A regional wealth management firm with 42 relationship managers spent a combined 400+ hours per week generating client reports. The Aizen Event found 5 of 7 workflow steps were automatable. The narrative drafting step was the highest-value opportunity: and the most technically interesting.
7 steps, 7 staff roles, 8 hours per report average. Each portfolio required review, calculation, narrative composition, and compliance sign-off. Senior relationship managers spent 2.5 hours on narrative drafting alone.
In the Aizen Event, we map each workflow step to an AI type: Deterministic (rules with predictable output), RAG (retrieval-grounded), Probabilistic (learned judgment), or Human Required. Classification drives build vs. defer decisions.
| Step | AI type | Recommendation | New state | Time saved | |
|---|---|---|---|---|---|
| 01 | Data collection from CRM/custodian Manual extraction from multiple systems |
Deterministic | Auto-pull via CRM/custodian APIs. Standardized feeds. One-time integration work, then fully automated. Returns data pre-formatted for reporting engine. | ● Automated | 85 min |
| 02 | Data validation & reconciliation Manual rule-based verification |
Deterministic | Rules-based reconciliation: flag variances >0.5%, check totals match, verify benchmark data currency. Exception queue for manual review only when threshold exceeded. | ● Automated | 40 min |
| 03 | Performance calculation & benchmarking Fragmented Excel-based calculations |
Deterministic | Unify into single calculation service. Performance, attribution, and benchmark comparison computed in one system. Minor time savings, high consistency gain. | ● Automated | 20 min |
| 04 | Narrative drafting Manual portfolio commentary by RM |
Probabilistic | Highest-value step. LLM generates commentary from portfolio data, market context, and client-specific voice profile. RM reviews, edits, approves. Full audit trail. Average edit rate: 22% of sentences. | ● Augmented | 2.1 hrs |
| 05 | Compliance check Manual regulatory language review |
Deterministic | Run regulatory language library against draft. Flag prohibited phrases, check required disclosures present. Compliance officer reviews flagged items only (~1.8% of reports). | ● Automated | 30 min |
| 06 | RM review & approval Final RM sign-off and accountability |
Human Required | Final review and approval cannot be removed: RM is accountable. But AI pre-populates edits and highlights only changed sections, reducing review from 30 to 8 min average. | ○ Unchanged | ~22 min |
| 07 | Client delivery & formatting Manual template application and portal delivery |
Deterministic | Auto-format to client template, generate cover letter, deliver via secure portal, log in CRM. Zero manual handling for standard workflows. | ● Automated | 18 min |
We plot each step by implementation complexity (X) against business impact (Y). Steps in the top-left are Quick Wins: start here. Top-right are Strategic Investments: build after wins are live. Bottom-right: skip entirely.
API integration. High impact, moderate complexity. Phase 1.
Rules engine. Fast build.
Library matching. Built alongside validation.
LLM with RM voice profile. Highest ROI. Phase 2.
Required approval. Not automatable.
The redesigned workflow after the Aizen Event implementation. Every step with a new state shows the before design, what changed, and who now handles what.
Build cost, deployment timeline, and three-year return: based on 42 RMs × 50 weeks/year, blended labor rates, and measured time savings post-deployment.
This case study represents a composite of typical enterprise portfolio reporting engagements. Metrics are based on observed industry benchmarks and comparable implementations. All investment figures are illustrative ranges based on the Otonmi service tiers.
Tell us about the process your team is spending the most time on. We'll classify every step and tell you what's worth building: and what isn't.