B2B spare parts portal: digital sales for machine builders

In short: for a machinery manufacturer, digitizing sales means putting four things online: a B2B portal where customers order the right spare part, a configurator that produces consistent quotes, an automated flow for quote requests, and the use of installed-machine data in after-sales service. Start with spare parts, where orders are repetitive and easy to measure. The Data Act, the Machinery Regulation and the Cyber Resilience Act make machine data, documents and software part of the service a manufacturer has to organize.

Key takeaways

Spare parts are a machine builder's most repetitive order, so they are the first candidate for self-service.
A B2B portal works when it starts from the machine's serial number, not from a generic catalog.
Configurators and automated quoting cut technical errors in offers first, response times second.
Under the EU Data Act, customers are entitled to the data their machine generates: after-sales service has to earn their loyalty.
Treat the spare parts catalog as structured data (serial number, BOM, compatibility, customer part numbers), not as a PDF to publish.

A machine builder sells equipment that runs at the customer's site for years. The machine sale is rare; parts, service and consumables are an ongoing relationship that often still runs on phone calls, nameplate photos and PDF exploded views. That is where digitization pays off.

According to Istat, Italy's national statistics institute, 16.4% of Italian companies with 10 or more employees used at least one AI technology in 2025, up from 8.2% in 2024; among large companies the figure reaches 53.1%. Machines already produce data through their sensors: the goal is not to add AI everywhere but to connect that data to sales and service. The rules of B2B e-commerce apply, with one complication: the product is technical and every customer runs a different machine.

What is a B2B spare parts portal and what should it do?

A B2B spare parts portal is a login-protected area where a business customer finds their own machines, the parts that fit each one and their own prices, and orders without going through the sales office. Every user belongs to a company account with its own price lists and permissions.

Search by serial number: from the installed machine's exploded view straight to the part number.
Customer-specific pricing: price lists and discounts pulled from the ERP, with up-to-date stock availability.
Quick reorder from part numbers or past orders, for consumables and wear parts.
Roles: who builds the cart on the shop floor and who approves the order in purchasing.
Documents in one place: manuals, safety data sheets, invoices, shipments.

The foundation is the bill of materials (BOM), the structured list of a machine's components: linked to the serial number, it tells the customer exactly which part their unit uses. Mainstream platforms cover the commercial side: Shopify, for example, sets up B2B customers as companies and lets you personalize pricing, products, payment and shipping methods for each one. The real work lies in the data and in B2B payment terms (see how B2B payments are changing).

There is a competitive reason too. Parts buyers also consider non-OEM aftermarket suppliers, which are often cheaper and easier to source. According to commercetools, a commerce platform vendor, downtime in industrial settings often costs far more than the part itself, so the supplier that makes it easiest to identify the right part and verify fitment wins. The portal's first job is to stop losing orders to friction, before it sells more.

How do you build a digital spare parts catalog?

Start from the technical data you already have (bills of materials, exploded views, ERP item master) and turn it into a structured catalog. Publishing exploded-view PDFs is not enough: customers need to search, filter and check compatibility with their own machine.

A useful catalog links every part to:

The machine, assembly and sub-assembly it is fitted to, by serial number and version.
Technical attributes: dimensions, material, compatibility, lifecycle status (active, superseded, discontinued).
Alternative part numbers: the customer's own code in their system and the number that superseded an old one, so search finds the part even from a reference that is not yours.
Complementary parts: seals, lubricants and small hardware needed with the part, offered as maintenance kits.

These are the same elements commercetools lists for a modern parts catalog, noting that a manufacturer's part number rarely matches the one in each customer's system. A structured catalog is also readable by software and AI agents, so it is worth building properly even if advanced automation is not on your roadmap yet. The work sits mostly with engineering and technical documentation, not web development: decide who updates BOMs and part supersessions when the machine changes.

Which sales processes should you digitize first?

The repetitive ones, with clear rules and a visible cost in office hours: spare parts, standard quotes and documentation. Configuring special machines comes later, once master data is clean.

Process	Without a system	With a digital system	What to measure
Spare parts orders	Email with a nameplate photo, manual part number lookup	Portal with serial number search and customer pricing	Share of self-service orders, wrong part numbers
Quote requests	Generic email, data retyped into the CRM	Structured form, CRM updated automatically, draft quote	Time to first response, quotes won
Machine configuration	Spreadsheet, verbal checks with engineering	Configurator with compatibility and pricing rules	Revisions per quote, errors found in production
Documentation	PDFs sent on request	Manuals and exploded views by serial number in the portal	Document requests to the service desk
Field service	Phone calls, untracked interventions	Tickets tied to the machine, service history, operating data	Response time, downtime hours

The portal depends on warehouse and shipping (see logistics digitization for SMEs): if the availability shown is unreliable, self-service creates complaints.

How do you connect the portal to the ERP and to customers' procurement systems?

