Portfolio-Scale Research

A 60-minute intelligence sprint works for one customer. What happens when you have thirty? Fifty? Portfolio-scale research requires a fundamentally different architecture — one that automates the repeatable steps, templates the analysis, and reserves human judgment for the interpretation that AI cannot yet do.

The shift is from artisanal to systematic. Individual deep dives are handcrafted intelligence. Portfolio-scale research is an intelligence factory. The quality per customer may be 80% of a bespoke deep dive, but the coverage is 100% of your accounts instead of the 5-10 you have time to research manually. In most revenue organizations, 80% coverage at 80% depth beats 100% coverage at 10% of accounts.

1. Tiered Account Classification Not every account deserves the same research depth. Tier 1 (top 10% by revenue potential) gets full deep dives with human analysis. Tier 2 (next 30%) gets automated collection with analyst review. Tier 3 (remaining 60%) gets automated monitoring with alert-based escalation. The tier determines the research investment.
2. Templated Research Workflows Build research templates that agents can execute independently: company overview, competitive landscape, dark asset scan, value gap analysis. Each template is a structured prompt that produces consistent output across accounts. CLAWMANDER coordinates the agent fleet that executes these at scale.
3. Automated Refresh Cycles Tier 1 profiles refresh monthly. Tier 2 refreshes quarterly. Tier 3 refreshes on trigger events only. The refresh cycle ensures intelligence stays current without overwhelming analyst capacity. Stale intelligence is worse than no intelligence — it creates false confidence.