Practical framework / 5 MIN READ
Plan the evidence horizon before old detail disappears
Match each future analysis to the minimum data it needs and an authorized retention plan.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-06
A dashboard’s visible history does not guarantee that its underlying detail remains available for new analysis. Reporting continuity requires an explicit horizon, lawful purpose, access control and a tested retrieval path.
What you will leave with: Create a retention requirements matrix that distinguishes aggregate history, event detail and permitted exports.
Use this when
- An annual review needs detail older than the exploration tool retains.
- A team proposes exporting everything indefinitely just in case.
This lesson is not legal advice or permission to extend retention, create access grants or copy personal data.
Start with future decisions
List recurring analyses and when they occur: annual planning, a long sales-cycle review or a campaign reconciliation. For each, identify the minimum fields, level of detail and historical horizon required. A market-level total may answer a question without retaining individual identifiers.
Separate a genuine commitment from speculative future curiosity. Write the business purpose and owner for each requirement. If nobody can name a supported decision, challenge the proposed retention rather than treating unlimited storage as the safest default. Consider whether a documented aggregate can satisfy the need.
Inspect what each product actually retains
Google Analytics distinguishes user/event-level retention from standard aggregated reporting and describes product-specific limits. A long trend chart therefore does not prove that all old event detail can be queried. Inspect the current product, report type and account configuration.
Map each required analysis to its actual available source. Record what happens when retention expires and whether historical export is supported. Do not promise a later warehouse connection will backfill information already unavailable. Check the separate rules for linked advertising data or other specialized fields.
Design an authorized minimum archive
If an export is necessary, agree its fields, destination, access and deletion rules with the responsible owners before enabling it. A warehouse copy needs its own controls; moving data does not erase privacy obligations or automatically inherit source deletion behavior.
Choose the least detailed artifact that preserves the intended evidence. Store calculation definitions, mapping versions and source cutoffs alongside permissible data so an aggregate remains interpretable. Avoid retaining secrets or unnecessary personal fields. Plan how corrections and authorized deletion requests propagate to downstream copies where required.
Test retrieval before relying on it
Run a small approved retrieval exercise: can a second analyst reproduce the required prior-period total and explain its scope? Check that permissions work for the intended role and that files are not trapped in a departing employee’s private account.
Schedule a review when the analysis purpose, provider limits or legal obligations change. If history is already missing, disclose the gap and adapt the decision rather than manufacturing a precise reconstruction. A documented prospective plan is more credible than claiming an archive is complete because a folder contains some old exports.
Make the reasoning concrete.
A hypothetical business holds an annual review of a nine-month sales cycle. Its chosen exploration view cannot provide the required older detail. The team only needs monthly campaign counts and stage outcomes, not names or contact information.
Export all records forever
Opening: Copy every available field into a shared drive.
Hypothesis: Some future questions might be answerable.
Risk: The copy creates unnecessary exposure and lacks a deletion policy.
Define a minimum evidence set
Opening: Approve a controlled monthly aggregate with definitions and a retention horizon.
Hypothesis: The stated review can be supported with less data.
Risk: Some later individual-level questions will remain intentionally unsupported.
Confirm the aggregate meets the decision, then implement only the approved archive and test its retrieval. State any historical gap that predates the plan.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Decision
Which future review requires history?
Minimum
What fields and granularity are truly needed?
Availability
Which product retains that evidence for how long?
Controls
Who approves access, export and deletion?
Test
Can the intended reviewer reproduce one past period?
Watch for the wrong conclusion.
- Assuming visible aggregate history means retained event detail.
- Expecting a future export to restore expired data.
- Keeping all identifiers indefinitely.
- Treating a warehouse as exempt from retention rules.
Your next action
Write a retention matrix for one annual analysis, including a minimum-data alternative and a retrieval test.
Sources & scope
Frameworks, worksheets and hypothetical examples are original editorial synthesis. Linked primary sources support only the stated concepts. Source checking is not an independent audit of effectiveness. No named human author or reviewer is claimed.
- Google Analytics: Data retention →Primary technical documentation · Supports: Explains retention for user/event data and different treatment of standard aggregated reporting, with product-specific limits. Does not authorize longer retention or imply a warehouse inherits the same deletion policy.