Electrical engineering, drive technology, cables and wiring, seals, hydraulics: in many technical assortments an item is not a fixed entity but the result of ten or more attributes. Voltage, frame size, protection class, connection, length, material -- every choice constrains the next, and not every combination is technically possible. Such variant-rich products are hard to map in a classic web shop: a dropdown list with thousands of theoretically possible combinations overwhelms the buyer and produces incorrect orders. The way out is a rule-based product configurator with guided selling in front of it. It leads the buyer to the right item through question logic, permits only technically valid combinations and generates price and quote end to end. That business customers are ready for this is clear from the numbers: 73 percent (McKinsey) of B2B buyers now place orders of USD 50,000 or more in a single online transaction. Anyone selling variant-rich technical products can only unlock this self-service with a configurator and guided selling.
Key takeaways
- Technical assortments break the flat catalog: six attributes with five options each produce more than 15,000 arithmetic combinations, only some of which can actually be manufactured. As an item list this is neither maintainable nor sellable and it produces wrong orders.
- A configurator's rule engine works with four rule types: dependency rules force follow-up choices, exclusion rules block impossible combinations, enforcement rules set defaults, and pricing rules tie every attribute to a price contribution.
- Guided selling asks about application, environment and standard and translates that need into technical attributes; the configurator then checks validity, price and orderability. A filter cannot do this because it assumes the buyer already knows the attributes.
- The PIM supplies attributes, permitted values, rules and data sheets; the ERP supplies base prices, attribute surcharges, quantity tiers and availability and receives the order including its bill of materials. Without both connections the flow stays a calculation exercise.
- CPQ requires the valid configuration to flow into a quote with frozen terms and an expiry date, without any manual re-entry. Where that step is missing, buyers configure online and still place the order by phone.
- The hardest part of such projects is not the technology but making explicit the rules that currently live in the heads of the sales team. Starting iteratively with the most variant-rich product group lowers risk and creates a blueprint for further assortments.
Why Variant-Rich Assortments Fail Online
B2B online trade among manufacturers and wholesalers in Germany reached a volume of 509 billion euros (ECC Cologne) in 2024, up 7 percent (ECC Cologne) year over year. The online shop is the most-used channel: 93 percent (Bitkom) of retail companies with online trade sell through their own shop on their website. Yet this is exactly where the classic catalog hits its limit as soon as a product is configurable. A servo motor, a made-to-order cable assembly or a hydraulic cylinder does not exist as a single item but as an attribute space: six attributes with five options each yield more than 15,000 combinations arithmetically. As a flat item list, that is neither maintainable nor sellable.
The result is a process that is digital in intent but analog in practice: the buyer cannot find the right variant, calls or emails, the sales desk clarifies questions and creates a quote manually. This quote ping-pong fits poorly with real buying behavior. B2B buyers today use an average of ten interaction channels (McKinsey) across the buying journey, twice as many as in 2016, and more than half expect to switch seamlessly between them (McKinsey). Anyone selling variant-rich technical products only via phone callbacks breaks this expectation at the very point where the purchase decision is made. For how this channel question fits strategically, see the article on marketplace or own shop as a platform strategy.
Typical Symptoms of a Missing Configurator
What a Product Configurator Delivers in B2B
A product configurator is not a prettier filter but a rule-based selection flow. The buyer chooses attribute by attribute, and with every choice the system automatically blocks all options that do not technically match the selection so far. The result is not a wish list but a valid, manufacturable configuration with a clear item identity, correct net price and the basis for a binding quote. The configurator thus takes on the role that the experienced sales desk plays in the analog process -- around the clock and without waiting time.
Valid Combinations Only
Dependency, exclusion and enforcement rules permit only technically manufacturable variants. Impossible combinations cannot even be selected in the first place.
Price in Real Time
Every attribute carries a price contribution. Base price, attribute surcharges and quantity tiers immediately yield a traceable net price -- based on the terms from the ERP.
Seamless Quote
The valid configuration converts into a quote or order with a single click, including bill of materials, drawing and item number -- without re-entering anything.
Variants from the PIM
Attributes, permitted values and rules come from the product information system. The configurator thus stays consistent with catalog, data sheet and sales.
