Rigid price lists rarely match the reality of B2B. Every customer has different terms, quantities fluctuate, purchase prices move, and competitors change the situation often week by week. Anyone who still works with one price list across all customers gives away margin in one place and loses orders in another. This is exactly where dynamic pricing comes in: rule-based and AI-driven prices optimize the margin per customer and order in the logged-in shop instead of forcing an average price on everyone. The leverage is bigger than many assume. As a rule, a one percent higher price at equal volume lifts operating profit by around 8.7 percent (McKinsey), in B2B typically within a range of 6 to 14 percent (McKinsey). Data-driven dynamic pricing typically raises the margin by 5 to 10 percent (McKinsey) and revenue by 2 to 5 percent (McKinsey). At the same time online trade keeps growing: the e-commerce share of German retail is set to rise to up to 16.5 percent (HDE) in 2026. This article shows how mid-market firms implement dynamic pricing in Shopware traceably, lawfully and without discount chaos.
Key takeaways
- Dynamic pricing in B2B does not mean surprise prices but steering the existing customer-specific pricing logic on rules and data: customer groups, tiers, framework contracts and promotion or run-out logic behind the login.
- Price is the strongest profit lever: a one percent higher price lifts operating profit by around 8.7 percent (McKinsey), and in B2B by 6 to 14 percent (McKinsey) depending on the cost structure.
- Rule-based prices provide order and traceability, while AI suggestions take over in the grey areas. Data-driven price steering typically adds 5 to 10 percent (McKinsey) of margin.
- No good prices without clean data: purchase prices, terms, tier thresholds and framework contracts belong in one leading source of truth, usually the ERP, instead of parallel price maintenance.
- Every AI suggestion needs fixed price and margin floors plus human approval outside the corridor. In consumer business there is an additional duty to disclose personalised prices (Art. 246a EGBGB).
Why Dynamic Pricing Is Not Discount Chaos in B2B
Dynamic pricing has a mixed reputation in B2C, because there the public end price often fluctuates within hours. In B2B the starting point is different: prices are customer-specific anyway, they sit behind the login, and they follow negotiated terms, customer groups and framework contracts. Dynamic pricing here does not mean surprising customers with random prices but steering the existing pricing logic systematically, based on data and traceably. Instead of handing out discounts by gut feeling, a clear set of rules emerges: who gets which price, at which quantity, in which period and under which condition? These rules can be maintained, checked and proven.
The second difference is commitment. In B2B the price is part of a business relationship, not a shop window. A price that changes constantly without reason damages trust; a price based on clear rules and visible in the customer account strengthens it. That is why every dynamic pricing strategy in B2B needs a clean foundation of customer groups, roles and permissions. How to map this structure in Shopware is shown in detail in our article on customer groups, roles and permissions. Without this order, dynamic pricing really does become chaos – with it, it becomes a controllable instrument.
Dynamic Pricing Is Not a Surprise Price
What Sets Rule-Based Apart from AI-Driven Prices
At the start of every dynamic pricing strategy stands rule-based price finding. It translates commercial decisions into unambiguous if-then conditions: customer group A gets twelve percent off the product group, from 500 units the next tier applies, in the framework contract a fixed price holds. Such rules are transparent, verifiable and easy to explain. AI-driven pricing adds a layer on top: it evaluates historical orders, quantities, purchase prices, availabilities and seasonal patterns and suggests prices or price corridors that fit the situation. The AI does not replace the rules, it complements them with a data-driven recommendation where rigid rules are too coarse.
| Attribute | Rule-based pricing | AI-driven pricing |
|---|---|---|
| Basis | Fixed if-then conditions | Patterns from historical and current data |
| Strength | Transparency and traceability | Fine control and reaction speed |
| Maintenance | Manual upkeep of rules | Training and ongoing monitoring |
| Typical use | Customer groups, tiers, framework contracts | Price corridors, promotions, phase-out items |
| Risk | Rules age unnoticed | Recommendation without context |
| Approval | Directly applicable | Suggestion with human approval |
In practice the combination works best. The rule base creates order and reliably covers the majority of cases. The AI suggestion steps in where many factors interact and a rigid discount either gives away margin or endangers the order. Data-driven price steering typically delivers 5 to 10 percent (McKinsey) more margin, because it hits exactly these grey areas better than a blanket terms list. The decisive point is that the AI does not decide alone: it delivers a suggestion that a human checks and approves.
The Margin Lever: Why One Percent Moves So Much
Price is the most effective profit lever of all, more effective than volume growth or cost cutting. The reason is simple: at equal volume a higher price flows almost entirely through to the margin, because variable costs stay unchanged. That is why a one percent higher price lifts operating profit on average by around 8.7 percent (McKinsey), in B2B by 6 to 14 percent (McKinsey) depending on the cost structure. The reverse holds too: every unnecessarily granted discount point costs a multiple in profit. This is exactly where many B2B traders quietly lose money – not through one big mistake but through many small, unjustified reductions.
How large this gap is is shown by the realization rate of price increases: on average, companies actually push through only around 28 percent (Simon-Kucher) of planned price increases. The rest leaks away in exceptions, special terms and discounts that nobody tracks systematically. Data-driven price steering and clear approval limits win back part of that: even a structured control of terms and discounts typically brings more than 200 basis points (McKinsey), that is around two percentage points, of additional margin. The prerequisite is being able to see where the margin stands per line item, customer and order at all – a topic closely tied to the metrics and KPIs of a B2B shop.
