B2B webshop business case for manufacturers

A manufacturer calculates a B2B shop on Salesforce over three years: GMV fee, buyer licences, implementation, ERP integration and run costs on one side, saved inside sales time, fewer order errors and additional revenue on the other. The important part is that the GMV fee grows with the store’s revenue, so the costs rise exactly when the store succeeds. In the worked example below, time and error savings alone cover only about a third of the costs. A store usually pays off through additional revenue and through inside sales capacity you actually redeploy. Both need honest numbers, not vendor averages.

Five cost items, and most estimates miss two

Five items, and most estimates miss at least two:

1. GMV fee. Salesforce charges a percentage of gross merchandise value, the store revenue minus tax and shipping. The pricing page1 shows “Contact for Pricing” for Growth and Advanced and publishes no percentage. Details are in B2B Commerce pricing: what the GMV model costs.
2. Buyer licences. Buyers need Customer Community Plus. On the Experience Cloud for Manufacturing price list2 it costs 35 euros per member or 15 euros per login per month, billed annually.
3. Implementation. Store setup, catalogue, buyer groups, entitlements, price books, checkout, tests.
4. ERP integration. Price and stock calls, order transfer, status and invoices back. In manufacturing this is often larger than the store itself.
5. Run costs. Catalogue maintenance, support for dealers, three Salesforce releases a year with regression tests.

What belongs on the benefit side?

Four items, in the order in which I trust them:

1. Inside sales time per order line. Measure how long typing in one line takes today, including the query when a part number is wrong. This is the most reliable number, because you can stop the clock.
2. Order errors. Wrong part, wrong quantity, wrong address. Each one costs a return, a credit note and a second shipment.
3. Fewer status calls. When dealers see order status and history in the store, some calls disappear.
4. Additional revenue. Orders placed at night or at weekends, dealers who used to buy from a third party, orders that were never placed because the email got lost. This is the largest and least certain item.

One rule for item 1: saved time is only money when you use it. If inside sales keeps the same headcount and does the same work, the saving stays on paper. Decide in advance what happens with the freed capacity, for example active calls to dealers who order less than last year.

Worked example: spare parts store for 250 dealers

Assume a manufacturer of conveyor systems sells spare parts through 250 dealers. The numbers:

• 40,000 spare parts order lines a year, average value 150 euros per line, so 6 million euros net.
• Inside sales needs 4 minutes per line at a full cost of 55 euros per hour: about 3.67 euros per line.
• 2 percent of manually entered lines contain an error that costs 60 euros to fix: 1.20 euros per line.
• The GMV fee is assumed at 2 percent.
• One member licence per active dealer at 35 euros a month.
• Implementation including ERP integration 250,000 euros, run costs 40,000 euros a year.

With a digital share rising from 25 to 70 percent over three years, the result looks like this:

Year	Store share	Lines in store	Time and error savings	GMV fee	Buyer licences	Run costs	Costs
1	25 %	10,000	48,700 €	30,000 €	42,000 € (100 dealers)	40,000 €	112,000 €
2	50 %	20,000	97,300 €	60,000 €	73,500 € (175 dealers)	40,000 €	173,500 €
3	70 %	28,000	136,300 €	84,000 €	105,000 € (250 dealers)	40,000 €	229,000 €

Over three years, savings are about 282,000 euros. Costs are 514,500 euros plus 250,000 euros implementation, so 764,500 euros. The savings cover about 37 percent. To close the gap of roughly 482,000 euros, the store has to bring about 160,000 euros of additional contribution margin a year. At a contribution margin of 30 percent after the GMV fee, that is about 536,000 euros more parts revenue a year, almost 9 percent on top of 6 million.

That may well be realistic. But it is the number the decision rests on, and it should come from sales, with a reason per dealer group, not from a slide.

Why does the GMV fee grow with success?

Look at the scenario per order line. The store saves 4.87 euros per line in time and errors. The GMV fee at 2 percent of 150 euros is 3 euros per line. In year three, buyer licences add another 3.75 euros per line. Per line, the store costs more than it saves in time.

Two effects are behind this. First, the fee scales with the value of a line, the saving with the number of lines. A line of 1,500 euros costs ten times as much fee as a line of 150 euros but takes the same 4 minutes to type in. Stores that sell many cheap consumables do better on this side than stores selling expensive assemblies. Second, the licence per dealer is fixed while small dealers order few lines. A dealer who logs in on two days a month is cheaper on a login licence at 15 euros, because according to Salesforce Help3 several logins on one day count as one. Licence type per dealer group is a real lever in the model.

That is why I always model three years with a rising share. A one-year view hides that the fee rises while implementation is already paid.

The Forrester study shows levers, not results

Salesforce promotes a Total Economic Impact study by Forrester Consulting4, commissioned by Salesforce and published in April 2024. Its headline figures:

289 %

three-year ROI reported by the study

under 6 months

payback reported by the study

4

customer representatives interviewed as its basis

The four interviews are combined into one composite manufacturer with 1 billion dollars in revenue. Forrester itself advises readers to use their own estimates.

I read such studies as a list of levers: which benefits did those customers see, which costs did they name. The numbers do not transfer to a manufacturer with 250 dealers and 6 million euros in spare parts revenue. Your own stopwatch in inside sales tells you more than a composite company does.

First step: four numbers from the last twelve months

Collect four numbers from the last twelve months: order lines by channel, average value per line, minutes per manually entered line and the cost of order errors. Then build the three-year model with the GMV rate as a variable at 1, 2 and 3 percent. Whether a store or a portal fits at all is covered in B2B Commerce or dealer portal: when a shop pays off, and the numbers to measure after launch are in B2B shop KPIs: what to measure after go-live. The overall decision is in the masterclass chapter on whether B2B Commerce is worth it, and an independent check of the numbers is part of my work on make-or-buy and licence forecasts.

Sources4

1. 1pricing page (opens in a new tab)salesforce.com
2. 2Experience Cloud for Manufacturing price list (opens in a new tab)salesforce.com
3. 3Salesforce Help (opens in a new tab)help.salesforce.com
4. 4Total Economic Impact study by Forrester Consulting (opens in a new tab)tei.forrester.com