---
title: "Support analytics"
meta:
description: "418 answered questions from 297 threads, 13 Apr to 25 Oct 2026"
---
- Questions
- 418
- Threads
- 297
- Customers
- 8
- Answerable today
- 78%
- With action drafts
- 87%
- Median first reply
- 1.3 h
- With a screenshot
- 37%
## What stands out
[#Section titled "What stands out"](#what-stands-out)
1. **78% of questions** can be answered today from the thread, the docs and code, and product data. Action drafts lift that to 87%. See [Sources](#sources).
2. **Data missing** is the largest type, 103 questions (25%). See [Question types](#question-types).
3. **Contoso** asks 93 questions, and 9 of them need someone from the team. See [Customers](#customers).
## Sources
[#Section titled "Sources"](#sources)
Each step adds one source an answer can draw on, from what the thread already holds to what only the team knows. What stays below 100% at the last step needs a person.
*Share of questions answerable, adding one source at a time*
All questions, and the two customers who ask the most.
thread only+ docs and code+ product data+ action draft
0%20%40%60%80%100%
ContosoallNorthwind
Data
| step | all | Contoso | Northwind |
| --- | --- | --- | --- |
| thread only | 20 | 20 | 16 |
| + docs and code | 55 | 49 | 55 |
| + product data | 78 | 78 | 76 |
| + action draft | 87 | 90 | 87 |
## Customers
[#Section titled "Customers"](#customers)
*Questions per customer, by the last source a reply needs*
The first three steps are answerable today; team only needs a person.
ContosoNorthwindTailspinFabrikamWingtipLitwareAdventure WorksProseware
1927271191230168101517101613710815575137
thread only
docs and code
product data
action draft
team only
Data
| customer | thread only | docs and code | product data | action draft | team only |
| --- | --- | --- | --- | --- | --- |
| Contoso | 19 | 27 | 27 | 11 | 9 |
| Northwind | 12 | 30 | 16 | 8 | 10 |
| Tailspin | 15 | 17 | 10 | 1 | 9 |
| Fabrikam | 8 | 16 | 13 | 4 | 7 |
| Wingtip | 10 | 15 | 9 | 1 | 8 |
| Litware | 8 | 15 | 5 | 6 | 3 |
| Adventure Works | 7 | 13 | 9 | 3 | 5 |
| Proseware | 4 | 13 | 7 | 4 | 4 |
*Questions per customer and type*
Darker cells hold more questions.
2020181310720198108111109743131078431013541310613681044545965334
Data missingHow it worksForecastPurchasingBugAlerts
ContosoNorthwindTailspinFabrikamWingtipLitwareAdventure WorksProseware
Data
| customer | Data missing | How it works | Forecast | Purchasing | Bug | Alerts |
| --- | --- | --- | --- | --- | --- | --- |
| Contoso | 20 | 20 | 18 | 13 | 10 | 7 |
| Northwind | 20 | 19 | 8 | 10 | 8 | 1 |
| Tailspin | 11 | 10 | 9 | 7 | 4 | 3 |
| Fabrikam | 13 | 10 | 7 | 8 | 4 | 3 |
| Wingtip | 10 | 13 | 5 | 4 | 1 | 3 |
| Litware | 10 | 6 | 1 | 3 | 6 | 8 |
| Adventure Works | 10 | 4 | 4 | 5 | 4 | 5 |
| Proseware | 9 | 6 | 5 | 3 | 3 | 4 |
## Question types
[#Section titled "Question types"](#question-types)
*Questions per type*
Sorted by volume.
Data missingHow it worksForecastPurchasingBugAlertsFeature requestSetup
10388575340342716
Data
| type | questions |
| --- | --- |
| Data missing | 103 |
| How it works | 88 |
| Forecast | 57 |
| Purchasing | 53 |
| Bug | 40 |
| Alerts | 34 |
| Feature request | 27 |
| Setup | 16 |
*Share answerable today, per type*
Setup, feature requests, and bugs need the team most often.
80%86%77%75%65%88%63%62%
answerable today
Data missingHow it worksForecastPurchasingBugAlertsFeature requestSetup
Data
| type | answerable today |
| --- | --- |
| Data missing | 80 |
| How it works | 86 |
| Forecast | 77 |
| Purchasing | 75 |
| Bug | 65 |
| Alerts | 88 |
| Feature request | 63 |
| Setup | 62 |
## Over time
[#Section titled "Over time"](#over-time)
*Questions per month, by type*
The four largest types, and the rest together.
May 2026JunJulAugSepOct
91311201313171110131915171029181572518201018
Data missing
How it works
Forecast
Purchasing
other
Data
| month | Data missing | How it works | Forecast | Purchasing | other |
| --- | --- | --- | --- | --- | --- |
| 2026-05 | 9 | 13 | 4 | 6 | 11 |
| 2026-06 | 20 | 13 | 6 | 6 | 13 |
| 2026-07 | 17 | 6 | 11 | 10 | 13 |
| 2026-08 | 19 | 15 | 17 | 10 | 29 |
| 2026-09 | 18 | 15 | 7 | 9 | 25 |
| 2026-10 | 18 | 20 | 10 | 10 | 18 |
*Time to first reply*
Median and 90th percentile per month, on a log scale.
May 2026JunJulAugSepOct
1 h2 h5 h10 h
p90median
Data
| month | median | p90 |
| --- | --- | --- |
| 2026-05 | 2.04 | 15.7 |
| 2026-06 | 1.23 | 8.8 |
| 2026-07 | 1.04 | 6.0 |
| 2026-08 | 1.33 | 8.4 |
| 2026-09 | 1.01 | 4.8 |
| 2026-10 | 1.02 | 3.1 |
## Team
[#Section titled "Team"](#team)
*Replies per person and type*
Ana answers most of every type.
52422526241726191315910162012946977331
Data missingHow it worksForecastPurchasingBugAlerts
AnaBenChenDara
Data
| person | Data missing | How it works | Forecast | Purchasing | Bug | Alerts |
| --- | --- | --- | --- | --- | --- | --- |
| Ana | 52 | 42 | 25 | 26 | 24 | 17 |
| Ben | 26 | 19 | 13 | 15 | 9 | 10 |
| Chen | 16 | 20 | 12 | 9 | 4 | 6 |
| Dara | 9 | 7 | 7 | 3 | 3 | 1 |
| Person | Replies | Median first reply |
| --- | --- | --- |
| Ana | 215 | 1.4 h |
| Ben | 101 | 1.1 h |
| Chen | 71 | 1.1 h |
| Dara | 31 | 1.2 h |
## Method
[#Section titled "Method"](#method)
How the numbers were computed
The export holds one row per customer message. A question is a message that asks for something; thanks and scheduling are left out. Each question was classified by its type and by the last source a sendable reply needed, in the order the line chart adds them. First reply is the time from the customer's first message to the team's first reply, in calendar hours. Every number on this page comes from one script over the export, analyze.py, and nothing was typed by hand.