Outsourced ProgrammersPowered by StealthAgents

What share of offshore support calls expose unmasked customer data?

A protocol for sampling support calls, classifying screen exposure, and separating a brief masked field from a full customer record left visible.

What share of offshore support calls expose unmasked customer data?
WORK_LANE / RESEARCH BRIEF / DATAREVIEW QUEUEACCEPTANCEPASSowner review requiredEVIDENCE TRAILTICKET → BRANCH → REVIEW → CLIENT RELEASE
Table of contents
  1. Research context
  2. Practical application
  3. Operating controls
  4. Sources
  5. Frequently asked questions

Research context

This protocol measures the share of offshore support calls where an unmasked customer record appears on a shared screen. The unit is one support call with a screen share, identified by the session record and the ticket. Exposure means a value visible long enough to be read or captured. The study describes screen hygiene, not the intent of the agent.

Define the call population from the support platform, including calls with a customer and internal calls that use a shared screen. Record the session identifier, the agent, the ticket, and the share duration. Exclude a call with no screen share. Test the population against the platform export before sampling.

Define exposure by category, such as a full name with an account number, a card number, a credential, or an internal hostname. A masked field with the last four characters is not exposure for this measure. Record the category and the duration. The definition should be tested on two recorded calls.

Use a sample that covers agents, shifts, and call types. Set the sample size and the selection method before watching any call. Record the sampling frame and the calls that could not be reviewed. A sample drawn only from convenient calls is not a sample.

Record the evidence for each call, including the recording reference, the time code, and the visible field. Do not copy the customer value into the study file. Use a category and a short description instead. The reviewer should be able to confirm the finding without seeing the data.

Practical application

Distinguish a brief exposure from a sustained one. A field that appears for a second during a window switch is different from a record left open for the call. Record the duration and whether the customer could see the screen. Report the two groups separately.

Report the number of calls reviewed, the number with exposure, and the share by category. Show the counts by agent group and by call type. Present the calls with no recording and the calls that could not be classified. Use the same denominator for every rate.

Separate a missing tool control from a missed habit. A platform without a window share or a blur feature is a tool gap, while an agent who shares the desktop is a habit gap. Report the two categories separately. Record the evidence beside each finding.

Inspect the exposures around new-hire periods and tool changes. A cluster may mean the training did not cover the current platform. Compare the pattern with the onboarding dates. A single call can be a mistake, while a cluster is a training gap.

Do not treat a low exposure count as proof that the practice is safe. Agents may avoid sharing screens when they know a review is running, or the sample may miss the calls that matter. Report the sampling method and the limitation. A narrow review produces a narrow result.

Operating controls

The record cannot show whether a customer captured the screen or whether the data was used. It can show only whether an unmasked value was visible during the reviewed call. A visible field is not proof of harm, and a clean call is not proof of control. The protocol also cannot review a call that was not recorded.

A useful additional view is the share of calls where the agent used a window share or a blur control. That share describes the available practice rather than the outcome. Record the tool and the action. That field turns a count into a review of the control.

Give the result to the client support owner and the offshore delivery lead with one observable change. The change might be a window-share default, a blur shortcut, a pre-call checklist, or a redaction drill for new agents. 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 most common exposure category, not the easiest call to review. Record the expected effect and the evidence that will prove it. A screen rule that is never observed is only a poster.

Sources

  1. NIST SP 800-53 security and privacy controls
  2. CISA Secure by Design
  3. OWASP ASVS
  4. Google SRE Workbook

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.

Build a programming lane your technical owner can review.

Bring the stack, backlog, acceptance checks, and release boundaries. We will help scope the role around the work.

Discuss your programming needs