Website operations: responsibilities and practical exercise
The webmaster oversees publishing and day-to-day site operation: content, components, access and incidents within the agreed scope. This role coordinates changes with relevant specialists and checks journeys afterwards. Technical, editorial and hosting responsibilities need owners even when different providers handle them.
These sheets describe activities and teaching exercises. They do not constitute job offers, certifying courses or recruitment promises.
Keep a website operating and route necessary interventions across publishing, updates, monitoring and incidents within a defined operational scope.
Typical activities
Check essential pages, forms, certificates and alerts.
Prepare updates and verify backup and restoration.
Maintain change records and handover arrangements.
Responsibilities to agree with the team
The role varies with site size. Specify authorised operations, specialist interventions and escalation; responsibilities should be clear before an outage.
A first piece of work to demonstrate
Create a maintenance procedure with checks, owners, tested restoration and documented rollback.
This is an editorial description of common activities. Job titles and scope vary across teams; it is not a certification framework.
Prepare a site update and check recovery if a check fails.
Work on a copy or demo environment. Define the data used and the expected result before you begin. Keep your decisions and the difficulties encountered in order to be able to explain the process.
Inventory components, owners and dependencies.
Prepare backup, tests and backtrack before the change.
Control important routes after publication.
Record result, incident and next action in the log.
Deliverable and checks
Provide the site inventory, change preparation and checks before and after updating. Describe backup scope, tested recovery and relevant contacts. Keep sensitive credentials in their access-management system rather than copying them into a shared log.
Simulate a failed journey check after updating and follow the decision to fix or roll back. Then check content, forms and access within the recovered scope. Another person should be able to continue operations using the retained instructions and results.
An available backup does not show that a restore is working. The procedure should specify what is restored and how the result will be verified.
Collaboration to explore: Back-end development.
Fictional deliverable excerpt : An intervention log records the content owner’s decision, affected links, index to refresh and public check performed. Technical credentials stay out of the published log.
What passes between the roles : The reproduction record accompanies the intervention and corrected version. Separate the removal decision, operational action and post-change check, with an identifiable owner for each step.
Fictional teaching material: this excerpt illustrates part of a deliverable. It is not a client reference, a complete specification or evidence of professional qualification.
Sample excerpt
Fictional check: restore to an isolated copy.
Scope: pages, media and required data.
Checks: opening, test sign-in and simulated contact journey.
External effects: disabled on the copy.
Record: restored version, outcome and unverified items.
Boundary case: text returns, but an image and required configuration are missing.
Recorded outcome: partial restoration, listing missing items and affected journeys.
What the author should be able to explain
An archive existing does not prove the website can restart.
The copy should avoid real messages and other effects on production services.
Unverified items remain visible at handover.
Does opening the homepage establish successful restoration?
It confirms only that first check. Inspect media, data and the journeys agreed in scope, then record limitations. Keep the test separate from the production website. Compare the backup inventory with the restored copy and replay a journey using the missing items. Record what needs recovery and who owns it. Keep secrets out of the report; restoring files alone does not demonstrate that all required configuration is available.
Assess the quality of the work
The exercise uses an update that appears successful in administration but damages a public journey. The webmaster should be able to locate components, owners and recovery methods. Examine maintenance as a verifiable chain from preparation to post-change checks with distinct responsibilities for hosting, external services and content.
Scroll the table horizontally to read every column. With a keyboard, focus the table area and use the arrow keys.
Assess the quality of the work
Skill
Evidence to examine
Failure to spot
Usable inventory
Identify active components, versions, owners, renewals and required external services. Distinguish a used resource from something merely present on the server.
An essential domain or dependency has no known owner, or a file is deleted solely because its name appears old.
Verified recovery
Describe elements to restore together, necessary access and journeys to check. Use a protected copy and data appropriate to the exercise.
A backup is declared sufficient without restoration, or only tables are recovered while media and configuration remain missing.
Monitoring
Useful alerts lead to action and verification.
A large log is never reviewed or exposes unnecessary data.
Post-publication check
Repeat an important task after an update and check content, links and message receipt.
A successful administration change is also checked on the public page with an unprivileged account.
A focused next step
After a fictional update, replay a page, media item and end-to-end enquiry using an unprivileged account. If the result differs from expectations, verify recovery on the test copy. Distinguish team-controlled work from actions needing the host or a provider without placing secrets in public instructions.
