AGENTS
Agents and dashboards inside a property manager's operations
AGENTS & DASHBOARDS
Yardi, EliseAI and the pricing sheet feed a pricing report. The leasing lead signs it, and it is sent to the team.
8+ hours by hand → 10 minutes
per report; running about a month, against building it by hand in Yardi
What was built
Where it runs
Tools
By hand, the daily pricing report takes 8+ hours of copying from Yardi, EliseAI and the pricing sheet. With the agent, it's compiled and checked, the leasing lead signs, and it goes to the team in 10 minutes.
Towne Properties manages communities from Cincinnati, and every month someone senior pulled the same reports from two systems by hand, lined them up, and read the story off the numbers — for a pilot group of properties whose results leadership was watching closely. The numbers mattered and the read was real. The ritual was the problem: a morning of pulling before a minute of thinking.
A dashboard that mirrors that monthly workflow exactly — the two systems' reports side by side, on one screen, refreshed on the schedule the team already keeps — so the read happens in minutes instead of a morning, and the definitions the team used to argue about are written down once.
Around it, an AI program backlog: the workflows worth automating next, ranked with leadership — management dashboards, ticketing and triage, scheduling — tracked in the open and built one at a time. And before any vendor's AI was bought, a roadmap review that keeps the vendor's claims, the publicly confirmed facts, and the open questions visibly separate.
The team also needed daily and weekly pricing reports, and building each one by hand in Yardi took 8+ hours. Now an agent pulls rents and occupancy from Yardi, leasing conversations from EliseAI and the team's pricing sheet into one report every morning, checks it against yesterday, and queues it for the leasing lead to sign before it goes to the team.
It takes 10 minutes per report. It went live in 1 week and has been running for about a month.
The red box is the only step with a person in it; every other step is an agent.
Case 01 · Towne Properties · property management
Rents and occupancy from Yardi, leasing conversations from EliseAI and the team's pricing sheet come together in one report every morning, checked against yesterday, signed by the leasing lead.
Format of the run log · sample, no client data
Nothing here replaces a decision. The dashboard shows the same numbers the same person used to pull, with the definitions written down once — the agent never redefines a number. The backlog is prioritized by leadership, not by the agent. The roadmap review ends in decision controls — approve, hold, needs confirmation — so a person, not a model, decides what goes ahead. The pricing report goes out only after the leasing lead signs it.
The round-one priorities set with leadership: management dashboards first, then contract management, ticketing and triage, and a scheduling app — each one a pilot with its own approver, on the systems the team already runs.
Credits
