Newsrooms Deserve to Have Fun with Their Numbers
One of the most exciting things for me around 2012 to 2015 was watching the GA realtime numbers spike up after I published a story. A dopamine spike, sure. But it was more than that. I’d publish a story, and once it got a slot on the homepage or went out on social media, I kept checking the GA realtime numbers to see if it climbed. After a while, I started looking at both the realtime traffic and the engagement metrics for my stories. If I saw a dip in time on page, or an increase in bounce rate, I’d go back to the story, add or remove something, change the title, and republish.
Then I started to obsess about comments. I always looked at how many useful comments a story got and how many of them added new perspectives to it. I felt the inner need to write stories that would satisfy all of it: a spike in realtime traffic, better time on page, fewer bounces, more useful comments. Thinking about it now, I feel contradicted: the story should come first, not the numbers, and yet chasing those numbers was the most alive I felt in the job. For me, back then, this was all fun. The more hands-on I got with analytics, the more fun I had.
This was the seed for my interest in the business of news. Once I moved to products, I started seeing the other side of the newsroom. Till date, I have seen many analytics and dashboard tools for newsrooms, and I always sense friction when they’re presented to editorial. You either overload the cognitive space of the editorial folks and flood their brains with data, or you ask so many questions that they feel intimidated. I’ve had very productive metrics discussions with newsrooms, and bad ones too.
The one question I still carry: why are they not excited about this the way I used to be? Why can’t I replicate that emotion now?
It occurred to me yesterday that the metrics and the dashboards are not helping the newsroom ask better questions of itself. None of it appeals to their intuition. And maybe some of them are excited, but there’s no space where they can safely express it.
Some tools have come close. But I think the newsroom needs a much more personal space to play around with metrics, so they can introspect and ask better questions of themselves.
So I put all of this into a document and built it as a skill file that interprets newsroom metrics. Journalists, sub-editors, digital editors, and editors can load it into their choice of LLM tool and play around with it.
What the skill does when you talk to it:
Translates metrics into plain editorial language. “Engaged sessions” becomes “people who actually read it.”
Warns you about the trap in each metric, the wrong conclusion each number invites.
Compares your work against your own history, never against made-up industry benchmarks.
Suggests questions to bring to your next editorial meeting, about the work, never about a person.
Refuses to rank journalists by traffic or tell you what to cover. That judgment stays yours.
The primary user for this skill is journalists and editors, not digital business. Do get permission from the relevant HODs before using it, from a data security and governance perspective. If you want it to pull numbers automatically from tools like GA4 or GSC, you might need some tech help to connect them. You can also manually import metrics and analyze them. The skill guides you through both.
I built this hoping it gives you what those realtime numbers once gave me. Try it, and have fun with it. That’s the whole point.
If you know a newsroom that could use this, share it with them. And if you try it, tell me what worked and what didn’t. I’m still building.
Get the skill here: https://github.com/RajaRamamoorthy/newsroom-metrics-interpreter (the README covers installing it on claude.ai, Claude Code, and other surfaces).