The four heights
The same task, four distances: today's deadline, the next reviewer, the stuck moment, the pattern.
Execute — do the immediate task
+I have the quarterly P&L and transaction export from last month and a list of recurring manual…
Execute — do the immediate task
+I have the quarterly P&L and transaction export from last month and a list of recurring manual adjustments we make every close. Send the cleaned-up workbook and a one-page summary of proposed automation steps to Marco in FP&A and to Priya in Ops for approval, with a Wednesday close request. Check that formulas recalc and the pivot snapshots match the source before sending.
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Improve — make it easier to accept
+Before I show this to the CFO, make the efficiency case obvious: put current run rate and headcount…
Improve — make it easier to accept
+Before I show this to the CFO, make the efficiency case obvious: put current run rate and headcount cost per process on the front sheet, surface the savings estimate from automating the three manual adjustments, and flag which assumptions would make the ROI miss. Make the conversion from time saved to dollars explicit and call out implementation risk.
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Decide — diagnose the stuck moment
+The quarter close is on Friday, Marco in FP&A wants a recommendation today, and I cannot prove…
Decide — diagnose the stuck moment
+I just found a $12k monthly manual adjustment buried in a hidden sheet.
The quarter close is on Friday, Marco in FP&A wants a recommendation today, and I cannot prove whether the adjustment is valid without contacting Ops. What is the most defensible recommendation to make now that keeps the books defensible and buys time to validate the adjustment? Who should I tell, what temporary entry should I make (if any), and how to phrase the ask so Ops responds fast?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+Across the last six closes we spend two days per close on three manual tweaks that could be…
Become — change the pattern
+We keep redoing the same manual adjustments each close and losing time to reconciliation.
Across the last six closes we spend two days per close on three manual tweaks that could be automated. I want a habit change that reduces close touchpoints and preserves auditability. Should I standardize and script the adjustments into a single macro/template, demand source system fixes with a prioritized ticket, or lock the close with a one-person review signoff and automation backlog? Outline short steps and the trade-offs for speed, audit risk, and chance of getting Ops to commit.
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Where the evidence lives
Who was seen doing this, and what people really ask.
Bank Reconciliation AnalystBudget AnalystCost AnalystFp And A AnalystTreasury AnalystProduction Manager Eventsalso: Develop ways to improve efficiency
Software tasks in the LLOS Work Atlas come from evidence, never a feature list: careers attested to do the work, real job descriptions, and the questions people actually ask (with their view counts). Facets — feature, workflow, troubleshoot, administer, deploy, scale — are open metadata: the work decides, not a taxonomy.
Copyright © LLOS.ai · 2026 — original pedagogy, voice, and design — all rights reserved.
The rest of the map
Same library, five ways in.