B2B pricing software and business rules engine: Commerce Rules Engine (CDL Workspace)
Management · Sales director
A conflict between two rules shows up, and the committee decides.
The auditing agent cross-checks the rules in force, simulates the impact and escalates the conflict. It rewrites no rule: it proposes, and a person confirms.
Illustrative scene with a real CWS Platform screen in a demo version. People, stores and on-screen data are fictitious.
In the sales session · B2B sales rep
Above his role's ceiling, the discount asks for a reason.
Within his approval limit, the discount goes straight through. Above the ceiling written in the panel, the order only moves on with a reason code.
Illustrative scene with a real CWS Platform screen in a demo version. People, stores and on-screen data are fictitious.
On the portal · Reseller buyer
The cart the agent built after hours, already at her price.
The purchasing agent builds it under the same rules as the sales rep: contract price, tax from the warehouse that serves her and the credit limit checked. She reviews and closes.
Illustrative scene with a real CWS Platform screen in a demo version. People, stores and on-screen data are fictitious.
At the counter · Building materials sales rep
The project quote, with freight and tax already in the total.
The price that applies to that builder, the freight and the state sales tax come in through the rule, before he sends the quote.
Illustrative scene with a real CWS Platform screen in a demo version. People, stores and on-screen data are fictitious.
Deterministic module
Your commercial policy, with its discount, credit and tax rules, lives scattered across a spreadsheet, the ERP and the sales rep's head. And now AI threatens to hallucinate a price on top of that. The Commerce Rules Engine is the business rules engine that turns those rules into an executable constitution, applied the same way every time, that the AI respects because it queries, it does not invent.
In short
Commercial policy is the set of rules that says who can sell what, at what price, with what discount and on what terms. On CWS Platform it becomes executable configuration in the Commerce Rules Engine (CDL Workspace), a business rules engine that also governs B2B pricing, with 1,029 configurable parameters, applied the same way to sales reps and to the AI Agents, and every exception sent for approval carries a Reason Code.
- The AI Agents consume the engine via MCP, under the same rules as a human rep: they suggest, and the human decides. Source: the 6 AI Agents.
- The price derived by Contextual Pricing obeys the same rules. Source: B2B pricing engine.
- The rep grants discounts inside the Approval Workflows, with a Reason Code. Source: sales rep software on the Sales Hub.
- Every discount has an author and a justification, and approval limits are by profile. Source: guided selling.
Rules as configuration, not code, is one of the criteria in the guide to choosing a B2B ecommerce platform. Three challenge pages show the rule at work: price list management, discount approval workflow and B2B payment terms.
What it is
The deterministic heart of the platform. It holds the business rules, approval limits, credit, tax compliance and agent permissions, as 1,029 configurable parameters that execute in code, at zero inference cost, and are exposed to the AI Agents via MCP (Model Context Protocol). The philosophy is configuration over code: commercial exceptions carry a Reason Code, critical actions stay in an audit trail, and new points for recording a reason can be built when the project calls for it.
The Commerce Rules Engine in 1 minute
The capability no one replicates
Governance by infrastructure
Approval Workflows by seniority, Pre-Submission Guardrails that block the unviable proposal before it reaches the manager, Discount Stacking Prevention and Reason Codes.
Who enforces the rule?
The rule blocks on its own, it does not depend on a memo.
The answer to Shadow AI
AI in silos, with no audit trail, becomes governed and auditable AI: the agent queries the same engine in real time, under the Autonomy Gradient (green, yellow, red).
Where does the market fall short?
Where the market falls short: fixed-price platforms have no native discount authority by seniority; in the CRM model, the equivalent requires building CPQ, approval workflows and pricing rules separately and custom.
Persona · Sales manager and discount approver
Only what truly needs a decision reaches me, and the reason comes with it.
In the Commerce Rules Engine (CDL Workspace), Pre-Submission Guardrails block the unviable proposal before it reaches the manager, and every exception sent for approval carries a Reason Code.
Persona · Sales rep
I negotiate within my own limit, and the rule is the same every time.
In the Commerce Rules Engine (CDL Workspace), each sales rep has their own discount limits, and commercial policy is applied the same way every time.
Approval matrix
The payment term and method each role can release.
The same approval matrix that holds each role's discount ceiling, by customer type and price list, sets the maximum payment term and the payment methods each role releases within its own approval limit.
One engine, the five dimensions
It is the module that governs price (derived by the B2B pricing engine), delivery, payment, quantity and composition, with the same mechanisms for human sales reps and for the AI Agents.
The way into the ecosystem, not only the rules engine5
This is what a single market term does not capture. The same environment that holds the rule also distributes access to it, and that is what lets an operation grow without scattering systems.
- Role-adaptive interface
- The same door serves distinct profiles.
What does each profile stop seeing?
