Three threads of design work exist for Smart City. All of it is paid for and none of it is currently moving. Every one is blocked on a decision rather than on design or engineering time. This page lists what exists, what is already settled, and the specific questions that would unblock each thread.
Three threads, stopped at three different points for three unrelated reasons. Only one of them is a design question, and it is already answered.
20 screens across three walkthroughs, reviewed by the team on 28 May 2026. A costed build menu with three tiers was delivered on the same day. No tier was ever selected.
Four of eight screens delivered 8 April 2026, all approved through Checkpoint 3. Held for a post-launch review that has not taken place in four months.
Search and responsive CRM patterns designed across 2025, including five user-testing sessions with Smart City agents. Confirmed on 17 August: most features are built and in staging, with the full prop build still in progress. No major open UX questions.
This is not a reporting problem or a dashboard problem. Cost per acquisition is spend divided by acquisitions, and the figure is only meaningful next to the attempts that did not convert. In the current model a request record is created when a placement succeeds. The attempts that went nowhere are therefore not in the data at all.
There is no denominator. No amount of BI tooling or dashboard work produces the number, because the entity that would carry it does not exist yet. This is the reason a competent team can spend a call on it and not converge.
The same constraint was named from the other direction in that meeting: "did the client you could have helped this year go with you or not, and why not?" That question and cost per acquisition are the same question. Neither is answerable until an opportunity can exist in an unconverted state.
The surface for it was scoped and never built. The company-forecasting roll-up frame rolls agent annual goals against the company sales target and marketing's lead target, which is where acquisition cost becomes visible and actionable rather than merely reported.
These decisions are not a checklist. They are a chain, and answering them out of order wastes build time. Anything to the right of an unanswered question gets built on an assumption that may not hold.
Agent Money sits outside this chain. It depends on one unrelated question, whether the comp plan shipped as designed, and it can move in parallel the moment that is answered.
Each of these is a business, data or ownership question. None of them is a design question. Design decisions are already made and documented; these are the ones only Smart City can answer.
Everything below has been delivered and reviewed. Nothing here is a proposal.
The fastest path is not a build. It is a short, defined engagement that settles the opportunity-state question and the CPA definition, confirms what the CRM already stores, and produces a sequenced plan with the first tier scoped for development.
That work turns six open questions into a set of decisions with owners and dates, and it makes the existing 24 screens buildable rather than theoretical. Everything above was designed against the live SC-CRM design system, so nothing needs to be redrawn to start.
Pick this up →