Chapter 8: Anyone Can Sell a Crane
It's ten o'clock at night and a homeowner is configuring an elevator.
Not browsing. Configuring: choosing a cabin style, comparing finishes, trying the lighting options, watching the price update. Somewhere beneath that pleasant surface, a thousand rules about shaft clearances, load classes, drive types, and safety codes are silently doing their work. The homeowner never sees a single one of them. One question ("Where will it live?") has already set the size envelope, the compliance regime, the power requirements, and the ventilation.
They feel possibility. The system handles reality.
That division of labor is the whole secret of selling complex products through channels you don't control. Manufacturers rarely fail at self-service because they lack rules. They fail because they surface them: sixteen-step wizards, part-number menus, red error walls. The buyer wanted to make eight or ten meaningful choices. They were handed an engineering exam instead, and there's a number for what happens next: 68 percent of B2B buyers abandon or delay online purchases out of fear of ordering the wrong thing. Self-service doesn't fail on convenience. It fails on trust.
One brain, many faces
The structural principle that fixes it: configure once, deploy everywhere. One product brain (the constraint model, the pricing waterfall, the bottled expertise of the last three chapters), with every channel as a different face on the same brain. The rep's screen shows engineering depth. The dealer portal shows commercial guardrails. The website shows the homeowner cabins and lighting. Your channels should disagree on style, never on truth.
The moment the logic forks, the tax begins. I've watched the scene play out at a manufacturer: the rep demos an option in the CRM, the partner quotes the same machine at a different price with a missing compliance package, and the website is selling last quarter's variant. Nobody lied. The logic lived in three places.
For a lift manufacturer I work with, the sequencing was the strategy: first quarter, internal CPQ on one product line, salespeople quoting correctly. Then the same rule backbone was pushed outward, market by market, into a build-your-lift web experience and distributor tooling. No double maintenance, no drift. Their smaller distributors used to have exactly one person who could produce a correct quote. Now a distributor rep can quote correctly, with drawings, on their first day. That's not a UX achievement. It's a structural one: the correctness is guaranteed under the surface, so the surface can afford to be simple.
And it changes what selling looks like physically. Configuration with live CAD: the customer picks options and watches the actual machine update, dimensions and all. Two different visual jobs, worth keeping apart: the plant engineer needs parametric drawings that answer "will it fit, will it connect, will it pass inspection?", while the homeowner needs a rendering that answers "will I love it?". Same brain, different faces again.
Partners without chaos
Dealers and distributors are where trust in structure gets stress-tested, because you're handing pricing power to people with their own incentives. The governance pattern that works is bounded autonomy. Partners get real freedom inside hard fences: their own catalogs, their own discount bands with hard stops, their own service bundles behind a review flag. Prices come as quarterly released price books, frozen at quote time, so nothing changes overnight underneath a customer conversation.
One capital-equipment maker runs 120 dealers on this model, and their most valuable governance rule costs nothing: dealers never see each other. No shared views, no leaked price lists, no cross-region shopping. That single choice, they estimate, prevents 80 percent of channel friction before it starts. Culture changes with quotas. Guardrails do not.
Anti-pattern: The Channel Fork. A copy of the product logic, made "just for the webshop" or "just for the dealer portal," maintained by someone else. It looks harmless on day one. It becomes a permanent tax: every product change now ships twice, drifts silently, and eventually contradicts your own reps in front of a customer.
Here is what I find executives underestimate most: the revenue ceiling this removes. If your complex catalog is only sellable through experts, you will only grow at expert speed. Structure is how you break that law.
One channel remains, the newest and the strangest: the customer who wants to just describe their problem in a sentence and have the machine configured from it. That works now. It also goes wrong in a very specific way.
When the system guarantees the machine can be built, you can let anyone sell it: dealers, distributors, even the customer.