Reviewing a call

The Phone calls page shows all calls. Each row is a call; click it to open the details window.

Call detail window

Status

At the very top of the panel, above the tabs, you'll see the call's current processing status. Hover it to see a full two-track timeline: one track shows events reported by your call tracking platform, the other shows Kaliper's processing stages:

Overview tab

QC Result

The combined pass/fail verdict across all Scorecards applied to this call. Three possible states:

• Pass — the call met the score threshold and all Critical Detectors passed

• Fail — the score fell below the threshold, or at least one Critical Detector failed

• N/A — no Scorecards are assigned to this call's campaign, or scoring hasn't run yet

When the result is Fail, a reason card appears.

Score The weighted quality QC score as a percentage, combining all Scorecards applied to this call.

Summary An AI-generated summary of the call transcript. If it's inaccurate or unhelpful, use the Report summary issuelink — this sends feedback to the Kaliper team.

Recording An inline audio player.

Flags, comment, and sharing

Flags Flags are manual — they're set by your team for internal purposes. Every flag change is logged with the user who made it.

Comment A free-text field for notes. It auto-saves as you type and also saves when you close the panel. Comments are visible to anyone in your org who opens the call.

Share Creates a public link to this call and opens it in a new tab immediately. The shared view is accessible without logging in — no Kaliper account needed.

Call info

Below the main actions, a collapsible accordion contains three sections.

Processed data Computed fields Kaliper derives from the call:

• Conversion Detected — whether a conversion was detected. Click the pencil icon to manually set it to Yes, No, or Unset (manual overrides use sentence case to distinguish them from the AI's uppercase YES/NO output). Changes are logged in the Change Log

• Timestamps for each processing milestone.

Source data The original call record as received from your call tracking platform — every field exactly as it came in. Useful when you need to verify a value at the source or troubleshoot a discrepancy.

Change log A record of every user-initiated change to this call. Tracked fields: flag, comment, Conversion Detected, and read/unread status. Each entry shows who made the change, when, and what the value was before and after. Automatic updates from scoring do not appear here.

Scoring tab

The Scoring tab breaks down the QC Result by Scorecard and by individual Detector.

Per-Scorecard summary

For each Scorecard applied to this call, you'll see:

• QC Result for that Scorecard (Pass / Fail / N/A / In Progress)

• Score percentage

• How many Detectors Passed, Failed, and were Skipped

Questions (Detectors)

Below each Scorecard summary, every Detector is listed with:

• The AI's output (YES/NO, a disposition category, or an extracted value)

• A colored dot — green if the output matched the target, red if it didn't

• A Critical label in red if this was a Critical Detector that failed

• A Conversion label in green if this Detector was marked Conversion Related and the AI's output matched the target — this is what set Conversion Detected = Yes on this call (see Scoring and Conversion detection)

Click any row to expand it and read the AI's full reasoning for that output.

Thumbs feedback

Each Detector result has thumbs up and thumbs down buttons. Use them to evaluate the AI's output.

What happens when you click

1. Your rating is saved — Kaliper records who evaluated the output and when

2. The call's QC Score is recalculated — quality metrics for the call (QC Score, Pass/Fail) update automatically to reflect your negative rating

3. Admins are notified — if your organization has notifications enabled, a super-admin will receive an email with the details: which call, which Detector, and what rating was given

Transcript tab

The full verbatim transcript of a call.