Client case study · Custom AI operations system

Chris Lapa
Chris LapaCustom AI operations system

An agency’s expertise, ready to use again.

A connected knowledge base and ten reusable AI skills for recurring agency work.

← See all case studies

The challenge

Recurring briefs, meeting follow-up, emails, and reports need the same context repeatedly: the client, the agency’s judgment, and Chris’s voice.

What we built

Ruth built Chris OS end to end: a Notion knowledge base and approval structure, an operating manual, and ten skills covering routing, advertising briefs, creative briefs, email drafts, strategy, meeting processing, reports, check-ins, and a morning digest. Automation integrations were also prepared.

The delivered result

A reusable system for turning source material into structured drafts with human review. The project records distinguish assisted workflows from staged automation. Handoff and post-transfer validation remain unfinished, so this is not a claim that every workflow is running unattended.

Estimated time savings

Illustrative capacity estimate: 3–5 hours per week, if the system is used for the workload below. This is a planning model, not measured client performance. It includes human review and excludes setup, maintenance, and exception handling.

Illustrative weekly workload; minutes include review
TaskWeekly volumeManual → assistedTime released
Meeting recap525 → 10 min75 min
Client email / check-in515 → 7 min40 min
Campaign / creative brief245 → 20 min50 min
Report draft140 → 15 min25 min

Model total: 190 minutes, or about 3.2 hours per week. At 1.5 times this workload, the estimate is 285 minutes, or about 4.8 hours. Actual use and review time must be measured before calling this a realized saving.

Start with your business

Bring the service.
We’ll look at the marketing.

Tell us what you sell, who buys it, and what keeps slipping. We’ll work out whether Back Office or a focused project fits.

Book a marketing call