3 ready prompts
Written for the moment this actually comes up.
Execute — do the immediate task
+We need session duration trends for the last six months to report to the retention team. Calculate…
Execute — do the immediate task
+We need session duration trends for the last six months to report to the retention team. Calculate average session duration by week from January 1 to June 30, break it out by new versus returning users, and highlight any consecutive two-week declines greater than 15%. Export charts and the raw table for the retention deck.
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 put session duration charts into the retention deck, make them easy to interpret for…
Improve — make it easier to accept
+Before I put session duration charts into the retention deck, make them easy to interpret for non-analysts: annotate weeks with marketing campaigns or site incidents, show new versus returning on the same axis, and call out plausible causes for any sharp drops. Provide a short bullet list of three recommended experiments if duration falls for returning users.
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
+Average session duration for returning users fell by 30% in May. The head of retention, Maria,…
Decide — diagnose the stuck moment
+Average session duration dropped suddenly last month for returning users.
Average session duration for returning users fell by 30% in May. The head of retention, Maria, expects a diagnosis. I don’t know whether this is measurement drift, a change in site content, or a cohort effect from a campaign. Compare session duration by device and traffic source for returning users, check for missing event timestamps or new redirects, and recommend the most likely cause plus the next measurement to confirm it before Maria asks for experiments.
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.