Custom Review & Rating Module for 1C-Bitrix

We often encounter clients trying to collect reviews using standard components like bitrix:catalog.element.vote and bitrix:forum. The result is poor: an unmoderated forum, ratings without purchase verification, and no structured data for Schema.org. A proper review module is a different story: verified reviews, multi-criteria rating, moderation, seller responses, and aggregated star microdata. Our team, with over 10 years of experience in 1C-Bitrix development and 40+ successful projects, offers a custom module that covers all these gaps.

Why Standard Solutions Don't Cut It

Bitrix has a bitrix:catalog.element.vote component for voting and bitrix:forum linked to elements — these are the building blocks typically used for reviews. They lack built-in verification of paid orders, no moderation, no multi-criteria ratings. Microdata isn't generated, robbing you of the SEO benefit of star snippets. Our module addresses all these shortcomings.

How Purchase Verification Works

The most valuable part of a review is the "verified purchaser" badge. We check against b_sale_order:

public function isVerifiedPurchase(int $userId, int $productId): bool {
    $order = \Bitrix\Sale\Internals\BasketTable::getList([
        'select' => ['ORDER_ID'],
        'filter' => [
            '=PRODUCT_ID' => $productId,
            '=ORDER.USER_ID' => $userId,
            '=ORDER.PAYED' => 'Y',
            '=ORDER.CANCELED' => 'N',
        ],
        'limit' => 1,
    ])->fetch();
    return (bool)$order;
}

When creating a review, order_id can be explicitly passed — then verification is unambiguous. Without order_id, we check any paid order containing that product.

Moderation and Spam Protection

All reviews are created with status pending and enter a moderation queue. Two modes are possible:

• Manual moderation — a moderator approves/rejects each review.
• Auto-approval — reviews from verified purchasers are published automatically; others go to the queue.

Spam filtering uses simple heuristics: links in the body, excessive capital letters, duplicate text, stop words. Suspicious reviews get status spam.

class SpamDetector {
    public function check(string $text): SpamResult {
        if (preg_match('/https?:\/\//i', $text)) {
            return SpamResult::spam('Links are forbidden');
        }
        if (similar_text($text, $this->getLastReview($text)) > 80) {
            return SpamResult::spam('Duplicate');
        }
        foreach ($this->stopWords as $word) {
            if (mb_stripos($text, $word) !== false) {
                return SpamResult::suspicious();
            }
        }
        return SpamResult::clean();
    }
}

Multi-Criteria Rating

Products can have multiple rating criteria (e.g., "Quality", "Matches Description", "Delivery Speed"). The overall rating is the average of all criteria:

SELECT entity_id, AVG(rr.value) AS avg_rating, COUNT(DISTINCT r.id) AS review_count
FROM b_vendor_review r
JOIN b_vendor_review_rating rr ON rr.review_id = r.id
WHERE r.entity_type = 'product' AND r.status = 'approved'
GROUP BY entity_id;

Aggregated data is cached with tag review_entity_{entity_id} and invalidated when a new review is approved.

Schema.org Microdata

For SEO impact, aggregated ratings must appear in microdata. As per Wikipedia: Schema.org, use AggregateRating for products:

<div itemscope itemtype="https://schema.org/Product">
  <span itemprop="name">Product Name</span>
  <div itemprop="aggregateRating" itemscope itemtype="https://schema.org/AggregateRating">
    <span itemprop="ratingValue">4.7</span>
    <span itemprop="reviewCount">143</span>
  </div>
</div>

The vendor:reviews.aggregate component generates this block from cached data. Sites with microdata see up to 5x more clicks on star snippets — proven in practice. You get stars in search results.

Usefulness Voting

Users can mark a review as "helpful" or "unhelpful". Protection against manipulation:

• One vote per IP per day.
• Authorized users: one vote per review forever.
• Usefulness rating influences the display order of reviews.

Development Process

1. Analysis and data model design (ORM tables).
2. Module development: infoblocks, criteria, verification.
3. Moderation, spam filter, voting.
4. Integration with microdata and site components.
5. Testing and deployment.

Comparison: Standard vs Our Module

Timeline

Total: 11 business days. Importing reviews from Yandex.Market or other platforms adds 1–2 days.

What's Included

• Review module with ORM tables, components, and admin interface.
• Purchase verification, moderation, spam filter setup.
• Multi-criteria rating and Schema.org microdata.
• Usefulness voting and seller responses.
• Documentation and staff training.
• Post-launch support.