Avoid Incorrect Orders
Because impossible combinations are excluded, follow-up questions, returns and rework decrease. The process becomes more reliable for both sides.
24/7 Self-Service
The buyer configures whenever they want, even outside business hours. Sales is relieved of routine requests and gains time for complex cases.
Guided Selling: From Requirement to Item
Not every buyer knows the technical designation of the item they need. Guided selling starts exactly here: instead of asking for attributes, it asks about the need. What is the product used for? What environmental conditions apply? Which standard is required? From the answers, the system derives the matching technical attributes and leads into the configurator, which then refines the selection. Guided selling thus translates the customer's application language into the manufacturer's product language.
This need-oriented entry matches the way B2B customers want to decide. McKinsey describes a stable rule of thirds: at any point in the buying journey, roughly one third of customers want in-person contact, one third remote interaction and one third full self-service (McKinsey). Guided selling serves the self-service third without excluding the others: anyone who gets stuck is handed over to the right contact person without losing the configuration state so far.
- Needs questions instead of attribute lists: the entry is via application, environment and standard, not via technical jargon.
- Step-by-step narrowing: every answer reduces the possible attribute space and makes the next question more relevant.
- Explanatory help: each question can show tooltips, diagrams or standard references so that non-experts can choose confidently.
- Handover to the configurator: at the end of the question logic there is a pre-filled configuration that the buyer only has to confirm or fine-tune.
- Advisory fallback: for complex requirements a click leads to the sales desk -- with the full context of the selection so far.
Guided Selling Is Not a Filter
Configurator and Guided Selling: the Distinction
Guided selling and product configurator are often mentioned in the same breath but solve two different tasks. Guided selling leads to the right product class and the matching base attributes. The configurator ensures that the derived selection is technically valid, priceable and orderable. In practice, the two interlock: guided selling is the guidance, the configurator the safeguard. The following comparison frames the roles.
| Aspect | Guided Selling | Product Configurator |
|---|---|---|
| Starting point | Customer's application need | Technical attributes of the product |
| Guiding question | What is it used for? | Which combination is valid? |
| User group | Also non-expert buyers | Expert and guided users |
| Core logic | Question logic and recommendation | Dependency and price rules |
| Result | Pre-filled product class | Valid item with price |
| Handover | Into the configurator | Into quote and order |
Rule-Based Logic: How Valid Combinations Emerge
The heart of a configurator is the rule engine. It knows four basic types of rules. Dependency rules enforce a selection once another has been made -- whoever chooses 690 volts necessarily needs a higher protection class. Exclusion rules block combinations that do not technically fit together. Enforcement rules set default values when a selection allows only one sensible continuation. Price rules, finally, link every attribute to a price contribution. From the interplay of these rules, a valid, priced configuration emerges in real time with every choice.
{
"product": "servo-motor-series-x",
"attributes": [
{ "key": "voltage", "type": "enum", "values": ["230V", "400V", "690V"] },
{ "key": "protection", "type": "enum", "values": ["IP54", "IP65", "IP67"] },
{ "key": "connection", "type": "enum", "values": ["terminal-box", "plug-m23"] }
],
"rules": [
{ "if": { "voltage": "690V" }, "require": { "protection": ["IP65", "IP67"] } },
{ "if": { "connection": "plug-m23" }, "forbid": { "protection": ["IP54"] } },
{ "price_add": { "when": { "protection": "IP67" }, "amount_eur": 84.00 } }
],
"result": { "valid": true, "net_price_eur": 1284.00 }
}Technically this is a constraint problem: with every choice, the engine checks all stored rules and determines which values of the remaining attributes are still permitted. Invalid options are greyed out or removed, open mandatory attributes are shown. This check has to be fast and traceable so the configuration stays fluid. The rule base itself belongs in the hands of the technical department: it describes the company's technical knowledge and is maintained, not programmed. Sound B2B e-commerce consulting at the start of a project clarifies which rules even exist and where they currently sit only in the heads of the sales team.
Rules Are Company Knowledge
Integration with PIM and ERP
A configurator is only as good as the data it works on. Two systems are decisive. The PIM (Product Information Management) provides the attributes, permitted values, dependencies and accompanying content such as data sheets and drawings. The ERP provides prices, terms, availability and finally receives the order. Only the connection of both worlds turns a selection flow into a sellable process. How structured product data is prepared for this is described in the article on product data and PIM in B2B; the technical connection itself is the topic of ERP integration.
