Platform dependency is an operational question: which part of your work stops when a service becomes unavailable? Counting the number of profiles you maintain does not answer it. Several channels may still rely on the same login, payment route or file store. Map the complete journey from discovery to customer support before deciding where to invest effort.
Choose one real journey to inspect
Start with a specific activity, such as someone finding a tutorial and buying a workshop place. Write every step in order: discovery, offer page, inquiry, payment, confirmation, delivery and support. Include manual steps, such as copying registrations into a participant list, because those can also depend on a single person or device.
Trace an actual recent journey using records you are authorized to inspect. You may discover that an old profile link still directs visitors to an unused form, or that confirmations are sent from an inbox nobody checks. Record observed behavior separately from what you intended the process to do.
Give each dependency a useful record
Create a private inventory with one entry per service or essential resource. Record its purpose, the person responsible, the account used to administer it and where its operating instructions live. Do not put passwords, recovery codes or customer data into an ordinary shared planning sheet.
- What task requires this resource?
- What other resource must work first?
- What information would you lose access to?
- Which current commitment would be affected?
- What is the available fallback, if any?
A folder of original video files and a published video channel serve different purposes. Keeping the originals protects your ability to work with the material, but it does not reproduce the audience or comments. Describe the fallback accurately so a backup does not create false confidence.
Look for shared points of failure
Draw connections between the inventory entries. If the same email account controls your domain, shop and scheduling service, losing access to it can affect several otherwise separate tools. If all customer instructions exist only in social messages, an unavailable messaging service can interrupt delivery even when payment still works.
Include people and devices in the map. A collaborator may be the only person who knows how to send an access invitation. A laptop may hold the only editable copy of a paid resource. These are actionable dependencies even though they are not public platforms.
Rank consequences before buying replacements
Assess impact using your actual work: an interrupted paid session, a missing project file or an unanswered support request. Then consider how quickly you need a response and whether a tested alternative exists. A simple high, medium or low priority is often enough to identify the first repair.
Choose a small change that addresses the dependency directly. Save an authorized export, document the delivery procedure, assign a second responsible administrator where supported, or create a stable contact destination. Adding another promotional channel will not repair missing delivery instructions.
Assign ownership and review triggers
Every chosen fix needs an owner and a checkable completion condition. Replace vague tasks such as improve resilience with a concrete task such as verify that the workshop instructions can be opened from the approved backup location. Record what was tested and the remaining limitation.
Review the inventory when you launch an offer, replace a service or change collaborators. Keep the review proportional to the work involved. Use the continuity drill to test your highest-priority fallback, and the unavailable-account response guide when an interruption is already happening.
