Not hustle. Compounding.

Measure: Daily voice share and voice plus typed character volume; coverage, weekly change and weighted period averages.

What we know: October 4: 40,161 voice / 72,618 total characters = 55.3%. Ghostty on one Mac only. Hardware typing excludes pasted text; completed Dictation Pal transcription is not paste confirmation. Typing coverage starts September 21.

Measure: Clarification rounds, correction rate and elapsed time on comparable tasks.

What we know: Practice supplied by Thomas. No paired task comparison is recorded for this playbook.

Measure: Task completion time, errors and window switches on matched work.

What we know: Workspace practice supplied by Thomas. Screen count alone does not establish a productivity gain.

Measure: Time without connectivity, detection-to-recovery time and successful unattended handovers.

What we know: Independent recovery and notification workers are installed. The earlier physical trial had approximately nine minutes offline. Repeated physical handover with the revised system is still being validated.

Measure: Manual interventions per completed task, rework and recoverable failures.

What we know: CLI/API workflows exist; no comparable before-and-after intervention series has been assembled.

Measure: Daily sleep duration and device score, bedtime consistency and observed-night coverage.

What we know: Sleep source checked October 5: 78 nights with score and duration over the endpoint’s July 7–October 5 window (91 inclusive dates). The existing pillar chart uses the last logged night per week, not a weekly average. Productivity effects have not been established.

Measure: Completed 13-minute sessions and a consistent attention task before and after an eight-week observation period.

What we know: The referenced study reported effects after eight weeks, not four, versus its podcast control. This research does not establish Thomas’s personal outcome.

• Huberman focus episode (13-minute practice at 18:10)
• Basso et al., Behavioural Brain Research (2019)

Measure: Exact product, reason for use and evidence review; distinguish purchase, use and observed outcome.

What we know: Topics specified by Thomas: fish oil/omega-3, already-owned creatine, and alpha-GPC. Matching October 5 order entries identify Sports Research Omega-3 Fish Oil and Nutricost Alpha GPC. A purchase does not establish use or benefit. The episode links an observational stroke-risk association for alpha-GPC; association is not proof of causation. No cognitive benefit is established here. Discuss suitability with a qualified clinician.

• Episode: omega-3 and creatine at 27:00; alpha-GPC at 29:00
• Alpha-GPC observational study, JAMA Network Open (2021)
• NIH: omega-3 evidence and safety

Measure: Repeated task time and errors, minus build and ongoing maintenance effort.

What we know: User-supplied design principle. No consolidated return-on-time calculation has been made.

Measure: Meeting minutes per week, decision latency and rework from misunderstandings.

What we know: Calendar and task sources exist. A validated meeting-versus-async comparison has not been assembled.

Measure: Decision and handoff latency, blocked time and ownership completeness.

What we know: Thomas’s operating philosophy. Metrics require definitions and a baseline before a chart can show improvement.

Measure: Monthly outcomes, report completion and blocker age, paired with initiative and learning evidence.

What we know: User-supplied operating model. No employee performance or compensation data is published.

Measure: Demonstrated new capabilities, increased autonomy and useful outcomes over successive reviews.

What we know: A leadership philosophy using a football-player analogy. Capability growth, market value and retention are distinct outcomes; none is inferred from compensation alone.

Measure: Completed and blocked handoffs, repeated briefings, source freshness and stale-context corrections.

What we know: Tao is the Head Coach; Thy is the main executive assistant. Scoped source connections and memory exist. Context can be incomplete or stale; aggregate outcome reporting still needs a baseline.

Measure: Capture-to-search delay, successful retrieval and corrections, with consent and source provenance.

What we know: The matching October 5 order entry identifies Plaud NotePin S, a wearable recorder for in-person conversations. Its connection to the personal context engine has not been verified. Capture and ingestion still need a measured trial; no raw recordings or private context are published.

• Plaud NotePin S: official product description

Measure: Actual Zone 2 minutes overlapping listening, followed by next-day recall or an applied idea.

What we know: Run and cycling activity records include a zone_2 field; its presence alone does not prove time in that zone. Listening overlap and recall are not measured. Use the named zone system from the training source.