The Core in One Sentence
Mapping Pricing Logic Cleanly in Shopware
Shopware open source brings the building blocks to map a dynamic pricing strategy cleanly. Customer groups and roles control who sees which prices; tier and quantity prices, customer-specific price lists and promotion logic can be configured deliberately and complemented via extension with AI suggestions. The important thing is to think of price finding as a coherent system rather than a loose collection of discounts. How price lists and tier prices are built in principle is covered in our article on price lists and tier prices in Shopware. Dynamic steering builds on top of that.
Customer Groups
Each customer group sees its negotiated terms. This is the base order on which every further price rule builds.
Tier Prices
Quantity tiers reward larger purchases and steer the margin per unit without anyone having to calculate by hand.
Framework Contracts
Fixed agreed prices and quotas from framework contracts apply automatically and run cleanly against the agreement.
Promotion Logic
Time-limited promotions and phase-out prices apply on a rule basis and end automatically, so no discount is left standing by accident.
AI Suggestion
For grey areas the AI proposes a price or corridor and justifies it from quantity, history and availability.
Approval
Suggestions outside defined limits pass into an approval before they become binding for the customer.
So that these rules are correct not only in the shop but across the whole business, prices, terms and availabilities must be reconciled with the leading system. In most mid-market firms the ERP is the price source. A robust connection ensures that the price shown in the shop matches exactly what appears in the order confirmation and invoice – a core topic of every ERP integration for the B2B shop. Without this reconciliation, price differences arise between channel and document that quickly lead to complaints and loss of trust in B2B. So that field sales, too, work with the right customer-specific prices on the move, the same rules apply in the field sales app for mobile and offline captured orders.
No Good Prices Without Clean Data
Every price rule and every AI suggestion is only as good as the data beneath it. Anyone who wants to steer prices dynamically needs reliable figures on purchase prices, costing bases, customer groups, quantities and availabilities. If these are missing or contradict each other across shop, PIM and ERP, even the best pricing logic produces wrong results. That is why dynamic pricing does not begin with an algorithm but with data quality and a clear source of truth for every figure. How much well-maintained product and price data relieves the entire process is shown in our article on product data and PIM in B2B.
- Purchase prices and costing bases up to date and maintained per item
- Customer groups, roles and negotiated terms clearly stored
- Tier limits, packaging units and minimum quantities kept consistent
- Framework contracts mapped with price, quota and term
- Margin per line item, customer and order analyzable at any time
- One leading source of truth, usually the ERP, instead of parallel price upkeep
A frequently underestimated point is duplicate data upkeep. Anyone who maintains prices in parallel in the shop and in the ERP sooner or later produces deviations. An automatic reconciliation via a robust interface prevents this and creates the basis for price rules to take effect reliably at all. Which data integration paths make sense for your shop is something we classify within our integrations and interfaces.
AI Suggestion, Human Approval
An AI that suggests prices is a tool – not a substitute for commercial responsibility. That is why every AI-driven price finding needs a clear approval process. If a suggestion moves within defined limits, it can apply automatically. If it leaves the corridor – for example because the AI recommends an unusually low or high price – it passes into an approval by sales or purchasing. This keeps control with the human while the AI does the groundwork. This mechanism resembles the approval processes for authorizations and budgets that many B2B shops already use for orders.
Define Limits Before the AI Starts
Pricing Lawfully and Traceably
Customer-specific prices are the norm in B2B and legally unproblematic: two business customers may receive different, negotiated terms. The case is different as soon as a shop also sells to consumers. Then, for personalized prices based on an automated decision, an information duty applies: the consumer must be informed that the price was personalized (Art. 246a EGBGB). This duty expressly concerns consumers, not the pure B2B relationship. Anyone running a hybrid shop for B2B and B2C must separate both worlds cleanly and fulfil the consumer duties where they apply.
Regardless of the audience: prices must be stated correctly and be traceable in the document. The German Price Indication Ordinance (PAngV) governs the requirements for price indication in consumer business, for example on total and unit prices. In B2B, traceability is paramount: every price should derive cleanly from list, customer group, tier and any promotion and be viewable in the customer account. As with other current obligations in online trade – such as the packaging duties of the PPWR in the B2B shop – clean documentation is the best protection. This is exactly what rule-based price finding delivers – it makes every price justifiable instead of letting it arise in the dark.
Consistency Between Shop, Quote and Invoice
Step by Step to a Dynamic Pricing Strategy
The path to dynamic pricing does not begin with a model but with transparency. First it must be clear where margin arises today and where it is lost. Then comes the rule-based order, then the data-driven fine control and finally the AI suggestion with approval. Those who keep this sequence build a system that stays explainable and grows with the business. A step-by-step approach also lowers risk, because each stage works measurably before the next begins.
- Make the actual margin per line item, customer and order transparent and find loss sources
- Map customer groups, tiers and framework contracts as a clear rule set in Shopware
- Reconcile prices, terms and availabilities with the ERP as the leading source
- Set price and margin floors as well as approval limits commercially
- Add an AI suggestion for grey areas, at first as a proposal with mandatory approval
- Measure the effect on real orders, refine rules and expand step by step
Those who make the margin visible and controllable from the start win twice: the contribution margin per order rises, and prices become justifiable to the customer instead of arbitrary. This pays off in a more stable margin and in less friction with purchasing and sales. Which pricing logic fits your assortment and your customers can be classified against comparable project approaches; our services around the B2B shop and a look at pricing show the frame. The concrete implementation is something we are happy to discuss in direct contact.