The data connection also pays into personalization. Already 34 percent (Statista) of B2B companies in Germany used AI-supported product recommendations to make online purchasing more efficient; the most effective personalization method in B2B e-commerce is considered to be personalized search within the shop (Statista). On this basis, a data-driven configurator can suggest matching accessories, spare parts or frequently co-ordered variants. This is exactly where the configurator connects with aftersales -- a well-designed digital spare-parts catalog uses the same attribute and rule base for reordering.
- PIM as the source for attributes, permitted values, rules and accompanying documents
- ERP as the source for base prices, attribute surcharges, quantity tiers and credit limit
- Real-time check of availability and delivery time during configuration
- Writing back the finished configuration as an order including bill of materials
- Consistent item identity across configurator, catalog, quote and document
From Configuration to Quote and Order (CPQ)
Configure, Price, Quote -- CPQ for short -- describes the seamless chain from configuration through pricing to the quote. This chain decides whether variant-rich technical products are actually sold online or merely configured and then ordered by phone after all. That large volumes are handled digitally is no vision of the future: according to Forrester, in 2025 more than half of large B2B purchases of one million US dollars or more will be processed through digital self-service channels, which explicitly include the vendor's website (Forrester). McKinsey also confirms the willingness to place large online orders: 20 percent (McKinsey) of B2B buyers are willing to place orders over one million US dollars purely digitally.
For implementation this means: the valid configuration must flow into a quote without a break -- with frozen terms, a validity date and the option to accept or reject line items. This transition is the same one that carries classic quote management in quote-to-order. The price basis in turn comes from clean pricing logic, as described in the article on price lists and tiered pricing. This way, the configuration does not become a PDF draft but a binding, orderable quote.
A configurator that only configures is a toy. Only when configuration, price and quote interlock without a media break does variant diversity become a sellable self-service.
The Benefit: Fewer Incorrect Orders, More Self-Service
The economic core of a configurator lies in error avoidance. Because technically impossible combinations cannot even be ordered, follow-up questions, wrong deliveries and returns on configurable items decrease noticeably. In our projects, with cleanly modeled rules we observe a clear reduction in order-related follow-up questions in the low double-digit percentage range (project experience), depending on assortment and starting point. Every avoided incorrect order saves twice: return transport and rework on one side, lost sales-desk time on the other.
The second benefit is relieving sales. Where standard configurations run in self-service, the sales desk gains time for genuinely complex cases and active advice. This also matters competitively: 67 percent (Bitkom) of retail companies see growing price pressure from cheap marketplaces. Anyone who differentiates through advice, technical depth and a smooth configuration process escapes this pure price comparison. A configurator is thus not only an efficiency tool but a differentiation tool.
- Fewer returns and rework because impossible combinations are excluded.
- Shorter response times because the quote arises directly from the configuration.
- A relieved sales desk that hands routine requests over to self-service.
- Higher data quality because every order corresponds to a valid, unambiguous configuration.
- Differentiation through depth of advice instead of pure price.
Pragmatic Entry
Implementation with Shopware Open Source
On the basis of Shopware Open Source, a configurator and guided-selling flow can be implemented independent of any single vendor. The open architecture makes it possible to integrate a rule engine, adapt the selection flow to your own assortment and cleanly connect the valid configuration to the existing checkout and quoting function. Because the source code is open, the merchant stays independent and can adapt the rule logic exactly to their technical requirements instead of submitting to a rigid standard. We implement exactly such flows in custom Shopware development and connect them via integrations to ERP and PIM. For sectors such as electrical engineering and electronics, this is the lever that makes variant-rich technical products sellable online in the first place.
In practice, an iterative approach proves itself. First a product group is modeled, the rule base is gathered together with the technical department and the configurator is connected to PIM and ERP. Then follow the guided-selling entry, quote conversion and further assortments. From more than 50 B2B projects (project experience) we know that this step-by-step build reduces risk and delivers visible benefit early. Whether a configurator, a full B2B portal with self-service or initially just a guided-selling flow is the right entry point, we clarify in a personal conversation.
Sources and Studies