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.
Article5 min readViews0

Catalog of 10,000 Products: How to Assess Data Readiness Before Migration

The number of rows in an export poorly predicts the effort required to launch. We examine product samples, filling rules, and a manual calculation exercise to distinguish between migrating data and preparing the catalog.

Comments 0

Two fabric swatches with empty product tags and an empty space with a lime-colored fastener on a plum background
In this article

"We only have ten thousand products; we will send the export." This statement is insufficient for assessing migration. One group has photos and specifications, another has only the supplier name, and a third has sizes recorded in the description. Technical file upload may take less time than deciding exactly what the store is ready to show to shoppers.

Before finalizing the total estimate, it is useful to check a small but diverse sample and count recurring actions. The result is separate estimates for migrating ready data, bulk transformation, and manual preparation. This helps the client understand which part of the work a developer can accelerate and which requires a person who knows the products.

What Counts as a Ready Product Card

First, agree on the minimum set of information for each category. For clothing, size and color variants are critical; for technical products, compatibility and specifications matter; for consumables, the sales unit and quantity per package are key. A single list for all products is insufficient. Legally required information must be verified separately for each specific assortment; this analysis does not replace such verification.

Value rules are also needed. "Gray," "gray.," and a supplier code may describe the same color but remain distinct rows for filtering. First, the catalog owner approves a mapping dictionary, then the team applies it. Automatically merging entries based on similar names without verification can incorrectly combine different products.

The CSV import documentation for 1C-Bitrix separately covers the unique identifier and item name, field mapping, and configuration preservation. This forms the technical basis for loading. A successful import alone does not confirm that the product card contains the required photo, a clear sales unit, or a correct set of attributes. We do not modify import settings here, nor do we claim execution on your store.

The sample must demonstrate different types of work

Do not select only ten ideal cards. Include different categories, suppliers, and product types: simple products, models with variants, kits, items without photos, and records with incomplete characteristics. Also include known complex cases. This sample helps identify types of work but does not automatically become a statistically accurate assessment of the entire catalog.

For each sample, note the initial state, the required action, and who can perform it. Number formats can be converted according to the rule. Missing characteristics cannot be reliably obtained through conversion; they must be confirmed by the supplier or a specialist. If a person spends twenty minutes searching for information, that search time also counts toward preparation.

  • Ready for transfer: data is filled in and complies with agreed rules.

  • Fixed en masse: there is a clear transformation and a control example.

  • Requires manual work: information must be found, verified, or created.

  • Excluded from launch for now: mandatory data is missing or no clear owner is identified.

Calculate the manual preparation queue

In a training catalog of 10,000 items, after verification, it is assumed that 7,000 are ready, 2,000 are being corrected through bulk processing, and another 1,000 require manual work. These proportions are given as an example; they cannot be transferred to your project without observations. Let an average of six minutes be spent on a manual card, including result verification. This results in 6,000 minutes, or 100 hours.

If a specialist can dedicate four hours a day to the catalog, that equals 25 working days of their time. Two specialists with independent tasks and the same productivity theoretically yield 12.5 days of combined work. A calendar plan requires rounding, schedule adjustments, and time for approvals. Doubling the workforce does not guarantee doubled speed: a single expert or waiting for a supplier response can remain a bottleneck.

Suppose a recheck reveals that a complex card takes nine minutes. The volume becomes 150 hours—50 hours more. For one specialist with four available hours, this is already 37.5 working days. Such sensitivity is more useful than promising an exact date based on the number of lines in a file. For planning, fix a range and revise it after the first batches are actually processed.

Who decides on disputed items

A developer may notice that a package lists different quantities but is not required to guess the correct one. Assign a data owner from the store side and set a response deadline. The question card needs only the product identifier, conflicting values, sources, and the resolution. Correspondence disconnected from a specific record quickly turns into rechecking the same issue.

The adopted rule must be returned to the original process. If only the storefront is corrected, the next export may restore the old value. Where the final version is stored depends on the project: in the accounting system, the supplier catalog, or a separate site field. The source is determined for each data type, not by a blanket statement like "we will take everything from 1C".

Launching the entire catalog at once is not required

If a standalone product group is ready, its launch can be discussed ahead of the others. It must support selection, pricing, stock availability, order placement, and fulfillment. The decision requires validating the assortment and customer expectations: empty but important categories cannot be hidden behind a total count of loaded products.

Batch acceptance is conveniently summarized in a short record: how many items were accepted, how many were excluded with reasons, which transformation rules were applied, and which control cards were verified. The number of rows and the count of products ready for sale may differ. It is the latter number, together with the list of unresolved issues, that indicates whether the catalog is ready for the promised launch.

Discussion 0

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

No comments yet. Start the discussion.