Yandex Metrica ecommerce tracking: see orders from search

Why ecommerce reports matter for a store

Standard traffic stats answer "how many came." Ecommerce reports answer "what sold and where from." For an electronics shop in Tashkent or a fashion store delivering across the regions, that is the gap between "traffic is growing" and "revenue is growing."

It matters even more when you pay for search, Instagram ads and Google Ads at the same time. Every channel is happy to claim the same order, and a product report helps you see where real sales happen and where there are only clicks.

How the data reaches your reports

Metrica knows nothing about your catalog by itself. Your site has to send it what shoppers do, through a data array on the page, usually called dataLayer. In the counter settings you switch ecommerce on and enter the name of that array; from then on Metrica reads whatever lands there.

If nothing is sent, the reports stay empty no matter how many visits you get. So setup is a shared job: the marketer specifies the events and fields, and the developer adds sending on the product page, in the cart and on the order-success page. Many platforms and plugins do this automatically, but you still need to verify the result by hand.

What to send with each event

The tidier the structure, the more useful your reports will be six months from now. A minimum set for a product and an order:

• product ID and name;
• category, in the same hierarchy as your catalog;
• brand, if your range is built around brands;
• price as a plain number, with no spaces or "UZS" text;
• quantity;
• order number at the moment of purchase.

What to look at for search and ads

A product report filtered to search engine visits shows which categories and product pages sell from organic search. That leads to clear decisions: strengthen the sections where search already sells, and stop spending effort on pages nobody looks for.

Another signal is products with many views and few purchases. The cause is usually the page itself rather than promotion: weak photos, unclear price, no information on stock or delivery. If people add to the cart and then vanish, look at checkout, not SEO.

When orders happen in Telegram or by phone

In Uzbekistan, a good share of buyers never reach the cart. They message a shop in Telegram, call to confirm availability, and pay on delivery or through Payme or Click. Ecommerce tracking cannot see those sales, so the channel that brought them looks useless.

Add goals for phone calls, messenger button clicks and lead form submissions. For a full picture, upload closed orders from your CRM into Metrica if your system allows it. Then your report shows real sales instead of just enquiries.

How to check the data is correct

Compare the number of orders in Metrica with your store admin for the same week. A small gap is normal because of ad blockers and tabs closed before the data is sent. A large gap means a bug. Typical causes:

• the purchase fires on every refresh of the thank-you page;
• the event does not fire when a buyer returns to the site after paying via Payme or Click;
• categories are named differently across templates;
• data is sent before the counter has loaded;
• cancelled and unpaid orders are counted as revenue.

Routine and attribution

Once a month, compare orders and revenue from search by category with the previous period. Once a quarter, list products with many search views and a low purchase share: that is your queue for better descriptions, photos, prices and stock info. Keep a change log for the site too, covering new templates, price changes and delivery launches, so a sudden jump or drop in the report is easy to explain.

Do not judge search by the last source alone. A furniture buyer may land on a comparison article, come back a week later by searching your store name, and only then order. Compare attribution models in Metrica: the difference shows how much search helps people get to know your store.

If you want sales from search, ads and messengers in one clear report, the markdev team can help set up end-to-end analytics. Tell us about your store and how you take orders, and we will discuss where to start.