Professional-services AI Coordination Playbook
Coordination practices for AI agent teams managing professional services firms -- consulting engagements, time tracking, resource allocation, proposal pipeline, client reporting, knowledge management, and client relationship continuity. Built for the unique economics of selling and delivering expert human time.
Engagement Delivery
Engagement Health Dashboard for Partners
The partner dashboard agent compiles a single view across all active engagements: budget health (green/yellow/red), client satisfaction trend, resource utilization, risk count, and upcoming milestones. A partner managing 8 engagements should not need to attend 8 status meetings to know where problems are. The dashboard surfaces the 2 engagements that need attention.
What goes wrong without this
A partner manages 12 active engagements. They attend 3 status meetings per week and read 4 email updates. They miss that Engagement #7 is 40% over budget because that PM sends updates biweekly. The partner finds out at the monthly financial review. By then, the overrun is unrecoverable.
Engagement Phase Gates with Mandatory Sign-Off
The engagement agent enforces phase transitions: Discovery to Planning, Planning to Execution, Execution to Delivery, Delivery to Close. Each gate requires sign-off data from specific agents (scope confirmed by client, resources allocated by staffing agent, deliverables reviewed by quality agent). No phase advances without the gate conditions being met. Skipping gates is how firms deliver work the client never asked for.
What goes wrong without this
A consulting team starts execution before the scope document is signed by the client. Three weeks in, the client disputes the deliverables. The firm has to absorb $40K in rework because the planning gate was skipped and nobody confirmed scope alignment.
Risk Register with Automated Escalation
The risk agent maintains a register for every engagement: identified risks, probability, impact, mitigation plan, and owner. Risks are reviewed weekly. A risk that has been "high probability, high impact" for 2 consecutive weeks without mitigation progress triggers an automatic escalation to the engagement partner. Risk registers that are never read are organizational theater.
What goes wrong without this
The project manager identifies a data quality risk in week 2. They add it to the risk register. Nobody reviews the register. Week 6, the data quality issue materializes. The project is delayed 3 weeks. The partner asks "why did nobody flag this?" The PM says "it was in the risk register." The register existed but the escalation process did not.
Scope Change Detection and Pricing Trigger
The project agent monitors every client request against the signed SOW. When a request falls outside the defined scope, it flags the project manager with: the specific request, the SOW boundary it exceeds, estimated additional effort, and a draft change order. Scope creep does not happen in one big moment. It happens in 20 small "can you also..." requests that nobody prices.
What goes wrong without this
The client asks for "a few extra slides" in the final presentation. Then "a small additional analysis." Then "one more stakeholder interview." Each request takes 4-8 hours. Over the engagement, 60 hours of unscoped work accumulate. The project overruns by 25%. Nobody issued a change order because each individual request felt too small to push back on.
Stay in the loop
Get weekly coordination intelligence updates. No account required.