The four heights
The same task, four distances: today's deadline, the next reviewer, the stuck moment, the pattern.
Execute — do the immediate task
+Create a new reporting view for our campaign performance so the growth team does not accidentally…
Execute — do the immediate task
+Create a new reporting view for our campaign performance so the growth team does not accidentally use raw data. The view should include only traffic that arrived with the utm_medium equal to cpc OR utm_source equal to facebook, exclude referral spam from domains ask-dev.com and cheapclicks.net, and set the currency to USD. Confirm the view is returning conversions comparable to the master view for the last 30 days.
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 hand this new filtered campaign view to the paid team, surface the top three campaigns by…
Improve — make it easier to accept
+Before I hand this new filtered campaign view to the paid team, surface the top three campaigns by revenue, make the acquisition channel visible in the header, and flag any campaign where revenue per session is lower than last month. Also call out UTM inconsistencies such as mixed capitalization or missing source values that would make the view undercount traffic.
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
+I just built the cpc/facebook filtered view and reported conversions are 40% lower than the master…
Decide — diagnose the stuck moment
+I built the paid-only view but conversions look low
I just built the cpc/facebook filtered view and reported conversions are 40% lower than the master view for last month; Julian in paid media will blame tagging. I cannot check every landing URL and some legacy ads use inconsistent UTM values. What is the most likely cause and the fastest way to validate whether the view or tagging is wrong so I can answer Julian confidently?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+We repeatedly hand the paid team filtered views that undercount because of messy UTM tagging and…
Become — change the pattern
+Campaign views keep undercounting revenue
We repeatedly hand the paid team filtered views that undercount because of messy UTM tagging and exclusion filters. Where are we losing credibility and time, and what single operational habit change will fix it — including the exact naming convention, a 3-step verification routine, and who owns it?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Next to this one
Other web analytics work people do in Google Analytics.
Every task here came from the work, not from a feature list — which is why the prompts name what you want done and never the button that does it. The tool changes; the work does not.
Copyright © LLOS.ai · 2026 — original pedagogy, voice, and design — all rights reserved.
The rest of the map
Same library, five ways in.