A mirror clone backs up the Git repository and its history. Git LFS objects require an additional fetch, and a wiki has its own repository. For a handover, also inventory data and services that do not reside in Git.
Dates indicate when references were consulted. These facts are neither a site audit nor an overall compliance statement.
A review case to reproduce
Recovery misses attachments
Situation to examine
A backup restores pages and headings but linked documents and some media remain unavailable.
Proposed check
Choose a complete page with document and image. Restore it in isolation and follow its links. Inventory files, data and dependencies; check who can resume service with the required access.
Evidence to retain
A report lists recovered and missing elements and the tested procedure. A backup’s existence does not replace verified restoration.
Fictional finding example
The starting inventory has 12 images and 3 documents. The restored copy finds all 12 images, but only one document opens.
Interpret this outcome
Two documents remain to be recovered or declared missing. An accessible homepage and correct image totals do not validate attachments.
Three complementary checks
Fictional cases: adapt these expectations to your test contract and retain the limits of the check.
1Expected inventory
2Tested copy
3Handed-over gaps
Scroll the table horizontally if needed; with a keyboard, focus it and use the arrow keys.
Scenario variations and expected evidence
Variation to test
Expected result in this case
Evidence to hand over
12 images return but 1 of 3 documents opens
Flag the 2 documents still missing
Expected list and links actually opened
Required configuration is missing
Describe the restoration as partial
Missing dependency and assigned recovery owner
The copy can trigger external sends
Isolate effects before replaying the journey
Checked test settings without secrets in the report
Decision to hand over : Hand over missing elements and verified limits before announcing recovery of the planned scope.
In-depth exercise: Support a demo site with broken links, an old document and a too heavy photo. Make an inventory before correcting. Check backups in a test environment and document a restore without touching production. Replace the document keeping a policy for its old URL. Finally, control essential courses and languages. Successful maintenance leaves understandable evidence and a way to reverse a change.
What to deliver with your work
Pages, media and dependencies inventory
List of defects and corrections made
Backup and restore procedure tested
Assessment of the courses, limits and work remaining to be expected
Acceptance criteria
These criteria are a working basis to adapt to the context. For each, note the action taken, the expected result, the observed result and the information remaining to be confirmed.
Restoration has been tried in a test environment.
An old useful URL has explicit processing.
The corrected images have a controlled weight and display.
Changes are documented without publishing access settings.
Keep a log of interventions with date, scope and verification. Specify the items that depend on an external service provider and the contacts authorized to process them. A backup on a disk does not prove that it is recoverable: Indicate the result of the test performed. After each intervention, check the priority routes and language versions affected by the change.
Website operations maintain workflows and information. Compare editorial updates with technical changes to choose appropriate checks and rollback arrangements.
Tracked content change
When to choose it
Suitable for opening hours, links, documents or descriptions. Identify the source, pages and language versions before publication.
What to plan for
Check old links, previews and structured data. A correction can leave other versions in conflict.
Prepared technical update
When to choose it
Suitable for dependencies, configurations or components. Verify backup and restoration arrangements before intervention.
What to plan for
Test critical workflows and monitor after changes. Rollback must account for data produced since the update.
A risk to anticipate
A responding website may still contain an unusable form or outdated attachment. Monitoring should include representative user actions.
How to decide
Keep an understandable intervention log. Another person should be able to find the change, result and follow-up required.
Compare content, access, operation and recovered procedures. One publication differs from taking responsibility for a site with backups and incidents.
Recurring operation
Define regular checks, updates and incident response. Separate on-call service, correction and evolution; a maintenance label does not specify human availability.
Exit and handover
Request inventory, change log and recovery procedures. Check authorised access transfer and restoration capability without copying secrets into documentation.
What should be checked after an apparently minor change?
Find the component and dependent pages. Repeat the affected journey with a success and an error, then check relevant languages and media. An address change may affect the form, footer and structured data; checking one page does not cover those copies.
How can an absence be covered without sharing every credential?
Name a substitute and check permissions for necessary operations. Document support contacts, alerts and recovery in the intended internal tool. Use the authorised access mechanism; public instructions or reports must contain no secrets.
What should be checked after a seemingly minor update?
Check affected functions and critical workflows that might be impacted: reading, navigation, search or contact requests. Look for errors and unexpected changes. Adapt the scope to dependencies rather than assuming a version number indicates risk.