Project management of PIM implementation and catalog synchronization with Shopify and Ozon/WB in a company with 10k SKUs

The initial situation is an online electronics store with 10k SKUs and frequent changes in prices/attributes. Currently, product data is maintained in Google Sheets and partially in 1C, which is why there are discrepancies in names, attributes and balances between Shopify and marketplaces (Ozon/Wildberries). Team - business owner, content manager, 1C contract specialist, integration developer (part-time). It is required to build a managed PIM implementation project (based on Pimcore or Akeneo Community - choice at the analysis stage) and a migration plan to reduce errors and speed up the launch of new products.

Technical environment and limitations - Shopify as the main store, 1C:UT as a source of balances and purchase prices, Ozon/WB marketplaces through existing connectors (they cannot be changed without approval), distributed team, access to production is limited, development time is maximum 40 hours for the current developer for the entire project, budget for PIM licenses - only free editions. Personal data is not processed, but there is a trade secret - all project artifacts must be stored in a dedicated workspace (Notion or Confluence - we’ll choose together).

What needs to be done in stages
1) Diagnostics and staging - interviews of 3-4 stakeholders, map of current data flows, list of systems and data owners, recording of pain points and priorities. The result is an agreed Scope/Out of Scope document and a risk log.
2) Catalog data model - designing a target attribute structure for 3 key categories (for example, smartphones, laptops, accessories), mandatory rules, directories, options, multilingual/SEO (if necessary). The result is an attribute specification and validation rules.
3) Integration and migration plan - source/receiver scheme, synchronization strategy (master-data, update frequency, conflict resolution), wave migration plan, readiness criteria. The result is an integration diagram and a step-by-step cutover plan.
4) Project plan and execution management - WBS, schedule for 4-6 weeks, RACI, communication plan, task backlog for the developer and 1C specialist, weekly statuses and decision protocols. The result is a work plan in Jira/Linear/YouTrack (optional) and a repository of solutions.
5) Acceptance and transfer - checklists for testing uploads for Shopify and 1 marketplace in the pilot category, a list of defects and a correction plan, instructions for the content manager on introducing goods and maintaining data quality. The result is a package of artifacts for support and scaling to other categories.

Measurable acceptance criteria
- Prepared and agreed upon: Scope/Out of Scope, risk register, target data model for at least 3 categories, integration scheme, migration plan by waves, project plan with tasks and owners.
- The task tracker contains a backlog of at least 40 tasks with priorities, dependencies and estimates, and at least 4 status meetings with protocols have been held.
- For the pilot category, validation rules and a checklist are described that allows you to reduce the share of cards with critical attribute errors from the current level (measured at the start) by at least 50% on a sample of 200 SKUs.
- All materials are in a single space, links are structured, decisions are recorded in the decision log.

Professional tools - Jira/YouTrack/Linear, Miro/FigJam for diagrams, Notion/Confluence for documentation, Google Sheets/CSV data profiling, basic understanding of API/ETL, experience in PIM and e-commerce directories.

It’s important that what’s expected is not just a plan on paper, but a controlled delivery to pilot readiness with artifacts that the team can use to continue implementation without losing context.