Closing the deal in the driveway

The problem was never the signature. It’s where the decision happens — and the fact that every existing tool assumes the one place the agent isn’t: a desk, on Wi-Fi, after the buyers have gone home and cooled off.

A competitive market is won in the ninety seconds after “we love it.” KeySign sits in that moment and runs one loop on the whole transaction: carry forward → draft & disclose → verify & sign → submit → open escrow. It reuses the CRM data already on the phone, drafts a compliant offer with one competitive decision left to make, and works entirely offline — so the agent acts while the buyers are still standing there, not from the office two hours later when the competing offer has already been accepted.

Translating the field insight into a single job-to-be-done: “When a buyer commits at a showing, help me lock in a competitive, compliant offer before the moment — or a competing offer — passes.”

The root causes weren’t laziness or bad tooling in isolation — they were structural: the work was split across disconnected products, all of which assumed connectivity and a desk, and none of which treated compliance as part of speed. KeySign’s definition collapses them into one offline-first surface where doing it fast and doing it right are the same motion.

The builder reduces a competitive offer to the single call that actually requires judgment — the escalation — and drafts the rest, scoring strength live as the agent moves.

A completeness ring anchors the running state; the price field formats money as you type; the C.A.R.-style escalation clause sets a dynamic cap; and a 0–100 strength meter reacts to the escalation, the gap to asking, and contingencies. Buyer-broker compensation is handled inline (RPA ¶3G + SPBB), reflecting the post-Sitzer/Burnett rules. The benefit: a genuinely competitive, compliant offer with one real decision left to make.

A guard runs before signatures and states the one blocker plainly — because the most expensive error is the one discovered after four people have signed.

The missing page-6 lead-paint initials would force every buyer to re-sign if they slipped through. Pre-flight names it in plain language and offers two stacked, full-width choices: “Queue it & fix now,” which resolves the flag and flips the package to 8/8, or a manual review. The benefit: the system spends a moment of friction now to save a full re-signing later.

Signing is the screen, not a step crammed under other controls. A wide drawable canvas, one mark at a time, with progress made explicit across all four required marks.

The real prototype uses a DPR-scaled canvas that takes pointer and touch input, so the signature is crisp whether it’s drawn on a phone balanced on a car hood or a tablet on the hood of a truck. Four marks — each buyer’s signature plus the page-6 initials — advance one at a time with a clear active state. The benefit: nothing competes with the one thing the person is doing.

Most tools stop at “submitted.” KeySign carries the momentum forward into an open escrow file — the single thing that turns a signed offer into a closing deal.

Once accepted, the screen shows the accepted price and escrow number, fraud-safe earnest-money wiring, the contingency deadlines on a visible clock, and a parties hub for the buyers, listing agent, escrow officer, and lender. Why this matters: the driveway moment only pays off if the deal actually closes, and putting the clock and the parties on screen keeps the whole transaction moving. The benefit: driveway-to-close, in one place.

Offline isn’t a degraded mode bolted on at the end — it’s the assumption the whole app is built on. Work continues at zero bars and reconciles itself the moment signal returns.

Toggle the signal and the prototype keeps working: drafting, disclosing, identity, signing, and submit all queue locally, then auto-flush on reconnect with a clear running count. Why this is the best choice: the driveway is, by definition, where signal is worst — designing offline-first instead of online-with-a-fallback is what makes the core promise believable. The benefit: the agent never has to think about the network.