Should you really standardize five-point ratings for SEO?

Why does Google insist on ISO format for dates?

The ISO 8601 format is an international standard for representing dates and times. Essentially, it looks like "2023-03-15" or "2023-03-15T14:30:00Z." Google recommends this format because its bots analyze billions of pages written in different languages, with varying date conventions.

When a French site displays "15/03/2023" while an American site shows "03/15/2023," the ambiguity is total for a crawler. The meta tag helps eliminate this uncertainty: you display the date in the format you want for your users, but you give Google a standardized version it can interpret without error. The ISO format becomes the common language between your site and the engine.

Must ratings really be on a five-point scale?

Google refers to a recommendation, not a technical obligation. The schema.org vocabulary supports different rating scales through the bestRating and worstRating properties. Thus, you can technically tag a system on 10, 100, or even reversed scales.

Standardizing on a five-point scale is suggested because it is the most widespread scale in the rich snippets Google displays. In practice, if your internal system uses a different scale, you must explicitly declare the boundaries in your markup. Google will then convert for display in search results. But this conversion creates approximations, and the question remains: how much do these approximations really impact CTR?

What is the connection between this markup and display in SERP?

The schema.org markup for ratings allows for stars in rich snippets. These stars increase the visual footprint of your result and, in theory, the click-through rate. However, Google never guarantees that rich snippets will be displayed, even with perfect markup.

Several factors come into play: competition for the query, perceived content quality, consistency between markup and visible text, and opaque spam filtering criteria for reviews. A site can have impeccable markup and never see its stars displayed. It’s frustrating, but that’s the reality on the ground.

• ISO 8601 format required in a meta tag if the visible format differs from the international standard
• Standardization on a five-point scale recommended for ratings, but not strictly technically mandatory
• bestRating and worstRating properties allow for declaring custom scales in schema.org
• Rich snippet display is not guaranteed, even with compliant markup
• Consistency between markup and visible content is essential to avoid anti-spam filtering

Does this recommendation really reflect observed practices?

On the ground, many international sites already use the ISO format for dates in their meta tags, at least for reasons of internal technical standardization. Modern CMS often manage this conversion automatically. The real problem arises on legacy sites or custom developments where developers never thought about this layer of metadata.

For ratings, the situation is more nuanced. We see sites using scales of 10 (like TrustPilot in certain markets) or 100 still appearing in rich snippets. Google seems to convert on the fly, but the loss of granularity is real. A score of 87/100 turns into an approximate 4.4/5, which can misrepresent an important nuance for a potential buyer. [To be verified]: Google has never published data on the real SEO impact of a 5-point scale versus another well-tagged scale.

What inconsistencies remain in this directive?

Google speaks of a "recommendation" regarding standardization on a five-point scale, yet its own structured data testing tools do not flag any errors when a different scale is used. This ambiguity creates a gray area where practitioners wonder if they risk a penalty or simply suboptimal display.

Another point: the schema.org documentation itself does not favor any particular scale. It’s a Google overlay that may not align with the open standards it claims to follow. In practice, if you are optimizing for other engines or data aggregators, standardizing on five may even be counterproductive. Let’s be honest, no one systematically tests the impact of a scale change on CTR, so we’re flying blind.

In what cases does this rule not apply or pose problems?

For B2B sites with complex ratings or compliance scores, forcing a five-point scale is absurd. A technical certification score does not boil down to nice stars. In these cases, schema.org markup becomes a marketing veneer rather than useful information.

Similarly, some local markets (China, Japan) use culturally entrenched rating systems that do not map onto a Western scale. Google never mentions this, highlighting the limits of its global recommendations. Finally, if your site displays aggregated reviews from multiple sources with different scales, the conversion becomes a technical nightmare and a potential source of bias in data representation.

What should you implement concretely on your site?

For dates, add a meta tag with the content in ISO 8601 format every time you display a date in a localized format. Practical example: if you show "March 15, 2023" on the screen, place a meta tag with content="2023-03-15." Most CMS can automate this conversion through plugins or code snippets.

For ratings, ensure that your schema.org markup includes the ratingValue, bestRating, and worstRating properties. If you are using a scale of 10, clearly declare it: bestRating="10" and worstRating="1" (or 0 depending on your logic). Google will handle the conversion for display, but at least you will have provided the correct metadata. Systematically test with Google’s rich results testing tool.

What errors should you absolutely avoid?

Never allow a discrepancy between markup and visible content. If your page displays "4.7/5 (230 reviews)" but your schema.org states 4.2/5, Google may see this as an attempt to manipulate and not display your stars. Consistency is more important than technical perfection.

Another classic pitfall: using AggregateRating markup for reviews that are not genuinely aggregated or verified. Google has tightened its criteria in recent years and is increasingly filtering out self-reported or dubious ratings. If your reviews come from a single unverifiable source, expect your rich snippets to disappear sooner or later. Also, avoid tagging pages without actual review content: a category page with just an average score but no detailed reviews risks being ignored.

How can you check that your implementation works?

Use Google's rich results testing tool (formerly the structured data testing tool). Paste the URL of your key pages and check that all the required properties are detected without errors or warnings. Note that a successful test does not guarantee display in SERP, but a failed test guarantees that there will be nothing.

Also, monitor your impressions and CTR in Search Console by filtering by type of rich result. If your pages with markup do not get rich snippets after several weeks, dig deeper: either the markup is malformed, the content does not meet Google’s quality criteria, or the competition for the query does not allow for display.

• Add an ISO 8601 meta tag for all dates displayed in a localized format
• Explicitly declare bestRating and worstRating in the schema.org markup of ratings
• Check for strict consistency between visible data and structured markup
• Test each page with Google's rich results testing tool
• Monitor impressions of rich results in Search Console
• Avoid self-reported ratings without verifiable sources or actual aggregation