The Scenario
You are a voice AI operations manager. It is Monday morning and your weekly ops review is at 10 AM. You need to figure out which Synthflow agents have the highest no-answer rate before you walk in.
The data is all in Synthflow. Thirty days of calls. Four agents. Hundreds of records. Your call log workbook on OneDrive has been empty since last quarter.
The bad version of this morning:
- You log into Synthflow, filter to the past 30 days, and look for an export button
- The CSV download includes twenty columns you do not need
- You open it in Excel, delete columns, reformat date strings
- You build a COUNTIF to tally no-answer outcomes per agent model ID
- The formula breaks because three rows have a different spelling of "no-answer"
- You walk into the 10 AM review with a half-finished pivot.
The fast version is one prompt.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent inside your Excel workbook that reads your data and calls Synthflow directly, so you never have to touch a CSV import or write a cleanup formula by hand.
Open the SheetXAI sidebar and type:
Pull Synthflow call analytics for the last 120 days into the 'Call Analytics' tab. Write date, duration in seconds, outcome status, and agent model ID into columns A through D. Add a pivot-ready summary row at the bottom showing total calls, average duration, and no-answer count per agent model ID.
SheetXAI calls the Synthflow API, pages through all call records, writes the clean table to the Call Analytics tab, and adds the summary. By the time you open the review deck, the no-answer breakdown by agent is already there.
What You Get
A clean call log table in the Call Analytics tab, ready for a pivot:
- Column A — call date
- Column B — duration in seconds
- Column C — outcome status (answered, no-answer, voicemail, transferred)
- Column D — agent model ID
- Summary row at the bottom — total call count, average duration, and no-answer tally per agent
The no-answer rate by agent is calculated before you build a single pivot. If one agent shows a 40% no-answer rate and the others are at 12%, the summary row makes that visible before you even filter the data.
What If the Data Is Not Quite Ready
Call log workbooks have inconsistencies. SheetXAI handles them inline.
When outcome status values are inconsistent
Your data has "no-answer," "No Answer," and "missed" all meaning the same thing.
Normalize the outcome status values in column C of the Call Analytics tab so no-answer, No Answer, and missed all become "no-answer." Then recalculate the no-answer tally in the summary row.
When you only want calls above a certain duration
Short calls under 10 seconds are usually misdials. You want them out of the analysis.
Filter out any call rows in the Call Analytics tab where duration is under 10 seconds. Recalculate the summary totals using only the remaining rows.
When you need a per-week breakdown instead of a flat total
The ops team wants to see whether no-answer rates are trending up or down over the month.
Group the call data in the Call Analytics tab by week and write a weekly summary below the main table showing total calls, average duration, and no-answer count per week per agent model ID.
When you need to join call outcomes with agent names from the Agents tab
The call log only has model IDs, but the Agents tab has the display names.
Look up each model ID in column D of the Call Analytics tab against the Agents tab, matching on the ModelID column, and write the corresponding agent name into column E. Then regenerate the summary row using agent names instead of model IDs.
The pattern: instead of cleaning the import and then building the pivot, you ask for both in one prompt. The cleanup happens inline, and the summary is already there when you open the workbook.
Try It
Get the 7-day free trial of SheetXAI and open any blank workbook, then ask it to pull your Synthflow call analytics. The Synthflow integration is included in every SheetXAI plan. For related workflows, see how to pull call transcripts for QA review in Excel or the Synthflow AI in Excel overview.