need to support tiers that aren't Business-specific: Falabella CMR Premium or Banco Galicia Select are platform-level or CP-level tiers that affect pricing across all merchants. There needs to be a way to represent "this buyer has Platform tier X, which means they get Y installment plans everywhere."
What UCP Needs to Address
This isn't a fully-baked proposal — it's meant to start the conversation. But here are the areas that need design work:
1. Installment plan representation
Products need to carry installment plan information, at minimum:
2. Payment-method-specific pricing at catalog time
The catalog/search response needs to express that the effective price varies by payment method. This could be:
pricing_optionsarray on catalog items showing different price points per payment method type3. Trilateral promotion model
Promotions in LATAM involve three parties. The schema needs to support:
4. Recurring temporal conditions
Bank promotions run on schedules: "every Tuesday and Wednesday", "first week of each month", "every Cyber Monday". Simple date ranges don't cover this.
5. Platform-level and CP-level tiers
The tier system needs to support tiers that aren't Business-specific:
Example: What an Agent Should Be Able to Do
For this to work through UCP, the protocol needs to carry all this information in a structured, machine-readable way — from catalog search through checkout.
Questions for the Community
Scope: Should installment plans be a core protocol concern or an extension? Given that installments drive >60% of e-commerce transactions in LATAM and are growing in other markets (BNPL), we argue it should be core.
Where do installment plans live? Options:
Indicative vs. authoritative: Installment plans shown at search time would be indicative. The authoritative plan is confirmed at checkout. How much tolerance for discrepancies is acceptable?
Interest rate disclosure: LATAM regulations (Argentina's Ley de Defensa del Consumidor, Brazil's CDC) require showing CFT (Costo Financiero Total) / CET (Custo Efetivo Total). Should UCP mandate structured interest rate data, or is that a Business-level concern?
Multi-currency installments: In Argentina, some products are priced in USD but installments are charged in ARS at the exchange rate of each installment date. How should UCP handle installments where the currency of the installment differs from the product currency?
Backward compatibility: UCP already supports simple discount codes and automatic discounts at checkout time. How do we extend the protocol to cover the trilateral installment model without overcomplicating the simpler use cases?
Beta Was this translation helpful? Give feedback.
All reactions