The Challenge
The analyst worked at a growing fintech in Canary Wharf — the kind of place where a team of six covers the work that larger firms spread across twenty. His role involved producing weekly sector reports: scanning regulatory filings, earnings calls, market data, and news across three verticals, then distilling it all into a briefing for the investment committee.
"Each report took me about a day and a half," he said. "And honestly, most of that time was gathering and organising information. The actual analysis — the bit where I apply judgement — that was maybe three hours. The rest was grunt work."
He'd tried ChatGPT and Claude directly, but the process was too fragmented. Copy-paste in, copy-paste out. No memory between sessions. No connection to the data sources he actually needed.
The Setup
We deployed OpenClaw on a secure VPS with encrypted storage — a non-negotiable for financial services. The agent connects to his curated RSS feeds, a regulatory filing API, and the company's internal Notion workspace where past reports live.
The workflow we configured runs every Sunday evening:
- Source ingestion — the agent pulls the week's relevant filings, earnings transcripts, and news articles from his monitored sources.
- Trend identification — it highlights recurring themes, notable outliers, and cross-sector patterns, referencing the previous week's report for continuity.
- First-draft report — a structured document in the company's standard format, complete with data tables, source citations, and flagged items requiring human judgement.
The Human Still Matters
The analyst is quick to point out what OpenClaw doesn't do. It doesn't make investment recommendations. It doesn't assess risk appetite. It doesn't understand the nuance of a CEO's tone on an earnings call the way a seasoned analyst does.
"What it does is hand me a 70% complete report on Monday morning. All the data is there, organised, cited. I spend three hours turning it into something I'd put my name on. That used to take a day and a half. It's not replacing my judgement — it's giving me more time to use it."
The Bigger Picture
The investment committee noticed the improvement within weeks. Reports arrived earlier, were more consistently structured, and included more sources than before. The analyst was covering the same ground in less time, with better output.
More importantly, it freed him to do deeper work. He started producing thematic deep-dives — the kind of strategic analysis that actually moves the needle for investment decisions. The work he was hired to do, but never had time for.
"People think AI replaces analysts," he said. "It doesn't. It replaces the part of the analyst's job that isn't actually analysis."