- Sessions and pageviews describe attention, not value. Pair them with conversion data or you're flying blind on ROI.
- The real scoreboard has two halves: whether content touches pipeline, and whether your brand shows up in search and LLM answers.
- A mention in an AI answer is not the same as a recommendation. Track both, plus citations, separately.
- GA4 and Google Search Console answer different questions. You need both, set up with a report designed for content, not the default UI.
- For many B2B and dev-tool teams, roughly half of conversions touch content somewhere in the journey — a number worth checking against your own CRM data.
Introduction
Somebody on your team pulls up Google Analytics before a stakeholder meeting, and the report says sessions are up 12% quarter over quarter. Nobody asks the follow-up question, because nobody wants to admit they don't know the answer: up 12% and doing what, exactly?
Content analytics has a scoreboard problem. Most teams default to traffic because traffic is easy to pull and easy to explain. But traffic doesn't tell you if content moved someone toward a demo, a signup, or a purchase decision. It doesn't tell you if your brand shows up when a buyer asks ChatGPT what tool to use.
This piece lays out what content analytics should measure — pipeline influence and visibility in search and LLMs — and why sessions alone can't carry that weight.
Sessions are not a business outcome
Sessions, pageviews, and time on page measure whether people showed up and stuck around. They're useful diagnostic signals, but they stop short of the question a CFO or CRO asks: did this content contribute to revenue?
Content doesn't convert the way ads do. There's no clean click-to-purchase path, no A/B test you can run on an entire blog post, no consistent site of engagement the way there is with a paid campaign. Buyers read a blog post, come back three weeks later, read a case study, ignore your emails for a month, then request a demo. Traffic metrics can't trace that path. Something else has to.
That something else is a measurement layer built to connect content to pipeline and to visibility — the two things that actually indicate content is working.
The pipeline half of the scoreboard
The pipeline half of content analytics answers one question: is content part of the journey that leads to a conversion? Not "did this exact post cause this exact deal," which is a harder and less useful question, but "how often does content show up somewhere in the path to a converted lead?"
Generalizing a bit beyond any single client result: for a lot of B2B and developer-tool companies, something in the neighborhood of half of conversions touch content somewhere along the way. That's a big enough share that ignoring it in your reporting means you're systematically undercounting content's contribution.
To get a number like that for your own site, you need conversion tracking tied to content views, not just channel-level traffic. The four reports that show content ROI lay out how to build this: one on overall inbound engagement (content-touched conversions inside GA4), one on funnel-stage engagement, one on warm-lead inbound tied to CRM data, and one on LLM visibility and engagement. Each answers a different piece of the pipeline question, and together they replace the single traffic number most teams still lean on.
The mechanics matter here. A conservative version of this report looks at users who land on a blog or resource page and later complete a key event, like a demo request or a signup. That's not a perfect causal claim, and it doesn't need to be. It quantifies influence without overstating it, and it gives you a defensible number to bring into a stakeholder meeting instead of a vague claim that content is important.
The visibility half of the scoreboard
The second half of the scoreboard is visibility: does your brand show up when people search, and does it show up when people ask an LLM?
Traditional search visibility is still measurable through familiar tools. Google Search Console tells you about impressions, clicks, and query-level performance. But LLM visibility requires a different kind of measurement, because ChatGPT, Claude, and Gemini don't hand you first-party query data the way Google does.
To track LLM visibility, you run a set of strategic queries through the major models, then quantify how often your brand shows up in the answers. That gives you a visibility percentage by topic, and it lets you compare that percentage against competitors over time. The full process for how we track LLM brand visibility walks through picking topics, generating queries, running them through multiple models, and averaging the results into a usable metric.
Mentions are not recommendations
One distinction matters enough to call out on its own: a mention is not a recommendation. An LLM can mention your brand in a list of eight vendors, describe you as a legacy option, or even warn a reader away from you, and that still counts as a mention if the brand name shows up in the response. Recommendation scoring looks at the language surrounding your name — words like "recommend," "top choice," or "worth considering" — to figure out whether the model actually positioned you as a fit for the question being asked.
Citation scoring is a third, related signal: whether a model used one of your pages as a source, which can happen with or without a recommendation attached. Treating all three as interchangeable will flatter your reporting and hide where the real work is. The distinction between mentioned versus recommended is worth understanding before you build a dashboard around a single "visibility" number.
