Most B2B ecommerce platform projects start the same way: a team puts together a requirements list, three vendors present their platforms, and in the end the most convincing demo or the lowest quoted price wins. The outcome is familiar. Projects drag on, interfaces do not work as expected, and the system that impressed in the showroom struggles with the company's real-world processes.
The 2026 B2B Commerce Trends & Opportunities Report (July 2026) confirms what experienced consultants have observed for years: companies that select their ecommerce platform primarily through feature comparisons fail more often than those that first clarify their own processes, data and system landscape. Modernising only the storefront without touching the systems underneath does not create a solution, it creates a new layer of problems.
This article sets out five questions a managing director or decision-maker in a mid-sized company should answer before committing to a B2B ecommerce platform. No technical configuration details, no code listings. Only the questions that decide whether a project succeeds or stalls.
Question 1: How good is our product data, really?
In most mid-sized companies the honest answer is: not good enough. Product data often sits in spreadsheets, in the ERP system, in old catalogue PDFs, or in the heads of long-serving employees. The moment that data has to be transferred into an online shop, the gaps become obvious.
A shop platform can only be as good as the data it displays. Missing dimensions, incomplete descriptions or contradictory prices lead customers to pick up the phone instead of placing an order. That is the opposite of digitalisation.
What a good answer looks like: the company knows where its product data lives, what condition it is in, and who is responsible for maintaining it. Ideally a PIM system (product information management, a central system for managing product information) is already in place, or there is a clear plan for closing data gaps before go-live.
A company that cannot answer this question should pause the platform selection and start with an inventory of its own data. No shop platform in the world compensates for poor product data.
Question 2: How deeply does the shop platform need to reach into our ERP?
Connecting an online shop to the ERP system is, in practice, the most common cause of delays and cost overruns in B2B ecommerce. Not because the technology is missing, but because the real requirements for the interface are often only understood during implementation.
In a B2B context those requirements are considerably more complex than in classic consumer business. Customer-specific prices, individual payment terms, credit limits, real-time stock levels, order confirmations from the ERP: all of this has to be synchronised. If the shop platform and the ERP do not communicate cleanly, manual rework and errors follow.
What a good answer looks like: the company has a clear picture of which data must flow between shop and ERP in real time and which can be synchronised daily. There is a contact person on the ERP side who is involved in the selection. And the prospective vendor can demonstrate working reference projects with the same or a comparable ERP system.
An important insight here is that replacing the ERP is rarely a prerequisite for a good online shop. The gap can usually be closed without a system change.
A B2B platform comparison that does not treat ERP integration as a central criterion is not a serious comparison.
Question 3: Which processes are still manual today, and who will run them in future?
Many mid-sized manufacturers and wholesalers have built processes over years that work but depend heavily on individual people. Quotes are written manually, orders arrive by fax, delivery addresses are agreed by phone. These processes are not documented; they exist as implicit knowledge.
A shop platform cannot simply take these processes over. It can represent them, provided they have been clearly defined first. The question is therefore not what the platform can do, but what the company is willing to change.
What a good answer looks like: the company has described its relevant ordering processes in writing, knows which exceptions exist (special terms, custom orders, returns) and has decided which of these should be handled digitally and which will stay manual. It is also clear who will look after the shop after go-live: who maintains content, who handles complaints, who releases new products.
Without that clarity, the launch produces a system that works technically but that nobody inside the company really operates.
Question 4: How flexible does the platform need to be in three years?
Shop platforms are rarely chosen for a company's current state, but for what the company wants to be in two to four years. That is an important difference. A platform that is sufficient today can become a brake in three years if the company wants to enter international markets, test new business models or connect to marketplaces.
Scalability is not a technical question. It is a strategic one. Choosing a system today that can only handle the current catalogue means starting over at the next growth step. That costs time, money and patience.
What a good answer looks like: the company has a rough idea of how its digital sales should develop, whether that means more countries, more channels, more product groups, connections to B2B marketplaces or its own customer portal. The chosen platform demonstrably supports that direction without requiring a new project at every step.
Platform architecture (the underlying technical structure of a system) and data sovereignty (the question of who controls your own data) play a central role here. A company that cannot export or migrate its data becomes dependent on a single vendor in the long run.
Question 5: Who inside the company owns this project?
This is the most uncomfortable question, and the one most often skipped. Platform projects rarely fail because of technology. They fail because of unclear responsibilities, missing internal capacity, and because nobody truly owns the outcome.
In mid-sized companies the starting point is often the same: management wants ecommerce, the sales director is sceptical, IT is at capacity, and marketing has no experience with digital channels. In that vacuum the project ends up with an agency that delivers what was commissioned, but not what the company actually needs.
What a good answer looks like: there is a named person or a small team responsible for the project's success, with the capacity to actually carry that responsibility. That person has access to management, can make decisions, and is authorised to change internal processes. If no such person exists, that is not an argument against ecommerce, but a clear argument for planning external support from the outset.
Recommended sequence
Answering these five questions consistently already puts a company ahead of most organisations that jump straight into a platform comparison. The recommended sequence:
First: take stock of product data and the ERP landscape. What exists, what is missing, what needs to be built.
Then: document the key ordering processes and decide which of them should be handled digitally.
In parallel: appoint internal project ownership with realistic capacity.
Only then: select the platform based on your own requirements, not on vendors' feature lists.
Finally: include scalability and platform architecture as a decision criterion, not as an afterthought.
This route is slower than a direct comparison of quotes. It is considerably faster than a failed project that has to be restarted after 18 months.
Companies looking for support with this analysis will find that experienced B2B ecommerce consultants start with exactly these five questions, long before any platform is even mentioned.