The ERP remains the source of truth: the portal reads items, price lists, stock and payment terms, and sends back structured orders that nobody has to retype. Before choosing a platform, decide for each data type which system is the master and how often it syncs:

Real time or near real time: customer pricing and stock availability, because that is where visible errors start.
At intervals: item master, BOMs and technical documents, which change less often.
Back to the ERP and CRM: orders, quote requests and service tickets, tagged with the serial number.

Larger customers often buy through e-procurement platforms and ask suppliers for a "punchout" catalog: the buyer opens the supplier's catalog from their own purchasing system, builds the cart and brings it back into their approval workflow. One protocol used for this exchange is cXML, designed for business documents exchanged between procurement applications, e-commerce hubs and suppliers: it covers catalogs, punchout, purchase orders, order confirmations, ship notices and invoices. If your key accounts require it, plan for it from day one.

Then there is the sales network: if some customers are served by agents or distributors, the portal must know who owns each account, credit online orders to the right person and give distributors their own access with their own prices. These are commercial rules before they are technology; for the broader principles of B2B channels, see how B2B and B2C strategies differ.

How does a configurator connected to quoting work?

A product configurator guides the salesperson, or the customer, through options that are compatible with each other and calculates price and lead time. When it also generates the quote document, it is called CPQ (configure, price, quote). The main benefit is technical consistency: engineering writes the compatibility rules once (this motor does not fit that frame, this option requires that control panel) and every quote follows them.

You do not need an enterprise CPQ on day one: for catalog machines with a few options, a web configurator connected to the price list is enough. For complex deals the negotiation can stay human, as long as it is tracked. Adobe Commerce, for example, supports negotiable quotes: the buyer submits a request to negotiate the price from the cart, and the seller responds with discounts and adjusted quantities and items.

How do you automate quote requests?

Structure the request at the point of entry, move the data into the CRM automatically and leave pricing and specification decisions to a person. A typical flow, built with n8n or Make, has five steps:

A form with the fields engineering actually needs: model or serial number, application, material processed, required throughput, country of installation.
For email or PDF requests, an AI model extracts the same fields and flags missing ones without inventing them.
The request lands in the CRM with company and history, so the quote reflects machines already installed.
The system drafts a quote from the price list or configurator and assigns it to the right rep by territory or product line.
The rep reviews, approves and sends it; the system schedules the follow-up and records the outcome.

Automation prepares, people decide. To choose where to start, apply the criterion from the guide to business process automation: high volume, clear rules, costly errors. For exporters, AI translation speeds up multilingual data sheets and portals, but technical terms need human review.

What role does machine data play in after-sales service?

It turns service from reactive to planned. If the machine reports hours, cycles and alarms, the manufacturer can offer the maintenance kit before a breakdown.

Predictive maintenance means using operating data (vibration, temperature, power draw, cycles) to estimate when a component will fail and act before it does. It needs a history of data and failures, so for many manufacturers the realistic first step is threshold-based maintenance: an alert when a counter exceeds a value set by engineering.

Maintenance type	When you act	Data needed	What the manufacturer offers
Corrective	After the failure	None, just the report	Urgent part and field visit
Time-based preventive	At fixed time intervals	Installation date	Maintenance plan and scheduled kits
Threshold-based	When a counter exceeds a limit	Machine hours, cycles, alarms	Automatic alert with the linked kit
Predictive	When data shows degradation	Operating and failure history	Intervention before downtime, service contracts

The same data is valuable upstream. Digital orders show which parts are requested, where failures cluster and how demand shifts over time, which supports better inventory and production planning, as commercetools also notes: combined with the installed base (how many machines of which version, with how many hours), they become a forecast of the parts to keep in stock. And recurring failures on a component are valuable input for engineering, which can fix the design in later series.

Machines already installed without connectivity are not left out: hours and alarms can be collected through a retrofit (a gateway that reads the existing PLC) or, more simply, by asking customers to log counter readings in the portal during service visits. To decide where to invest, start with the machines that sell the most parts.

The Data Act (Regulation (EU) 2023/2854) has applied since 12 September 2025 and covers connected products such as robots and industrial machines. According to the European Commission, the user can access the data they generate with the machine free of charge and have it shared with third parties, such as repair and maintenance providers; the manufacturer, usually the data holder, cannot use non-personal data without the user's agreement. After-sales is no longer a de facto exclusive: customers stay when the manufacturer's service is the most convenient.

The new Machinery Regulation (EU) 2023/1230, mandatory from 20 January 2027, also clarifies the conditions under which instructions for use and the declaration of conformity can be provided in digital format, and adds cyber-safety provisions for compliance-relevant software, data and safety control systems, as well as rules for AI-powered safety functions: a good reason to link manuals, exploded views and parts to the serial number in the same portal. To measure the return, set a few indicators up front (see KPIs for measuring the value of AI).

