Table of contents
Research context
This protocol measures how often an outsourced demo environment contains live customer data. The unit is one demo environment at one observation time, identified by the environment name and the owner. Live customer data means a record that matches a real customer, account, or person. The study describes data hygiene in demo systems, not the quality of the demo.
Define the environment population before sampling. Include sales demos, technical proofs, training systems, and partner sandboxes that an offshore team maintains. Exclude a local mock that contains only synthetic records. Record the owner and the product version for each environment.
Define the observation method. Use a read-only inventory of tables, files, and search indexes, then sample records against an approved matching method. Do not export customer data to make the comparison. Record the sample size and the matching rule. The method should be repeatable by another reviewer.
Classify a record as live only when the match is strong. A name alone is weak, while an account identifier and an email together are strong. Record the evidence category and the confidence. Weak matches should be reported separately rather than counted as confirmed.
Record the environment source, including whether it was built from a production export, a seed script, or a manual import. A production export is a risk signal even when the current sample looks clean. Record the date and the person who ran the import. The source explains how the data arrived.
Practical application
Check the reset and deletion history for each environment. An environment that is reset weekly has less exposure than one that has run for a year. Record the last reset date and the evidence. Report the environments with no reset record.
Report the number of environments, the number with confirmed live data, the number with weak matches, and the number that could not be checked. Show the counts by environment type and owner. Present the environments built from a production export. Use the same denominator for every rate.
Separate a live record that is still reachable from one that exists only in a backup or a log. The first is an active exposure, while the second is a retention question. Report the two categories separately. Record the location and the access path for each finding.
Inspect the environments around a sales event or a product launch. A cluster of live data may mean the team copied production to meet a deadline. Compare the pattern with the import log. A single bad demo can be a mistake, while a cluster is a process.
Do not treat a clean sample as proof that the environment is safe. A small sample can miss a table, and a search index can hold a copy that the database no longer has. Report the surfaces covered and the ones that were not. A narrow check produces a narrow result.
Operating controls
The record cannot show whether a customer saw the data or whether the data left the environment. It can show only whether a live record was present at the observation time. A live record is not proof of a breach, and a clean sample is not proof of isolation. The protocol also cannot read an encrypted backup.
A useful additional view is the share of environments that can be rebuilt from a seed without a production export. That share describes the design, not just the current state. Record the rebuild method and the last successful run. That field turns a point check into a review of repeatability.
Give the result to the client privacy owner and the offshore delivery lead with one observable change. The change might be a seed-based rebuild, a pre-demo data check, a shorter environment lifetime, or an import approval step. Record the owner, the effective date, and the measure that will show whether the change worked. Repeat the protocol with the same definitions after the change.
Close the review with one change and a date to re-measure. The change should address the environment type with the most confirmed live data. Record the expected effect and the evidence that will prove it. A demo environment that is never checked is a shadow copy of production.
Sources
Frequently asked questions
What should happen first?
Begin with a bounded ticket, approved access, explicit acceptance evidence, and a named reviewer.
Who approves production changes?
The client's technical owner keeps final merge and release authority.
