REGISTER REGISTER TO LEARN

Cycle Closes → Pull KPIs → Compare To Prior → Explain Changes → Publish

Executive Summary and What Changed Reporting Agent

Generates a clear weekly or monthly narrative from KPIs and exceptions, what moved, by how much, and why, so leadership reads a summary instead of mining dashboards.

The Problem

Leadership gets dashboards. Nobody has time to read them.

Weekly and monthly reports are manually assembled. Commentary is written from scratch every time.

What Changes

A clear narrative: what changed and why, every cycle.

Generates clear narratives from KPIs and exception data. Your team reviews and adjusts the story - the first draft is already done.

Humans stay in control: Your leadership interprets and decides. The agent writes the what-changed narrative from the data. It removes the dashboard mining, not the judgment.

HOW THE AGENT WORKS

TRIGGERLLMAPIDECISIONACTIONWAIT
TRIGGERCycle ClosesWeekly or monthly reporting point
APIPull KPIsGather the period’s metrics and exceptions
LLMCompare To PriorCompute what changed vs last cycle
LLMExplain ChangesDraft the why behind each move
DECISIONAnything Notable?
ACTIONPublish SummaryDeliver the executive narrative
The agent doesn't set strategy. It turns the numbers into a clear what-changed narrative. Your leadership draws the conclusions.

HOW TO BUILD IT

Three levels of depth. Pick the one that matches where you are.

For the ops person who needs to understand what's involved and brief their team.

Your KPIsMetrics + exceptions
The AgentWorkflow engine + LLM
Your SummaryWhat changed, why

WHAT YOUR TEAM PROVIDES ONCE

  • KPI set - the metrics leadership tracks
  • Data access - where each KPI comes from
  • Comparison basis - prior period, budget, or target
  • Notability rule - what size of change is worth calling out
  • Summary format - length and tone leadership wants
What happens day-to-day: At each reporting point, the agent pulls the KPIs, compares them to last cycle, and writes a short narrative of what moved and why. Leadership reads a paragraph that points them at the two things worth their attention instead of ten dashboards.

For the IT person or technical lead who needs to scope the integration.

INQUIRY SOURCES

PMS / CMMS
Accounting
Other agents’ outputs

WORKFLOW ENGINE (N8N)

Trigger - schedule
API - Pull KPIs
LLM - Compare to prior
LLM - Explain changes
Branch - notable?
Publish summary
Assemble narrative

CONNECTS TO

PMS / CMMSRead: operational KPIs
AccountingRead: financial KPIs
Portfolio Insights / other agentsRead: exceptions and hotspots
Delivery (Slack / email / BI)Publish the narrative
WITHOUT PMS API

Agent reads a KPI export and writes the narrative against the prior export. The comparison logic still runs.

WITH PMS API

Agent reads live KPIs across systems and publishes a what-changed narrative each cycle. Best experience.

For the person who wants to build this agent step by step. Each node below maps to an n8n workflow node.

TRIGGERNode 1
Trigger - Cycle Closes
Cron weekly/monthly at the reporting point.
Output: { period, prior_period }
APINode 2
HTTP Request - Pull KPIs
Gather the KPI set for this period and the prior, plus exceptions from other agents.
Output: { kpis_now{}, kpis_prior{}, exceptions[] }
LLMNode 3
LLM - Compare & Explain
Compute changes vs the comparison basis and draft the why for each notable move.
Output: [{ kpi, now, prior, change, notable, explanation }]
You write a what-changed analysis for leadership.

For each KPI, compute the change vs the comparison basis. For changes beyond the NOTABILITY RULE, set notable=true and write a plain-language explanation grounded in the data and exceptions. No speculation beyond the data.

Respond ONLY with a valid JSON array.

NOTABILITY RULE: {notability_rule}
KPIS NOW: {kpis_now}
KPIS PRIOR: {kpis_prior}
EXCEPTIONS: {exceptions}
OUTPUT
[
  { "kpi": "avg_turn_days", "now": 9.2, "prior": 7.1, "change": "+2.1 days", "notable": true, "explanation": "Two turns at Oak Ridge stalled on vendor confirmation, pulling the average up." },
  { "kpi": "open_wo", "now": 61, "prior": 58, "change": "+3", "notable": false, "explanation": "" }
]
DECISIONNode 4
Condition - Anything Notable?
IF notable changes exist → build the narrative around them. ELSE → publish a steady-state summary.
Output: Branch: NOTABLE (narrative) or STEADY (brief)
ACTIONNode 5
Publish Summary
Assemble the notable changes into a short executive narrative and publish to the leadership channel.
Output: What-changed summary delivered.
MVP vs Advanced — Where To Start
MVP — START HERE
  • Manual run
  • KPI export
  • Prior-period compare
  • Fixed notability rule
  • Single narrative
  • Email delivery
ADVANCED — ADD LATER

    Full Video Walkthrough Coming Soon

    From KPIs to a what-changed narrative - step by step.

    Register to be notified

    Everything on this page is free to learn.

    When you're ready to build — do it yourself or bring us in.

    Register FOR WALKTHROUGHS
    Book a call to scope this

    OTHER AGENTS IN LEASING & MARKETING

    Owner Reporting Agent

    Drafts owner-ready summaries with variance explanations

    Daily Exceptions Brief Agent

    Morning list of what's stuck, who owns it, what's next

    Portfolio Insights Agent

    Flags hotspots - repeat failures, cost spikes, backlog trends

    Frequently asked questions

    Executive Summary and What Changed Reporting Agent is an AI agent that generates weekly/monthly narratives from KPIs and exceptions.
    Leadership gets dashboards. Nobody has time to read them. The Executive Summary and What Changed Reporting Agent addresses this — it generates weekly/monthly narratives from KPIs and exceptions.
    No — your team stays in control. Your leadership interprets and decides. The agent writes the what-changed narrative from the data. It removes the dashboard mining, not the judgment.

    This is 2 of 4 agents in Executive & Portfolio Operations. → · Explore all 10 departments →