What cybersecurity obligations apply to connected machines?

A machine that connects to a network or another device can qualify as a product with digital elements under the Cyber Resilience Act (Regulation (EU) 2024/2847), unless it falls under the exclusions for products already covered by other EU legislation: check case by case. According to the European Commission's summary:

from 11 September 2026, manufacturers must report actively exploited vulnerabilities and severe incidents, including for products already on the market;
from 11 December 2027, the main obligations apply, including vulnerability handling throughout the support period;
the end date of the support period must be stated clearly at the time of purchase.

For after-sales, this means software updates, security notices and instructions must reach the right customer for the right machine. A portal keyed to serial numbers is the natural channel, alongside parts and manuals. For the general approach, see a proactive view of cybersecurity; for the obligations that apply to your machine, get technical and legal advice.

Who should lead the project internally?

This is an organizational change more than a technology one: machine data is only useful if someone interprets it and turns it into decisions. You need at least four roles, even if a few people cover them:

After-sales or parts manager: owns the portal and its metrics.
Engineering and technical documentation: maintain BOMs, compatibility rules and manuals.
IT or the ERP provider: ensure integration with ERP and CRM.
Sales and the sales network: set prices, terms and rules for accounts served by agents and distributors.

Start with a pilot on a few customers and one machine family, prove its value with your own numbers, then extend it: that is the logic of the steps below.

How to apply it in your company: first steps

Take a hypothetical packaging machinery manufacturer with a few hundred machines installed across Europe, receiving spare parts requests by email every day:

Measure the baseline for one month: parts and quote requests received, average response time, orders corrected because of wrong part numbers.
Link serial numbers, bills of materials and price lists in a single master record: without this step the portal shows the wrong parts.
Open the spare parts portal to a small group of regular customers and measure the share of self-service orders after 90 days.
Connect the quote form to the CRM, with automatic drafts and sales rep approval; the metric is time to first response.
Pick one family of connected machines and set up threshold-based maintenance alerts, each tied to the matching parts kit.

Training sales and service staff is part of the project, not an afterthought: hands-on sessions on the portal and the data, repeated as features change. If the investment includes digital or AI technologies, also look at funding options for AI in SMEs (an Italian perspective). This is the kind of system I design in the websites, e-commerce and automation area, with ERP and CRM integrations.

Mistakes to avoid

Publishing the catalog without linking it to serial numbers: the customer picks the wrong part and calls anyway.
Showing prices and stock that are not synced with the ERP.
Automating quotes without human approval of prices and specifications.
Collecting machine data without deciding its purpose and who may use it.
Leaving agents and the sales network out of the project, so they see the portal as internal competition.
Not assigning anyone to update BOMs and part supersessions: within months the catalog is no longer reliable.
Treating embedded software security as a development-only issue, with no channel to get updates and notices to customers.

Frequently asked questions (FAQ)

Is a B2B spare parts portal worth it for a small or mid-sized machine builder?

Yes, if it regularly receives parts and consumables orders from customers who reorder. The return depends on how many repetitive requests currently take up the sales office's time, not on company size.

Do you need CPQ software to sell machines online?

Not at the start. For catalog machines with a few options, a web configurator connected to the price list is enough. Dedicated CPQ pays off when compatibility rules are many and change often.

Can AI write quotes on its own?

It can prepare the draft: extract data from emails and PDFs, suggest the right model, fill in the quote from the price list. Final price, technical exceptions and terms remain the sales rep's call, because a wrong quote on a machine costs more than the time saved.

What does the EU Data Act change for makers of connected machines?

Since 12 September 2025, the user of a connected product, industrial machines included, can access the data it generates free of charge and ask for it to be shared with third parties, such as an independent repairer. The user must be told before signing what data the machine produces, and the manufacturer cannot use non-personal data without their agreement. Micro and small manufacturers face lighter obligations; get legal advice on the contracts.

Are Shopify or WordPress suitable for an industrial B2B portal?

In many cases, yes. Shopify has native B2B features, such as company accounts with dedicated pricing and catalogs, available from the Basic plan up, with features varying by plan; with WordPress and WooCommerce, B2B features are added through extensions. The choice depends mostly on how deep the ERP and warehouse integration needs to be.

How much does a B2B spare parts portal cost?

It depends less on the platform and more on three factors: the state of your technical data (BOMs and part numbers already linked to serial numbers or still to be rebuilt), how deep the ERP and CRM integration goes, and the features customers require, such as punchout or distributor pricing. For a reliable estimate, start with a data review and a pilot on one machine family.

Can older, non-connected machines be included in the portal?

Yes. A spare parts portal works by serial number even without operating data. For threshold-based maintenance, hours and alarms can be collected through a retrofit, a gateway that reads the existing controller, or from readings logged by the customer or technician during service visits.

Sources

To find out which process to tackle first among spare parts, quotes and service, book a free assessment.