Relevant
Matches the executive’s field and point of view.
Case study 06 / Fractional executives
20,000 articles scored daily.
The right stories by 8am.
A news-intelligence system monitors 3,000 publications, ranks stories by importance and relevance, drafts a timely point of view, and delivers it to each executive in Slack.
3,000 sources monitoredExecutive-specific scoringSlack-ready first drafts
01
3,000
news sources monitored continuously
02
20K+
published articles processed each day
03
~667h
manual scanning equivalent every day
04
By 08:00
ranked stories and draft takes in Slack
The attention problem
Track every development.
Miss no commentary window.
Fractional C-level executives wanted to build thought-leadership brands around timely commentary in their specialist fields. Their authority came from reacting to developments while those developments were still relevant.
The limiting factor was attention. No executive had the bandwidth to monitor thousands of publications, separate repeated coverage from new information, judge what their audience would care about, and draft a useful point of view before starting client work.
This did not have a credible manual equivalent. Even a shallow two-minute scan per article would consume roughly 667 human hours every day, before research, prioritization, or writing began.
The system
Scan 20,000 stories.
Deliver the few worth saying.
Collection creates the universe. Deduplication reduces noise. AI scores what remains against the executive, their audience, the engagement opportunity, and the remaining commentary window.
News intelligence / daily run
From the entire news cycle to the right executive take
3,000
sources monitored
20K+
articles processed daily
~667h
manual scan equivalent
08:00
daily Slack delivery
Scoring before generation
Rank the opportunity.
Then draft the take.
The system did not generate commentary for everything it found. It first decided whether a story deserved the executive’s attention. Only the strongest candidates earned a draft.
Important
Material enough for their audience to care.
Engaging
Contains tension, novelty, or a useful disagreement.
Timely
Still inside the window for relevant commentary.
The human decision
Open Slack.
Choose write or skip.
Every morning, each executive received a compact brief containing the original story, why it mattered to their audience, an engagement score, and a first-draft take. The system removed discovery and blank-page work. The executive kept the final judgment and voice.
No feed browsing
The relevant story arrives instead of waiting to be found.
No blank page
A drafted angle creates a useful starting point, not a finished opinion.
Voice stays human
The executive decides whether to write, change the angle, or skip it entirely.
Scale, made tangible
Manual scan assumption
2 min
A deliberately shallow read per published article.
Daily article volume
20,000+
Before deduplication, relevance, and quality scoring.
Equivalent daily effort
~667h
More than 83 eight-hour workdays, every single day.
Back-of-the-envelope comparison: 20,000 articles × 2 minutes ÷ 60.