Devin · Cognition
Rank last week's blog posts by engagement from the analytics warehouse
At WorkOS, someone asked Devin in Slack to rank the past week's blog posts. It found the analytics tables, joined them and returned a scored ranking, and flagged that signups from the blog were not being tracked.
What they did
The request covered conversions, time on page, scroll depth and interactions for posts from the last seven days. Devin located the right tables, built the joins and combined the measures into one weighted score. Because blog-to-signup conversions were not set up as key events, it said so and used engagement time and interaction density as stand-ins.
How it went
The poster, who works in applied AI at WorkOS, says the answer came back in under six minutes, and that Devin is now one of the most used agents in the company's Slack after a failed first trial in 2024.
Worth knowing
One example told on an employee's personal blog, with no check of whether the ranking was right. The scoring weights were Devin's choice, and the missing signup tracking means the ranking leans on proxy measures.
Try it with Devin
Using read-only access to <your analytics warehouse>, rank the blog posts published in the last <number> days by <metrics, for example conversions, time on page, scroll depth>. Show the tables you used, how you weighted each metric and the SQL. If a metric is not tracked, tell me and explain what proxy you used instead. Do not change any tracking setup.
Swap the parts in angle brackets for your own details. This prompt is adapted from the case, not copied from the original.
What this label means: User story
Someone using it at work posted about it. Any numbers are theirs, so we write "the poster says".
Summarised from the original source; the link has the full story. The people and companies here are not Wize customers.