Most content teams
look at the same four numbers:
- Sessions.
- Pageviews.
- Bounce rate.
- Time on page.
We look at them regularly, we watch them go up and down, we are scholars of their whims and like astronomers we draw conclusions from their movement.
Most of those conclusions are, well. Wrong.
Not because the numbers are wrong — the numbers are numbers, they’re accurate. We’re mistaken because the numbers are being read in isolation from each other and from context that gives them greater meaning, and because the most important things GA4 and GSC can tell you about your content are not in the default dashboards.
Let’s put on a show. We’ll have a little awards show all about what we are missing.
The Metric You’re
Reading Wrong: Bounce!
Bounce rate is one of the most consistently misread metrics in content analytics. High bounce rate is assumed to be bad. This assumption is frequently incorrect.
A high bounce rate on a blog post that ranks for an informational query, fully answers that query in the content itself, and sends the reader away satisfied is not a performance problem. It is a performance signal.
The reader came, got what they needed, and left.
That is, you see, what good content does.
A high bounce rate on a product page, a lead gen page, any page where the intended outcome is further action — that is a problem worth investigating.
The question is not “is the bounce rate high?” The question is “does the bounce match the intent?”
Informational content should have high bounce rates.
Conversion-oriented content should not.
So you see how reading them the same way produces bad conclusions in both directions.
GA4 has largely replaced traditional bounce rate with engagement rate, which measures the percentage of sessions with meaningful interaction — at least ten seconds, a conversion event, or two or more pageviews.
This is a better metric for most content purposes. If you’re still reporting bounce rate to stakeholders, I would say stop.
The Metric You Are
Probably Not Watching:
Scroll Depth!
Scroll depth tells you how far down the page readers are getting before they leave. Most teams don’t have this set up.
This is a significant gap.
A page with five hundred sessions and average scroll depth of 20% has a different problem than a page with five hundred sessions and average scroll depth of 80%.
In the first case, something in the first screen is failing — the headline, intro, page load. Contentful paint is such a funny term to me. Not sure why. How are you feeling? Kind of Contentful. What was the name of your 2011 app again? “Contentful Paint.”
In the second case, the content is being consumed but it’s not converting to whatever the next action is. See, you thought I would say 80% is good, but sometimes it’s not. Sometimes it’s worse.
You cannot distinguish these problems from just sessions and bounce rate alone. You need scroll depth.
Set it up in GA4 as a custom event, or use a tool that tracks it natively. Run it for thirty days on your top-traffic pages and I promise what you find will surprise you.
The GSC Report
Nobody Opens:
Search Appearance!
Google Search Console’s Search Appearance report shows you which rich result types your content is appearing in — featured snippets, FAQs, sitelinks, video results.
Most teams check this either: occasionally or never.
What it actually says: where you are winning structured result placements — and where you are eligible to win but are not winning.
A page that appears in position three in standard SERP but has zero rich result appearances despite containing FAQ content, is telling you, “My schema implementation is missing or broken!“
The Search Appearance report is a schema audit in data form, so you know I’m into it. Run it monthly. Cross-reference it against your pages that should be generating rich results.
Fix the gaps. Get to Scheming.
The GSC Report
Hiding the Real Story:
Queries!
Every content team looks at the Queries report in GSC, of course. But most are looking at it wrong in my opinion.
The default view shows you the queries that are driving the most clicks. This is useful — It is not the useful thing.
Sort by impressions not clicks, and filter for queries with high impressions and low click-through rate.
These are queries where you are appearing in search results and not being chosen. This is a title and meta description problem, basically without exception. Your content is findable for these queries, good, but the search result snippet is not compelling enough to earn the click. Bad.
Now: filter for queries with zero or near-zero clicks despite ranking high, like on page one.
These are queries where your ranking is strong, but the query-to-content match is broken — searchers are seeing your result and immediately thinking “This is nice, but it’s not what I need.”
Are they right or are they wrong? Either way this is a problem, and it’s either a content mismatch or a title mismatch.
These two filters, run monthly, will surface more actionable optimization opportunities than any other report in GSC. Isn’t that grand? And the queries they surface feed directly into keyword research that actually tells you something — you’re no longer guessing at what your audience searches for, you’re reading it back from evidence. For a deeper orientation to GSC as an editorial intelligence tool rather than a traffic dashboard, your best content brief is already in Google Search Console and most teams don’t know it’s there.
The Thing no
Dashboard Shows:
What People Do
After They Leave!
This is the gap neither GA4 nor GSC can close, and it is worth naming. Why the Google I/O changes made the analytics blind spots worse is worth reading alongside this — the I/O announcements added a new category of zero-data content consumption that doesn’t exist anywhere in your dashboard. The structural version of this problem goes further still: it’s not just that certain journeys are invisible, it’s that the entire attribution model the dashboard is built on no longer matches the environment it’s measuring.
Your analytics show you what happens on your site. They do not show you what happens because of your site.
The reader who found your content, read it carefully, closed the tab, thought about it for two weeks, and then contacted you — this journey is invisible to your dashboard.
The attribution goes to whatever they did immediately before the conversion, which is often a branded search or a direct visit that obscures the original touchpoint.
This is the measurement problem that content teams deal with constantly, and that no tool has fully solved.
The honest and really unsatisfying answer is that content performance measurement is incomplete by definition, and any attribution model that claims otherwise is making assumptions in its own favor.
Report what you can measure.
Acknowledge what you cannot.
Build the qualitative evidence — customer conversations, sales team feedback, the questions that disappear from support tickets after a piece publishes — alongside the quantitative data that people will listen to.
The full picture requires both. You are required to master both, and to present both.
Therefore this is your problem to solve: How to supplement the incomplete data, which is still and will always be incomplete even after doing the cool GA4 and GSC things you have learned today, with a compelling argument that meets your interlocutor where they are and gets them to where you need them to be. A Content Problem, in other words.
Good thing content is your whole deal! For the full SEO context that makes this data meaningful: SEO for content teams.
I write about content strategy, editorial leadership, and the infrastructure that makes search performance possible. For inquiries: jacob@cliftoncreative.agency · Book a discovery call