Problem: Every third order is lost
The average shopping cart abandonment rate in e-commerce is 65–70%. Out of 1000 sessions with added items, 650–700 orders are not completed. With an average cart value of $40, that's up to $28,000 in lost revenue. The good news: up to 15% of these can be recovered with proper tracking and automated notifications. We set up the full cycle: from server-side detection to conversion reports. Based on our experience, the solution pays for itself within 2–3 months. We guarantee recovery of up to 15% of lost orders if our recommendations are followed.
Why customers abandon carts?
The main reasons are unexpected shipping costs, a complicated checkout form, or lack of a convenient payment method. Without data, you won't know the cause. Abandoned cart tracking gives you exact numbers: who, when, and with which product left. We use a combination of server-side methods and client-side events for complete data. Additionally, we analyze drop-off points through UX events.
How to set up server-side tracking in Bitrix?
Cart data resides in the b_sale_fuser (virtual user) and b_sale_basket (items) tables. To find abandoned carts, query \Bitrix\Sale\BasketTable. The code below checks carts not updated for more than an hour and, if an authorized user exists, pushes a task to the notification queue. According to the Sale module documentation, this approach is recommended for minimal load.
<?php $cutoffTime = new \Bitrix\Main\Type\DateTime(); $cutoffTime->add('-1 hour'); $abandonedFusers = \Bitrix\Sale\BasketTable::getList([ 'filter' => [ '<DATE_UPDATE' => $cutoffTime, '=ORDER_ID' => false, ], 'group' => ['FUSER_ID'], 'select' => ['FUSER_ID'], ])->fetchAll(); foreach ($abandonedFusers as $row) { $fuser = \Bitrix\Sale\FuserTable::getList([ 'filter' => ['=ID' => $row['FUSER_ID']], 'select' => ['USER_ID', 'DATE_UPDATE'], ])->fetch(); if (!$fuser || !$fuser['USER_ID']) continue; $user = \Bitrix\Main\UserTable::getById($fuser['USER_ID'])->fetch(); $email = $user['EMAIL'] ?? ''; if (!$email) continue; AbandonedCartQueue::push($fuser['USER_ID'], $row['FUSER_ID']); } After getting the FUSER_ID, we check if the user is authorized (USER_ID not null). If so, get the email and push the task to the queue.
Agent configuration details
Run the agent every 30–60 minutes to avoid missing any abandoned carts.
// Agent (Settings -> Agents) function checkAbandonedCarts(): string { AbandonedCartService::processNew(); return __FUNCTION__ . '();'; } AbandonedCartService::processNew() is your class that selects new abandoned carts (not marked as processed) and writes them to the notification queue.
For performance optimization, use indexes on DATE_UPDATE and ORDER_ID in b_sale_basket, and on FUSER_ID in b_sale_basket. This reduces load when dealing with large catalogs.
What data to analyze in the report?
After a few weeks, you'll have data: number of detected carts, communications sent, and recovered orders. A simple SQL query shows the trend.
SELECT DATE(detected_at) AS date, COUNT(*) AS detected, SUM(CASE WHEN status = 'recovered' THEN 1 ELSE 0 END) AS recovered, ROUND(SUM(CASE WHEN status = 'recovered' THEN 1 ELSE 0 END) * 100.0 / COUNT(*), 1) AS recovery_rate FROM local_abandoned_cart GROUP BY DATE(detected_at) ORDER BY date DESC; For example, with 1000 detected carts and 150 recovered orders at an average order value of $40, additional revenue is $6,000.
Comparison of server-side and client-side tracking
Server-side detection finds 1.2–1.3 times more abandoned carts than client-side, as confirmed by our projects. But client-side events help analyze UX and drop-off points.
| Criterion | Server-side | Client-side |
|---|---|---|
| Accuracy | High (based on database) | Medium (depends on JS execution) |
| Detection | All abandoned carts | Only during active session |
| Additional load | Minimal (database queries) | Depends on event volume |
| Use in reports | Yes, for recovery | Yes, for UX analytics |
What's included
| Deliverable | Description |
|---|---|
| Agent configuration for abandoned cart search | Setup interval, filters, notification queue |
| Status table for each cart | Record detection date, sent notifications, recovery status |
| Integration with GA4 and Yandex Metrica | Send events on add-to-cart and checkout initiation |
| Abandoned cart conversion report | Daily statistics with detection, send, and recovery metrics |
| Documentation and admin training | Description of system operation and setup instructions |
Implementation process: how we work
- Audit current cart implementation and identify bottlenecks.
- Set up server-side tracking: queries, agent, status table.
- Integrate client-side events (GA4, Yandex Metrica).
- Develop abandoned cart conversion report.
- Hand over documentation and train your administrator.
Estimated timeline
| Phase | Time |
|---|---|
| Server-side tracking + agent | 1–2 days |
| Client-side events | 4 hours |
| Status table and reporting | 1 day |
Full implementation cycle takes up to 3 days. You'll receive the first report a week after start. Contact us for a preliminary assessment of your project. Order the implementation and see the system pay for itself by recovering up to 15% of lost orders.
Our approach
Each task requires individual analysis and careful planning. We don't use template solutions — every project is adapted to specific requirements and existing infrastructure. Our team has experience with projects of various scales: from small shops to high-load platforms with millions of operations per day.
Guarantees and support
We provide a 12-month warranty on completed work. During this period, we fix any issues free of charge. After project completion, we deliver full documentation and training for your team. Technical support is available for 30 days after launch — we'll help resolve any questions.