Key points
- A B2B platform does not replace your sales reps. It takes repeat orders off their desk so their time goes back to negotiation, range expansion and winning accounts.
- Base the decision on last quarter numbers rather than instinct: share of repeat orders, volume of stock and price enquiries, invoice corrections, the hours orders actually arrive.
- B2B differs from B2C exactly where feasibility is decided: price attached to the customer, case packs and unit conversions, payment terms, approvals on the buyer side.
- A platform will not fix messy pricing or messy data, it will expose them. An inconsistent price list becomes something your customer sees, not something a rep quietly corrects.
Wholesalers rarely decide to build a platform because a competitor launched a good looking store. The decision comes when the current model stops adding up: reps work longer hours and the number of orders they handle stays flat. This guide is about moving wholesale selling online without breaking the relationships that carry it. The premise worth accepting up front: a B2B platform does not replace your reps, it takes the repeat orders off them. We start with diagnosis, because diagnosis decides whether this is the right investment now and, if it is, which part of the process should move first.
Signals that the current model has stopped scaling
None of the signals below is on its own a reason to build a system. Each one can be worked around for a while: add a head, ask the customer to wait, issue a credit note. The trouble is that workarounds get more expensive, and almost nobody prices them, because the cost is spread across several people and several departments. Walk through the table with last quarter data in hand, before anyone in the business asks what the catalog should look like.
| Signal | What sits behind it | How to check it in your business |
|---|---|---|
| Reps retype the same orders | The repeatable part of selling eats the time that was meant for new accounts and better margin | Count how many of last month orders were mostly lines the same customer had ordered before |
| Customers email to ask about stock and prices | Data that already exists in your system is reachable only through a person, and only in office hours | Read one week of a rep inbox and count the "is it in stock" and "what is my price" messages |
| Errors in orders and invoices | An order is retyped several times (call, email, spreadsheet, system) and every retype is a chance to get it wrong | Count credit notes and returns caused by the wrong item or the wrong quantity last quarter |
| Agreements live in inboxes and in people | A discount promised on the phone leaves no trace, and it walks out of the door with the rep who promised it | Pick a customer at random and try to reconstruct their terms without asking their account manager |
| Selling stops outside office hours | The customer builds the order on a Friday evening and gets an answer on Monday morning | Look at the actual distribution of hours when order emails arrive |
| A new market means new headcount | Sales grow in a straight line with people, so entering a market starts with recruitment | Estimate how many people the first hundred orders from a new market would need in the current model |
One signal is usually a matter of how work is organised. Three or more at once mean the constraint sits in the sales model, not in the people.
What a platform actually changes
A platform moves the repeatable part of selling into self service. A customer who already knows what they want places the order themselves: they see their price, their range and stock information, they repeat a previous order instead of rebuilding it, and they do it whenever it suits them, including evenings and weekends. The order arrives in one format, with no retyping from an email, so a whole class of mistakes disappears: wrong quantity, wrong item code, price from the previous list. Agreements stop living in inboxes, because terms belong to the customer account rather than to the memory of one account manager.
What it will not do matters just as much. It will not win accounts in place of a rep, negotiate annual terms, introduce a new brand into a buyer range or replace the visit where the relationship is actually decided. What changes is the proportion, not the headcount: repeat orders run through the system, and rep time returns to work the system cannot do. That is the right success measure for this kind of project, and it is worth agreeing on before anyone asks about catalog design, otherwise in six months the business will judge the platform by its feature count instead of by what changed in daily work. How a project like this unfolds over time we covered separately in the piece on digitising wholesale in 90 days.
Why B2B is not B2C with different prices
The most expensive misunderstanding at the start sounds like this: "we will build a normal store, just with wholesale prices". In B2B ecommerce a price is not a property of the product, it is a property of the relationship: one item code carries as many prices as you have customers, plus volume breaks, seasonal deals and individual arrangements. Products do not move in single units either, they move in case packs, cartons and pallets, with conversions and rounding up to a full unit. Payment is usually not immediate but deferred and capped by a credit limit, and the document at the end is an invoice, often with a pro forma along the way.
The other differences all point the same way. On the buyer side an order is often a process rather than one person deciding: the warehouse collects needs, a manager approves, procurement watches the budget. The cycle is long and repetitive, so the value is not in helping people discover new products, it is in letting them rebuild in seconds what they have ordered every fortnight for three years. The design consequence is simple. In B2C you optimise the path to checkout for someone visiting for the first time. In B2B you optimise it for someone who knows your range better than half your team and wants the whole thing done in three minutes.
A platform exposes the mess, it does not clean it
This is the part worth hearing before you sign anything. If your price lists have drifted apart and the real customer price is assembled in a rep head by correcting whatever the system shows, a platform will not heal that. It will do something else: it will show every inconsistency to the customer, instantly, without the intermediary who used to quietly straighten it out. The same goes for data. If stock levels are sometimes stale and descriptions or item codes differ from file to file, after launch that stops being an internal annoyance and becomes a sales problem.
A platform is an amplifier, not a cleaner. It amplifies order where order exists and amplifies chaos where it does not. That is why the sequence of work runs against intuition: first you settle your pricing rules and your source of truth for stock, then you build the front that displays them. Doing it the other way round usually ends with the system being switched off "for a moment" after the first complaints.
Four questions before you decide
Before you talk to anyone about systems, answer four questions and write the answers down. First: what share of your orders is repeatable, meaning it could exist without a rep if the customer had somewhere to place it. That share is what a platform takes over first, and it is also the measure of whether the project is worth doing. Second: for any given customer, can you state their price and terms today without asking their account manager. If not, you have work to do whether or not a platform ever gets built. Third: which system is the source of truth for stock, prices and documents, and how much do you actually trust it.
The fourth question is about people and is usually the hardest: what exactly should a rep do once repeat orders stop landing in their inbox. If there is no answer, the project will meet quiet resistance, because the sales team will read it as work being taken away rather than retyping being lifted off them. Together, these four answers are the input to the scope decision: they tell you what has to work in the first release and what can comfortably wait. Cutting that scope so it can be launched and maintained is the subject of the next chapter.
Questions
Will a B2B platform take work away from sales reps?
It takes away the retyping of orders the customer was going to place anyway. It does not take away negotiation, range expansion, handling a delivery problem, or the conversation that moves another product group over to you. In practice the shape of the day changes: less processing of known orders, more contact that actually grows revenue. Say this to the team early and plainly, because quiet resistance from reps stops a rollout more reliably than any technical issue.
Does a wholesaler with a few dozen customers need a platform?
Repeatability and frequency decide, not customer count. Thirty accounts ordering weekly from a similar list of item codes generate far more repetitive work than three hundred accounts ordering once a year. If your orders are rare, project based and negotiated from scratch every time, the benefit is smaller and the money is better spent first on getting prices and data in order.
What if customers say they do not want to buy online?
It is usually not about being online. It is two fears: losing contact with their account manager, and seeing terms in the system that differ from what was agreed. Both are defused if the platform is an additional channel rather than the closing of existing ones, and if after logging in the customer sees exactly the prices you agreed. The same customer who says no often ends up placing orders at ten in the evening, simply because it is faster.
