We had filed the document by its output
Agreement Sheets lived under Prints for an understandable reason: they eventually become paper or a PDF. That classification described the last technical step and ignored the job the document performs. A salesperson reaches for the sheet while comparing the deal, confirming the vehicle, discussing trade and products, and deciding what the customer should see. Nobody begins that thought by browsing a library of print formats.
The mismatch created a subtle handoff. Quote owned the changing decision, while Prints owned the artifact that was supposed to explain it. Each side needed enough knowledge of the other to reconstruct the same deal. A route could work and the product story could still feel wrong because the employee had to leave the business moment to find its output.
Ownership follows the decision
On August 6, Agreement Sheet ownership moved into the Quote domain. The route, navigation, agreement controls, and customer-presentation entry point now begin where the payment structure and trade conversation already live. Print generation remains a capability, but it no longer determines where the feature belongs.
That distinction sounds architectural and produces a practical result. The selected vehicle and current quote do not need to be rediscovered by a document browser. The employee can move from comparison to agreement while the assumptions are still visible. Future agreement behavior has one accountable product owner instead of being split between whichever module edits numbers and whichever module happens to render a page.
The move revealed two audiences, not two themes
The staff worksheet needs controls, internal working state, and the ability to change the structure. The customer presentation needs a calm explanation of approved choices. Those are different data boundaries, not light and dark versions of the same screen. Hiding employee controls with CSS would still send the browser a staff record and trust presentation code to keep it private.
LotPro’s pre-alpha design instead derives a narrower, server-written projection for the customer surface. Public identifiers are opaque and contain no customer information. Only approved fields cross the boundary, arbitrary labels are constrained, and temporary access can expire or be revoked. Host controls remain attached to the staff session rather than appearing in the guest application with a hidden flag.
- Staff edits and internal controls remain dealership-side
- Customer presentation receives a derived, approved field set
- Public routes avoid person information in identifiers
- Temporary access has an explicit lifecycle
One agreement, three rendering conditions
Relocation also made layout drift harder to ignore. The staff screen, customer view, and exported PDF should tell the same decision story even though they have different controls and physical constraints. The browser can scroll and respond. A guest view removes editing chrome. A PDF has fixed pages, print margins, and no interactive explanation to rescue a clipped total.
The shared hierarchy therefore needs to be defined in the product model: vehicle and dealership context, the compared structure, selected products, totals, disclaimers, and signatures where appropriate. Each renderer can adapt the presentation without inventing a different agreement. Page-break behavior, annotation placement, and dealership-configured sections are product work, not a final stylesheet pass after the calculation is finished.
What we gained by removing Prints from the center
Prints can now focus on reusable rendering tools and templates. Quote can decide when an agreement exists, which current structure it represents, and which customer-safe projection may be published. The PDF is still important; it simply no longer dictates the information architecture around it.
For the employee, the result is continuity. Shape the deal, review the choices, prepare the agreement, and publish or export from the same business context. For the customer, the result is a document and presentation that explain the approved decision without exposing how the dealership produced it.
Current status and the release gate
Agreement Sheet ownership now sits with Quote, and the isolated Live Quote proof of concept demonstrates the separate customer-surface model. That does not make public collaboration generally available. Each rollout still needs dealership-specific domain setup, tenant and role checks, sanitized-field review, retention and legal-hold behavior, audit coverage, and an end-to-end test with fictitious data.
The build supports a clear claim: the feature now belongs to the decision it explains, and the customer boundary is part of its architecture. It does not support claiming that every agreement format, jurisdiction, dealership configuration, or remote session is production-ready without rollout validation.