The metrics most website audits miss
Standard website analytics give you a surface-level picture. You can see how many people visited, where they came from and whether they converted. What they don’t show you is what happened in between: where people lost interest, which sections they actually read, what caught their attention and what they scrolled straight past.
That’s where the less obvious metrics come in. They’re not difficult to access, but most teams never look at them, which means the insights they contain stay hidden even when the tools to find them are already installed.
What counts as an obscure metric
The term is relative. Bounce rate and session duration are routine at this point. The metrics worth paying more attention to are the ones that sit a layer deeper.
Scroll depth tells you how far down a page users actually read. If 60% of your visitors never make it past the second paragraph of a blog post, the length, structure or opening of that content isn’t working, regardless of how much traffic it receives. That’s actionable information, but you only get it if you’re measuring it.
Time spent on specific sections of a page is different from overall time on page. A user might spend three minutes on a page but have that time concentrated entirely in one area, with the rest barely glanced at. Knowing which sections hold attention and which don’t tells you something about what your audience actually finds useful.
Exit rates on specific sections reveal where interest drops off. If users consistently leave from the same point in a flow, that section likely contains something confusing, off-putting or simply not relevant enough to justify continuing.
Interactions with non-clickable elements are another underused signal. If users are repeatedly clicking on an image or a piece of text that isn’t linked to anything, they’re telling you they expected it to do something. That’s either a missed opportunity or a design confusion that needs resolving.
What these metrics have revealed in practice
The findings tend to be surprising, which is exactly why they’re worth looking for.
One e-commerce site used heatmaps to analyse how users interacted with its layout. The data showed that 75% of visitors were ignoring sidebar promotions entirely. The sidebar had been designed as a revenue driver; in practice it was invisible. After redesigning the layout to remove it and redistribute that content more effectively, conversion rates increased by 15%.
A blog analysed scroll depth and found that 60% of readers weren’t getting past the first two paragraphs. Changes to the content structure, including getting to the useful information faster and making the opening more directly relevant to the reader, increased average time on page by 40%.
A service business used session duration data to identify discrepancies between how long users spent on different content types. The findings pointed to specific pages that weren’t holding attention. After overhauling those pages, bounce rate fell by 30%.
None of these findings would have emerged from standard traffic reporting. They required a closer look at what users were actually doing rather than just whether they arrived and converted.
How metrics shape decisions
The value of any metric depends on what you do with it. A high bounce rate on a landing page is a prompt to investigate, not a conclusion. It might mean the content doesn’t match what the visitor expected. It might mean the page loads too slowly on mobile. It might mean the headline isn’t doing its job. The metric points you to the problem; diagnosis still requires judgement.
A/B testing is how you validate the diagnosis. If you suspect a landing page headline is contributing to high bounce rates, test an alternative against the original under real conditions and let the data tell you which performs better. That’s more reliable than making changes based on instinct and hoping to see improvement.
The broader shift this enables is from reactive to proactive. Instead of waiting for performance to decline noticeably and then trying to work out why, you’re monitoring signals continuously and catching issues before they compound.
The tools that make this possible
Google Analytics provides the quantitative foundation: traffic volumes, session data, conversion funnels and user flow. Hotjar adds the visual layer with heatmaps, scroll maps and session recordings that show you exactly how individual users navigated through your site. Tools like SEMrush and Ahrefs extend the analysis to backlinks and keyword performance, which can surface unexpected insights about how your content is being found and by whom.
The practical recommendation is to set up dashboards that surface these metrics regularly rather than requiring someone to go looking for them. Patterns are easier to spot when you’re comparing week-on-week data than when you dip into the analytics occasionally and try to make sense of isolated numbers.
Getting your team engaged with the data matters too. Insights that sit in a report and don’t reach the people making content, design or product decisions don’t change anything. Building a habit of discussing what the metrics are showing, and treating that discussion as a normal part of planning, tends to produce better outcomes than treating analytics as a specialist function that reports upwards periodically.
Where this is heading
AI and machine learning are beginning to change what’s possible in website auditing. Rather than relying on analysts to notice patterns in data, these tools can flag anomalies and trends automatically, surfacing issues that might take weeks to spot through manual review. Emerging metrics like engagement scores and content dwell time are becoming more sophisticated, reflecting not just whether users stayed on a page but whether their behaviour suggests they found it genuinely useful.
The direction of travel is towards more personalised, more granular understanding of how different users experience the same site, which will make it possible to tailor content and experiences in ways that current analytics don’t support. The teams that build good habits around the metrics that are already available will be better positioned to make use of those capabilities as they develop.
Common questions
What are obscure metrics in a website audit? Data points that standard reporting doesn’t typically surface, such as scroll depth, section-level engagement, exit rates on specific page areas and interactions with non-clickable elements.
How can they influence strategy? By revealing what users actually do rather than just whether they convert. They identify content that isn’t working, layouts that confuse users and sections that drive people away, all of which informs more targeted improvements.
Which specific metrics are most worth tracking? Scroll depth, time on specific sections, exit rates by content area, internal link effectiveness and user flow within the site are all worth adding to a standard analytics review.
How can they reshape content strategy? By showing which content holds attention and which doesn’t, allowing you to focus effort on formats and topics that resonate rather than continuing to produce content based on assumptions.
What tools support this kind of analysis? Google Analytics, Hotjar and Crazy Egg all provide features relevant to obscure metric tracking, including heatmaps, session recordings and detailed behavioural analysis.