Payments are experiencing issues due to temporary restrictions in Russia. If your payment does not go through, please submit a support request.Our support team is available 24/7 — we are always here to help with hosting and server issues.We are now accepting requests for dedicated server rental and colocation services in our data center.Reminder: we recommend enabling backups for additional data protection.A new VPS/VDS lineup with NVMe storage and improved performance is now available.Maintenance work on some servers has been completed. All services are operating normally.
Article3 min readViews6

Maximizing Profit in Direct: Why Stores Need Accurate Order Data

Yandex is advancing its strategy with dynamic margin calculation. I examine how data errors hinder evaluating advertising based on actual business results.

Comments 0

Various products from the same store on a packing table
In this article

Two orders of the same amount can yield completely different results for a store. One product has a higher procurement cost, another is returned more often, and a third requires expensive shipping. A revenue-based ad report does not reveal these differences on its own.

In its September update roundup, Yandex highlighted the development of the "Maximize Profit" strategy, which exited beta in August 2026. Among its features is calculating margin as a share of actual conversion revenue. For stores, this is an opportunity to verify the quality of transmitted data before entrusting budget optimization to the system.

What Dynamic Calculation Means

In revenue share mode, the advertiser sets a percentage, and the system applies it to the amount of each specific conversion. If the order amount changes, the calculated value changes as well. However, the selected percentage does not automatically become the exact margin for each item.

Consider a hypothetical example: two orders of 10,000 rubles each. With a set margin share of 20%, the calculated margin would be 2,000 rubles per order. If the actual economics of the products differ, a single percentage will not reflect that. This example is hypothetical; its purpose is to demonstrate the limitation of input data, not to propose a configuration suitable for any store.

Therefore, the team must first agree on what they define as margin, which expenses are included, and how well the chosen value aligns with the promoted assortment.

Three areas where data loses meaning

First, duplicate events. Reopening a confirmation page should not appear as another sale. Check order uniqueness and how resubmitted information is handled.

Second, mixing stages. Ordering, payment, and fulfillment occur at different times. If all created orders are counted as completed sales, advertising will be evaluated based on intentions, some of which were never realized.

Third, incomplete amounts. Discounts, shipping, cancelled items, and changes made after a manager's call can differ from the initial value. A clear reconciliation process with accounting data is needed.

How to start the check

Take a small sample of orders and trace each one from the website to advertising analytics. Compare the identifier, amount, time, and status. Include cancellations and retry attempts to verify not only the ideal path.

After that, test the strategy on a consistent group of products and evaluate it over a sufficiently long period. Do not change the entire catalog, budget, and result-tracking method simultaneously; otherwise, the causes of changes will remain unclear.

The advertising algorithm receives exactly the signals the store prepared. The more precise the link between an event on the site and an actual purchase, the more meaningful the discussion about promotion effectiveness becomes.

Discussion 0

Share your experience and ask questions. Comments without links appear after editorial review.

No comments yet. Start the discussion.