LLM traffic behaves differently than search traffic
Beyond mentions and recommendations, there's the traffic that clicks through from an LLM citation. This traffic tends to be a small slice of total organic traffic today, but it also tends to convert at meaningfully higher rates than traditional search traffic, since someone who clicked a citation from an AI answer is often further along in their decision. A dashboard for LLM traffic shows how to isolate this segment inside your existing analytics so you're not lumping it in with generic organic traffic and losing the signal.
Why GA4 alone can't carry this
GA4's default interface makes it hard to answer even simple questions, like how one page performed last quarter, without digging through several layers of custom reports and exploration tabs. That's a UX problem, but it becomes a strategy problem when the difficulty of getting an answer means teams stop asking the question at all.
Google Search Console and GA4 also answer different questions, and neither one alone gives you the full picture. Search Console tells you how you're doing in Google Search specifically — impressions, clicks, query-level detail, indexing issues. GA4 tells you about behavior across all your channels once someone lands on your site. Neither one tells you whether content is driving pipeline or whether you're visible in LLM answers. You need Search Console next to Google Analytics, built into a report designed around content questions rather than the default dashboards either tool ships with.
Building the scoreboard without archaeology
None of this requires an entirely new tech stack. It requires instrumenting the stack you already have — GA4, Search Console, your CRM, and whatever LLM query tracking you set up — so the data answers the pipeline and visibility questions instead of just reporting traffic. That's the work of getting your team to instrument the marketing stack properly in the first place, so the numbers you pull are trustworthy rather than a patchwork of half-configured tags and default reports.
If you want a faster way to see this without digging through GA4 menus every week, the ércule app connects GA4 and Search Console into one interface so you can look at page-level and sitewide performance, including organic and LLM traffic, without re-building the same custom report every quarter.
Conclusion
Sessions tell you people showed up. They don't tell you if content moved anyone toward a decision, and they say nothing about whether your brand is visible when someone asks an AI model for a recommendation. The fix is a scoreboard built around two questions: did content touch pipeline, and are you visible — and recommended — in search and in LLMs.
Getting there means setting up conversion tracking that ties content views to key events, building an LLM visibility report you can run consistently, and putting GA4 and Search Console side by side instead of treating them as interchangeable. It requires the discipline to measure the right things and report them the same way every time.
Frequently asked questions
What's wrong with using sessions as the main content metric?
Sessions measure whether people showed up, not whether they moved toward a business outcome like a demo or signup. You can have rising sessions and flat pipeline influence at the same time, which is exactly the gap a report like how to show content ROI: 4 reports is built to close.
How do I know if content is actually influencing conversions?
Track users who land on content pages — blog, resources, docs — and later complete a key event like a demo request. That gives you a conservative but defensible estimate of content's contribution to pipeline, without claiming a single post caused a single deal.
Is LLM visibility worth tracking if the traffic volume is still small?
Yes, mainly because the traffic that does come through tends to convert at a notably higher rate than typical search traffic, since it often reflects someone further along in the decision process. A dashboard for LLM traffic can isolate that segment so you're watching it separately from generic organic numbers.
What's the difference between a brand mention and a brand recommendation in an LLM answer?
A mention just means your brand name showed up somewhere in the response, even in a list of competitors or a warning. A recommendation means the model's language actually positioned you as a good fit for the question. The distinction between mentioned versus recommended matters because treating the two as the same metric can make your visibility look stronger than it is.
Do I still need Google Search Console if I already have GA4 set up?
Yes. They answer different questions — Search Console covers your performance specifically in Google Search, while GA4 covers behavior once someone's already on your site. Reporting that puts Search Console next to Google Analytics gives you a fuller picture than either tool alone.
Why does GA4's default setup make this harder?
The default GA4 interface takes several steps to answer even a simple page-level question, and a lot of the data it surfaces isn't directly useful for content decisions. Getting to the reports described here means designing a specific report rather than relying on GA4's out-of-the-box views.
Where do I start if none of this is instrumented yet?
Start by getting your GA4, Search Console, and CRM data properly connected and flowing to the right places, since a lot of "we don't have the data" problems are actually instrumentation problems. That's the work involved in helping teams instrument the marketing stack before building any reporting on top of it.
Related reading
- How to show content ROI: 4 reports — walks through the specific reports referenced here, including funnel-stage and CRM-based tracking.
- A dashboard for LLM traffic — shows how to isolate and monitor LLM-driven traffic inside your existing analytics setup.
- LLM brand visibility: how we track it — details the query-based process for turning LLM mentions into a trackable visibility metric.
