Discounting, Multi-Currency & Subscription Metrics

Discounting, currency, and recurring metrics are where "the price" stops being a single number and becomes a set of numbers that must stay consistent across quote, contract, and bill. A quote carries a negotiated price in one currency; a contract fixes that price for a term; billing rates real usage and issues an invoice, sometimes in a different currency and always at a different point in time. Most commercial disputes — a customer disputing an invoice, finance unable to reconcile booked revenue against billed revenue, a renewal that quietly re-prices itself — trace back to divergence between those three numbers, not to any single system being wrong in isolation.

Discounting Logic

Not every reduction from list price is the same kind of discount, and treating them as interchangeable is how governance breaks down. The four types below differ in how they are triggered, how long they last, and who is accountable for the margin given away.

Approval Thresholds & Governance

Approval-threshold discounting exists to solve a specific problem: without it, discount authority is implicitly unlimited for whoever can edit a quote. Tiering approval to discount depth means the size of the margin given away determines how much scrutiny the deal receives before commitment — a defensible control, provided the thresholds are set against actual margin impact rather than an arbitrary percentage that has never been reconciled to cost.

Multi-Currency Layers

A single deal can legitimately carry four different currency roles, each fixed at a different point in the lifecycle and each capable of drifting from the others. Multi-currency support solves the problem of selling and billing across markets without duplicating the entire catalog per currency — it does not, by itself, guarantee that the four roles stay reconciled.

Subscription & Usage Metrics

CPQ and the catalog define what a metric is and how it is priced. They do not measure it. That boundary is the single most consequential ownership line in this section, because a metric that is perfectly defined at quote time can still never be accurately billed if the systems downstream have no means of measuring it.

• Recurring charge — a fixed periodic fee (monthly, annual) attached to a subscription line, defined once in the catalog and repeated on every billing cycle until the subscription changes
• Usage / consumption metric — a unit of measured consumption (minutes, gigabytes, API calls, sessions) for which CPQ defines the rate structure, but which CPQ itself never observes or counts
• Entitlement / allowance — a bundled quantity included in the subscription before overage charges apply, defined as a catalog parameter and tracked against actual consumption entirely outside CPQ

CPQ defines the metric; billing measures and rates it. This is the source-of-truth line: CPQ and the catalog own the definition (what counts, what the allowance is, what the overage rate is), while billing and usage mediation own the measurement (what actually happened) and the rating (what that measured usage is charged at). A metric can exist correctly in the catalog for years without a single accurate usage record ever reaching billing, if the underlying network or system feed that should populate it was never built or was decommissioned without anyone updating the catalog.

What This Does Not Solve

Discounting logic does not fix a weak value proposition — a well-governed approval workflow makes bad discounting more disciplined, not more necessary; if a product only sells at a deep discount, the discount is compensating for a pricing or positioning problem the workflow cannot see. Multi-currency support does not remove foreign-exchange risk — it defines where a rate is fixed, but the business still absorbs the gap between the rate fixed at contract signature and the rate prevailing when revenue is actually recognised or repatriated. And defining a subscription or usage metric in CPQ does not guarantee the network, platform, or service actually emits the usage records needed to measure it — a metric with no data feed behind it is a pricing fiction that only becomes visible the first time someone tries to bill it.

When It Becomes an Anti-Pattern

What Breaks First

Currency truth divergence between quote, contract, and billing breaks first. A quote issued in one currency, a contract that fixes an FX rate at signature, and a billing engine that rates usage at whatever rate applies on the invoice date can each be individually correct and still disagree with each other by the time an invoice reaches the customer. Under normal deal volume this drift is small and gets absorbed as rounding. Under scale, across many markets, or with long-term contracts spanning periods of real FX volatility, the gap compounds — finance cannot reconcile booked revenue to billed revenue, customers dispute invoices that do not match the number they negotiated, and nobody can say with confidence which of the four currency roles is authoritative for a given dispute, because no single system owns the reconciliation between them.

TMF Mapping

• TMF620 Product Catalog Management — ProductOfferingPrice models recurring, one-time, and usage-based price types, and price alterations express discount and promotional logic
• TMF622 Product Ordering — carries the agreed, priced configuration, including any negotiated override, from an accepted quote into fulfilment
• TMF635 Usage Management — standardises the shape of a usage record that feeds rating, though the rating and currency-conversion logic that turns a usage record into a rated, invoiced charge commonly sits outside strict TMF scope, implemented as vendor-specific billing and mediation logic
• This gap is honest, not a modelling failure: TM Forum standardises the price structure and the usage record shape, not the FX conversion or rating engine that applies them