Research question and scope

This protocol measures how often an outbound or inbound delegated call starts without the approved recording notice. The unit is one connected call with a human participant. A compliant call delivers the approved notice before identity verification or sensitive detail. The measure describes notice delivery, not whether the customer consented in a legally sufficient way.

Set the scope to one client, one phone queue, and a fixed period. Record the recording rule, the script version, and the regions covered. If the script changed during the period, split the sample at the effective date. A different script creates a different denominator.

Keep the measure separate from the customer's reaction. A customer may accept the notice without comment, or object and choose an alternative. The purpose is to test whether the desk delivered the notice. Report refusals and outcomes separately.

State the decision the result will support before collection begins. The client may use it to change the opening script, add a prompt in the call tool, or retrain a shift. Write the intended action beside the scope. A measure with no decision becomes a compliance story.

Population and unit of analysis

Define the population as every connected call handled by the queue during the period. Include inbound calls, outbound calls, and callbacks. The unit is the call, not the customer or the case. A customer with three calls is three units when each call has its own opening.

Exclude calls that never connected, wrong numbers, and immediate hang-ups before the specialist spoke. Record the excluded count and the reason. Mixing abandoned calls into a notice measure makes the result look worse for reasons outside the script. Report the exclusion.

Include calls handled by any team member and any recording tool. Record the queue, the shift, and whether the call was recorded. Leaving out a tool hides part of the customer experience. State the inclusion rule before drawing the sample.

Decide how to treat a call that was transferred before the notice. Score the first connected segment, then note the transfer. Record the transfer time and the receiving queue. A notice given only after a transfer may be too late for the first conversation.

Definitions that make the measure reproducible

Define the approved notice as the exact script sentence or an approved equivalent listed in the guide. Define a compliant call as one with the notice inside the first ten seconds and before verification. Define a late notice as one given after ten seconds but before sensitive detail. Define a missing notice as none heard before the call ends.

Define the ten-second window from the first human word, not from the ring or the automated greeting. Record the timestamp. A notice buried after a long menu is late even if it appears early in the agent's script. The reviewer should use the same start point.

Define an unclear notice as one with the approved words but poor audibility, speed, or overlap. A notice the customer cannot hear is not delivered in practice. Record the reason. The distinction helps the coaching plan.

Define a refused notice by the customer's explicit objection. Record the response and the alternative offered. A refusal is not a compliance failure by itself. The failure is continuing to record after a clear refusal when the policy forbids it.

Sampling and data collection

Draw the sample from the call log across queues and shifts. Record the export date, the filter, and the row count. Do not build the sample only from calls the client reviewed or complained about. A selected sample cannot show the everyday opening.

Capture the call identifier, the queue, the shift, the agent role, the recording status, the notice timestamp, and the disposition. Add the script version. Store the call or transcript according to the client's retention rule. The opening segment and the script are the core evidence.

Check a small overlap sample with a second reviewer. If the reviewers disagree about whether the notice was audible, the definition needs an audio test. Rewrite the rule and retest before the full sample. Agreement matters more than sample size.

Record the time spent listening to each call and the number of tools opened. A measure that requires a full call review will not be repeated. Note whether the call tool exposes the first thirty seconds. Collection cost belongs beside the result.

Measuring notice delivery

Report the share of calls that are compliant, late, missing, and unclear. Show the counts behind each share. The groups describe different failures and should not be collapsed. A single compliance rate hides the notice that was present but inaudible.

Break the result down by queue, shift, and recording rule. A shift with a different opening script may show a different rate. Compare only groups with the same script version. A rate without the group size is hard to interpret.

Report the most common failure and the most common time gap. A notice at twelve seconds may point to a script order, while a missing notice may point to a tool prompt. Record one example of each. The pattern should lead to one fix.

Report refusals and the alternatives offered separately. A refusal rate with no alternative record does not show whether the customer had a real choice. Record the option and the outcome. The customer experience matters alongside the notice.

Limitations and next action

The measure cannot show whether the notice satisfied every legal requirement in every region. It also cannot show whether the customer understood the notice. Treat the result as adherence to the approved script. Note the regions and call types that were not included.

Give the result to the client owner and the phone team lead with one change to test. The change might be a shorter sentence, a prompt in the call tool, or a pause after the notice. Choose one change and name the person who will apply it. Record the effective date and repeat the measure with the same definitions.

Review the boundary cases before tightening the rule. A customer who interrupts with an emergency may need a different opening. Fix the exception list first if that is the pattern. The goal is a notice the customer can hear and understand.

Schedule the next measurement for the same week in the following month. Keep the queue and shift filters unchanged so the comparison is fair. Record any script change in the note. A trend built from different wording is not a trend.