The dealer is not usually angry about the existence of a program. They are angry about a specific order.
A customer researched on the brand site, configured a product, maybe even started a cart. They finished at the dealer — or the dealer delivered, installed, or closed it after a lead the brand thought it "owned." At month-end the incentive file does not show the deal. Or it shows it as a house account. Or it shows it twice. Or the dealer in the next territory claims it.
The program was supposed to protect the network while the brand went digital. Instead, omnichannel produced a fight about whose sale it was.
B2B commerce is not a consumer storefront with a login. Manufacturer–dealer commerce adds a third party to that problem: the partner who still owns the relationship, the truck, or the service bay. Incentives are how the brand says that partner still matters. They only work if the company can tell a single sales truth across channels.
What Has to Be True Before Anyone Gets Paid
An incentive is a commercial rule applied to a transaction. If the transaction cannot answer a few questions, the rule is theater.
Who influenced the sale — a dealer salesperson, a brand campaign, a marketplace, a walking customer? Who fulfilled it — dealer inventory, brand warehouse, a 3PL, the store down the road? Which dealer owns the customer or the territory, and does "own" mean last touch, first registration, ship-to ZIP, or named account? Where did the demand originate if the click was brand and the close was local? Which channel is allowed to receive credit when two of those answers conflict?
Without those definitions, every new digital path creates channel conflict. Brand ecommerce looks like it is stealing. Dealers stop registering demand. They under-report. They hold inventory. They tell customers not to use the website. The brand reads that as analog resistance. Often it is rational: the partner cannot see how they get paid if the customer uses the channel the brand is advertising.
Assisted sales make this sharper. A dealer quotes from a portal, the customer pays the brand, the dealer delivers. Or the opposite. If "sale" means the payment event in one system and the delivery event in another, commissions and SPIFs will not match anyone's memory of the work.
Customer ownership is the long version of the same fight. The brand site captures an email. The dealer has served that plant for a decade. Who owns the next reorder? Programs that ignore the installed relationship in favor of last-click will look digital and feel extractive. Programs that freeze the website so dealers never lose a lead will stall growth. The usable middle is a recorded influence path, not a slogan about partnership.
Spreadsheets Are Where Trust Goes to Die
Many programs still run on uploads. Dealers submit. The brand matches. Exceptions pile up. A portal shows a balance that is three weeks behind the business. Disputes are a side process with no link to the order.
That is not an incentive-design problem. It is an attribution-model problem sitting on incomplete events. If sales events do not move, no rate card will be believed.
Territory rules, product-line rates, and period bonuses can be as complex as they need to be after the sale is a shared object: identity of buyer, selling location, fulfilling location, SKU, timestamp, and the influence path the company has agreed to recognize. Before that, complexity is how you multiply arguments.
POS, portal, and ecommerce have to contribute to that object or the program will credit only the channel that files the best spreadsheet. A store sale that never carries dealer ID is not a POS problem in isolation. It is an incentive hole.
Transparency is part of the operating model, not a courtesy dashboard. A dealer who can see, on the same day, which orders credited them and which were held — and why — will fight fewer ghosts. A dealer who learns at rebate time that a house website order used their customer will not.
Digital Does Not Have to Mean Direct-Only
Brands are not required to choose between a dealer network and a digital channel. They are required to choose an ownership model and encode it.
Some sales are dealer-fulfilled by design, even when discovery started on the brand site. Some are brand-fulfilled with a dealer spiff for the relationship. Some are protected accounts. Some ZIP codes route to a partner. None of those policies survive a cart that does not know the partner, an order that does not carry the dealer ID, or a finance file that only sees a credit-card sale.
House accounts and "national" customers need the same explicitness. If the brand will own those sales, say so in the transaction, not in a side list dealers discover when they are not paid. Ambiguous exceptions are how programs rot.
Replatforming the storefront without that model just makes the conflict faster. A better catalog will generate more demand the network cannot trust.
The work is unglamorous: define credit at the moment of order, not at the moment of argument; write dealer identity onto the transaction; calculate incentives from that object; give partners a view that matches operations, not marketing. Custom software is sometimes required because generic incentive tools assume one channel and a clean upload. It is not required because dashboards should be prettier. It is required when the commercial rule cannot be expressed in last month's spreadsheet.
If the brand cannot say, for a given order, who caused it, who fulfilled it, and who is owed, it does not have an incentive program. It has a monthly negotiation. Leadership should treat that as a channel-design failure, not as dealer politics.
A working program is boring in the best way: the order is born with credit already assigned, the dealer can see it, finance can accrue it, and disputes are exceptions with evidence rather than the monthly operating system of the channel.
Do not wait for a dealer council to design that object. The next digital order will create the dispute if the fields are not already on it. Assign influence, fulfillment, and payee when the order is created, then argue only the true exceptions. Everything else is delay dressed as governance. Credit assigned at creation is cheaper than credit argued at close.