Whoever enters sees no function they cannot perform and no data outside their own scope, and the adaptation is functional, not cosmetic.
- Structural base of the Seller Center
- The seller manages warehouses, balances, lead times and order status in a view of the same central interface, with no separate login and no system of their own.
- Structural base of the Marketplace Center
- The administrator's view governs catalog master data, approval thresholds, freight policy, commissioning and each seller's reach by geography, segment and catalog.
- Chained enablement
- The administrator enables the seller, and the seller enables their own team.
What does that spare?
The operation grows without anyone distributing credentials for five systems to five teams.
- Support with governed access
- The support team enters the specific environment to resolve an issue, by authority level, with no login switching, no request for a local password and no credentials in circulation.
Configurable commercial governance, from discount to access8
Commercial governance is B2B commercial policy applied to the B2B sales process every day, and the operating side of commercial excellence. Each company chooses how much control it wants and where: the rule decides what goes through, what goes to approval and what is blocked, and every decision is recorded.
- Three limits per sales rep
- Each sales rep has one limit that sends the order for approval, one that blocks the discount on the line and one that blocks the order, by category, store or price tag.
And what does that prevent?
The unviable deal stops before it reaches the manager.
- Tiered approval
- The order moves up level by level, with a reason and a history.
What does the approver see?
The approver sees what is pending or was denied below them and, if the company allows it, skips levels.
- Margin-based governance
- The rule looks at margin, not only at discount: below one minimum the order goes to approval, and below another it cannot be closed at all.
- Approval limits validated in your own system
- Companies that already keep their approval policy in the ERP or in their own engine can, store by store, have CWS Platform query that system before releasing the discount.
- Credit with a configurable decision
- When credit does not cover the order, the company decides by rule what happens: block it, send it for approval or release it.
- Who gets in and who buys
- The portal can stay closed to registered customers, open with registration under approval or be restricted to companies or to rural producers, and products under regulated sale require an identified customer.
- Security and audit trail
- Access by profile and by branch, 2FA for administrative users and an audit trail of who changed payment credentials or marketplace splits and who accessed customer personal data.
- Two-layer enablement
- CWS Platform enables each capability for the operation, and the company decides when to turn each parameter on.
And what about rollout?
The rollout moves in stages, without exposing from day one what nobody will use.
AI Agents
What each AI Agent does alone, proposes, or never does.
The Autonomy Gradient sets, action by action, what each of the 6 AI Agents executes alone, what it proposes for a human to confirm and what it never does. The zone is company configuration, not the agent's decision.
How the AI Agents work →
In operation
A distributor with millions of price rules and approval limits by seniority, today with no record of who granted which discount, now has every concession with an author and a justification. On integration, here is what customers say publicly:
"Excelente, estávamos a quase 2 anos tentando implantar uma solução B2B, com a CWS, implantamos em 60 dias."
Frequently asked questions
What governs negotiation on the CWS Platform?
The Commerce Rules Engine (CDL Workspace), with 1,029 configurable parameters. It is governance by infrastructure: whatever falls outside the Approval Workflows goes to approval or is blocked, according to the rule the company configured, without relying on instructing the sales rep or prompting the AI.
Is the rule written by the AI in the CDL Workspace?
No. The AI Agents query the engine via MCP, audit, detect conflicts and simulate impact, but never alter a rule directly. They suggest; the human decides.
How does it integrate with my ERP without becoming lock-in?
As an orchestration layer on top of the ERP, not a replacement, through APIs and webhooks. The ERP remains the vault; the Commerce Rules Engine is the commercial manager. Publicly associated integrations include SAP, TOTVS, Senior, Sankhya and Microsoft Dynamics.
What is a business rules engine in B2B commerce?
It is the system that holds the commercial policy, the rules that define who can sell what, to whom, at what price, with what discount, on what terms and with what credit, and applies them on every order. In most companies that policy lives in a document and is applied from memory; in the Commerce Rules Engine it becomes configuration the system executes.
How do you make commercial policy auditable?
Take the rule out of the document and put it in the system that processes the order, with a record of every exception. In the Commerce Rules Engine, a discount outside the approval limit does not go through without approval, every request for approval and every refusal carries a Reason Code, and critical actions stay in an audit trail, so each exception has an author, a date and a reason.
Is the Commerce Rules Engine B2B pricing software?
It is the part of B2B pricing software that holds the rules price has to obey. The price itself is derived by Contextual Pricing (Pricing Engine); the Commerce Rules Engine applies the approval limits, Discount Stacking Prevention and Reason Codes on top of it, the same way for human reps and for the AI Agents.
What should B2B pricing software control besides the price?
The discount on top of it, who can grant it and why. In the Commerce Rules Engine, a discount outside the approval limit does not go through without approval, Discount Stacking Prevention keeps discounts from piling up, and every concession carries a Reason Code, the same way for sales reps and for the AI Agents.