Faster answers didn't fix the retry rate. alibaba's field experiment gave support agents a gen-ai…
faster answers didn't fix the retry rate. alibaba's field experiment gave support agents a gen-ai assistant and got faster replies and better ratings, but customers re-contacted just as often, per the oct 9 write-up.
the metric that matters never moved.
Context
Retention Marketers on September 29 summarised service trials of AI agents and says the available studies do not establish a general improvement in customer retention. One service trial found shorter chats.
A research atlas lists a field experiment on agentic AI and human-in-the-loop interventions in Alibaba's customer service operations.
The specific Alibaba result in the note, faster replies and better ratings but the same re-contact rate, and the October 9 write-up were not found in the pages read, so they are unsupported here, not refuted.
The Alibaba paper that was found studies human-in-the-loop escalation, not a gen-ai assistant for agents.
'The metric that matters never moved' is the author's opinion.
Related work
- Retention Marketers: Do AI agents improve customer retention? ↗Summary of service trials.
- Agentic Work Atlas: Alibaba field experiment ↗The Alibaba study that was found.
- NBER: Generative AI at Work ↗Earlier support-agent study, 5,179 agents.
Watch next
- Find the October 9 write-up and read its re-contact numbers.
Sources
- Retention Marketers, Sep 29, 2026retentionmarketers.com
- Agentic Work Atlas, Sep 19, 2026ntlx.github.io
- Generative AI at Work, NBER working paper 31161docshare.wps.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 10 October 2026 at 01:03 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →