The complete pre-construction launch checklist (free download)

Most pre-construction launches fail in the same three places: the marketing pages aren't ready when the paid ads start spending, the APS template isn't approved by counsel until the day of broker preview, and the sales team has never seen the AI chatbot until launch morning.

T-8 weeks - Marketplace listing draft and inventory data structure. The sales platform listing is built in draft mode this week, even though pricing and availability will not be public for another two months. The developer's sales operations lead and the platform's implementation team agree on the inventory data model: how floors, stacks, exposures, suite types, square footage ranges, parking and locker availability, and price tiers will be represented - and crucially, how they will be updated during the launch when half of inventory moves in 48 hours. The brokerage's listing manager validates that the structure matches how their agents actually sell (by exposure and view, not just by suite type). Get this wrong and your platform spends launch day showing 'available' on suites that sold an hour ago, which is the fastest way to lose buyer trust.

T-6 weeks - APS template and disclosure package legal review. The Agreement of Purchase and Sale, the disclosure statement, and the schedules (features and finishes, deposit structure, assignment provisions, occupancy and interim occupancy terms, development charges cap if offered) go to the developer's Ontario real estate counsel for review. Counsel confirms the package aligns with the Condominium Act's 10-day rescission requirement and the developer's obligations under the warranty and licensing regime that governs new-home vendors in Ontario; the developer's project counsel and the brokerage's deal desk both need to sign off before the document is loaded into the sales platform's e-signature flow. The failure this prevents is the one that quietly kills deals: agents writing offers on the broker preview day against a template that legal hasn't blessed, then having to re-paper every signed APS at T-0.

T-4 weeks - AI chatbot knowledge base and handoff rules. The conversational layer on the listing - whether it answers buyer questions on the microsite, qualifies leads overnight, or handles the broker portal - gets trained this week, not launch week. The developer's marketing lead and the brokerage's sales manager jointly own the knowledge base: approved answers on price ranges, deposit structure, estimated occupancy, parking and locker availability, assignment policy, and what the bot must never say (anything that resembles legal, tax, or investment advice). The platform's AI lead owns the handoff rules: which questions route to a live agent, which questions trigger a CRM lead, and which questions get a 'we'll follow up' with a logged ticket. Without this week, the chatbot launches as a liability - confidently quoting the wrong maintenance fee estimate to a prospect at 11pm on launch night.

T-3 weeks - Sales platform UAT and CRM integration test. A dry-run week that does not appear on most checklists but should. The platform's implementation team, the developer's sales ops lead, and the brokerage's tech contact run end-to-end tests: a fake buyer registers interest, gets the welcome email, books a preview, receives an APS, signs it, deposits route to the trust account, and the inventory ticks down by one. Every integration in that chain (CRM, e-sign, payments, MLS feed if applicable, broker dashboard) gets exercised against a non-production environment. The failure mode this prevents is the one nobody admits to: a signed APS on launch day that never makes it into the CRM because a webhook was misconfigured, and the buyer's deposit sits in limbo for a week.

T-2 weeks - Broker preview and broker incentive structure. The brokerage runs its preview event for cooperating agents - typically in-person at the presentation centre or via a structured virtual walkthrough - and publishes the broker incentive package: commission structure, tiered bonuses by suite type or velocity, deposit schedule for buyers, and the assignment policy agents need to be able to explain. The developer's VP of sales and the brokerage's principal own the incentive math; the marketing agency owns the broker-facing collateral; the platform makes sure every previewing agent has a working login to the broker portal before they leave the room. The failure this prevents is brokers showing up to public launch without ever having seen the product, which is what produces the 'sales team has never seen the chatbot until launch morning' problem in the first place.

T-1 week - Final pricing, paid media in flight check, and sales floor staffing. Pricing goes from draft to locked this week with sign-off from the developer's CFO, VP of sales, and the brokerage principal. The marketing agency does a flight check on the paid program: Meta and Google campaigns are built and paused, audiences are loaded, UTM conventions match what the CRM expects, and creative is approved against the locked features-and-finishes sheet. The brokerage finalizes the launch-week staffing rota: who is on the sales floor at the presentation centre, who is on the broker hotline, and who is on overflow for the platform's live chat queue. The failure this prevents is the avoidable one - ads spending against a campaign whose UTM structure doesn't tag leads correctly, so the agency cannot tell the developer which channel actually produced the first 50 deals.

T-0 - Soft launch morning, paid media activation, sales floor live. Public launch runs in a deliberate sequence, not a single switch. First: the brokerage's database goes live a few hours before paid media so warm leads convert before cold traffic arrives. Second: the marketing agency activates paid ads against the locked landing page, with the sales platform's analytics team watching conversion in real time. Third: the sales floor (developer reps, brokerage agents, platform support) is staffed for a 12-hour day with a documented escalation path - pricing questions to the VP of sales, legal questions to counsel on standby, platform issues to the implementation lead's direct line. The chatbot handles overnight inquiries against the knowledge base finalized at T-4 and rehearsed at T-3. The failure this prevents is the one the intro of this post named: every preventable scramble on launch morning is the result of a milestone that quietly slipped weeks earlier.