How Publishers Can Test Engagement Signals to Improve On-Site Discovery

Publishers spend significant effort earning visits from search, social platforms and newsletters. What happens after readers reach a homepage, section front or topic page is equally important. The way stories are presented can influence which articles people notice and choose to read.

A reported tactic from The Athletic offers a straightforward example. Selected homepage stories show their comment counts, giving readers a visible cue that an article is generating discussion. As Search Engine Journal's analysis of the tactic explains, that small element can function as social proof: it helps visitors identify stories that other readers consider worth discussing.

The important takeaway is not that every publisher should add comment counts everywhere. It is that engagement signals can be tested as a distribution input on a publisher's own site, where teams can measure whether they improve content discovery and click-through rate.

Why visible engagement can affect story selection

Readers arriving on a content-rich page face a selection problem. Headlines, images, placement and topic relevance all compete for attention. A comment count adds another piece of information: other people have actively responded to this story.

That may make an article feel timely, debatable or especially relevant. In a sports setting, where analysis, club news and match reactions often prompt conversation, the signal can be particularly intuitive. The underlying principle is broader, though. A visible indicator of reader activity may help a visitor decide which item deserves a click.

This is an on-site discovery tactic, not evidence of a direct Google ranking factor. Comment counts may influence clicks on a publisher's owned pages, but the available reporting does not establish that displaying them improves search rankings, organic search click-through rates, subscriptions or retention.

What The Athletic example does and does not confirm

The Search Engine Journal article describes The Athletic as displaying comment counts on selected homepage articles and presents the approach as a way to surface discussion-worthy content. It also proposes using engagement data to influence the prominence of content across homepages, section fronts and topic pages.

There are meaningful limits to the evidence:

• No controlled experiment with numeric results is disclosed in the article.
• The Athletic has not publicly provided a formal rollout timeline for this particular homepage treatment.
• Public documentation specifically detailing comment-count signaling on The Athletic's homepage appears limited outside analysis and opinion coverage.
• A higher click-through rate does not automatically mean readers are more satisfied or more likely to return.

Those caveats make this a useful testing hypothesis rather than a proven universal playbook. The Athletic has also been associated with a broader engagement-focused publishing strategy. For context, TechCrunch's 2018 interview with The Athletic discussed the company's emphasis on subscriber engagement. That historical context does not independently verify the newer homepage implementation, but it helps explain why publishers may value engagement as more than a vanity metric.

A practical starter test for publishers

The simplest implementation is to show comment counts on a limited, representative set of homepage articles, then compare their performance with similar articles that do not receive the treatment. A limited test reduces the risk of redesigning the whole discovery experience around an assumption.

Choose a signal readers can interpret

Comment counts work because they are easy to understand. Other potential signals include likes or shares, as suggested in the Search Engine Journal framework. But a signal should mean something useful in the context where it appears.

For example, a high comment count can indicate active debate, but it can also be driven by controversy. Shares may suggest broad appeal, while likes may indicate approval without necessarily creating curiosity. Publishers should avoid presenting a metric as a quality label if it does not reliably help readers choose relevant content.

Set a clear comparison before launching

Measure click-through rate from the specific page or module where the signal appears. Also review downstream behavior that reflects whether the click was valuable, such as article engagement, return visits or other internally defined quality measures.

Compare like with like where possible. Breaking results down by content type, position on the page, audience segment and traffic source can prevent a prominent breaking-news story from distorting the result. A story at the top of a homepage has advantages that a lower-page story does not, regardless of its visible comment total.

Test prominence, not only the badge

The engagement signal can become an input to editorial distribution as well as a displayed label. A publisher could test whether content showing meaningful reader response deserves more prominent placement on a homepage, section front or topic page.

This requires restraint. Engagement volume alone is not a complete editorial value judgment. It should sit alongside relevance, freshness, reader needs and the publication's editorial priorities. The aim is to make strong content easier to discover, not to create a feedback loop that promotes only the most reactive stories.

Guardrails for using social proof

Visible engagement data can create incentives and unintended effects. Comment activity may be unavailable on some articles, uneven across beats, or affected by moderation policies. Newer stories also have less time to accumulate activity, which can make raw totals favor older content.

Useful safeguards include:

• Apply the treatment only where the metric is sufficiently meaningful and consistently available.
• Consider time-sensitive labels or normalized measures if older articles gain an automatic advantage.
• Keep moderation, trust and editorial standards separate from popularity signals.
• Review whether the treatment disadvantages specialist, service or evergreen content that naturally receives fewer comments.
• Evaluate outcomes over enough content and time to avoid drawing conclusions from a handful of exceptional stories.

A low or absent count can be a negative signal, so teams should decide whether it is better to hide the element below a threshold. That choice should also be part of the experiment.

What this means for search visibility teams

SEO and audience-development teams often focus on gaining the initial visit. On-site discovery is a related visibility problem: once a visitor lands, can they efficiently find the best next page? Better internal content discovery can increase the value of earned traffic without requiring additional paid distribution.

For publishers, this is especially relevant on pages that aggregate many choices. A homepage or topic hub can use reader behavior to complement traditional editorial curation. The best implementation will vary by audience and site architecture, which is why measurement matters more than copying another publisher's interface.

For businesses outside publishing, the same principle may apply to resource centers, product-learning hubs or community content, provided the engagement indicators are genuine, understandable and useful. Artificial or selectively presented social proof would undermine trust.

If you want to assess how discoverable and accessible your site is for both traditional and AI-driven search journeys, create a GEO Search Leads account. It provides a practical starting point for identifying GEO readiness and optimization opportunities, helping your team connect content improvements with the way people and search systems find information.

Frequently Asked Questions

Does showing comment counts improve SEO rankings?

No direct ranking benefit is established by the available reporting. The tactic is presented as a way to improve on-site content discovery and potentially homepage click-through rate, not as a confirmed Google ranking signal.

What should publishers measure in an engagement-signal test?

Start with click-through rate from the page or module showing the signal. Review downstream engagement and compare similar content, placements and audience segments to understand whether additional clicks are valuable.

Can likes and shares be used instead of comment counts?

Yes. Search Engine Journal's proposed framework includes testing other engagement signals such as likes and shares. Publishers should test which signal readers understand and which best predicts meaningful behavior on their own sites.

Should every article display an engagement count?

Not necessarily. The reported approach concerns selected homepage articles, and low, unavailable or uneven engagement data can be unhelpful. A limited test can determine where a signal supports discovery rather than distracting readers.

Conclusion

The Athletic example is a useful reminder that audience engagement can inform more than reporting dashboards. When used carefully, visible engagement signals may help readers navigate crowded content pages. The evidence supports testing the idea, not assuming a result: publishers should measure click behavior, reader value and unintended effects before expanding it across their sites.