Overview
The Agent Performance page helps you evaluate your support team’s activity and efficiency based on replies and tagged reviews. It provides detailed metrics for each team member and shows how automation contributes to review processing. The key indicators to track include:
- Number of Replies
- Reply Effect
- Agent Reply Time
- Platform Reply Time
- CSAT
- Reply Rate
Agent Performance is always available for Admin and User roles. Customer Support and Reviews Manager roles can receive access upon request. Please reach out to your account manager or our Support team for more details.
Agent Performance analytics are available across all plans, including Free, Essential, and Team. The level of detail may vary by plan—for example, some metrics can be shown as ranges or categories (such as Fast or Slow for reply time), providing a clear high-level view. More detailed metrics and exact values are available on the Enterprise plan.
Manual Agents and Automation Agents
The Agent Performance page has two tabs:
- Manual agents: reply, tag, and concern report metrics for your agents, described in the sections below.
- Automation agents: auto-reply and concern report results from your automation, described in Automation Agents.
To create a new agent, click Add agent in the top-right corner and select Automation agent or Manual agent.
Available Agent Data
On the Agent Performance page, you can select to view the data by All Agents, as well as compare the data separately by each agent. We show statistics for:
- AppFollow Agents (Team Members and Automation). Team Members who reply from the AppFollow interface. Automation shows the data by Auto-replies (including the automated concern report) and Auto-tags.
- Email Agents: We show the name and email address of users who reply via any Email or Helpdesk Integration connected to the AppFollow account. Note: We only show reply statistics for Email Agents.
- API: Data for replies sent via the AppFollow API.
- Zendesk Agents: By default, we show the Zendesk App as a single user. You can set up an additional integration to see Zendesk Agents; this allows you to see the names and email addresses of individual agents who reply via the Zendesk App Integration. To see the data by replies and tags, it's necessary to add the Zendesk API key to AppFollow: Zendesk Agent Performance Integration.
When you reply to a review from Zendesk, make sure to save the ticket assignee before sending a reply. This ensures that we will receive information about the assigned agent and show it in AppFollow.
Filters
Filters allow you to drill into the data you're interested in learning more about. Set up filters to:
1. Select agents: The list includes all agents who have ever responded to reviews.
2. Select sources: Here, you can select one or multiple sources from the current workspace.
3. Select a time period* for the analysis and the device country (in some cases, only worldwide data is available).
Device Country data for Google Play sources requires a Google Play Invite User connection with Countries enabled. Country-level data may be collected later than review data, which can affect the speed at which reviews become available for processing and automation. If faster review collection is a priority, you can use a separate workspace without Countries enabled and another workspace for country-level analysis.
On the Agent Performance page, the data is filtered by reply date (or tag date, report a concern date). It is different from how the data is filtered on the Reviews feed and Reviews analysis pages.
4. Select additional filters: Review language and Workspace.
*The "Today" button is unavailable as we display data with a 1-day delay to avoid discrepancies and facilitate page performance. We display information for the day before yesterday if the current time you are looking at this page is earlier than 3 AM (GMT +3). If it is past 3 AM (GMT +3), you are able to see data from yesterday.
Summary
The first section provides the most valuable metrics to track when working with reviews on the platforms.
These metrics are:
-
Replies submitted - the number of replies published on the platforms.
- For automation, this metric reflects its performance as a support agent and therefore only accounts for replies that actually ended up on the platform (i.e., it does not include replies generated for deleted reviews or replies that were not published).
- Agent reply time - measures how quickly agents respond to reviews in AppFollow, excluding platform moderation.
- Platform reply time - measures the total time from when a review is submitted until the reply is published on the platform, including moderation.
- Reply effect - shows the number of reviews that were updated after receiving a reply.
- Tags applied - the number of tags applied.
- Reviews reported - the number of reviews reported as a concern.
Some metrics may appear as ranges or categories depending on your subscription plan.
Agents Distribution
The Agents Distribution section appears only when the All Agents view is selected. It displays statistics for all agents, including team members and automation. You can customize the metrics shown using the dropdown menu in the top-right corner and sort the table by clicking any column header.
Available metrics by each agent:
- Replies submitted - the number of replies published on the platforms.
- AI direct replies - the number of replies generated by AI and sent to the platform without any edits at all.
- AI-assisted replies - the number of replies where AI features were used, but the agent made any edit before sending. Or Agent created a reply manually/with a template and used any AI feature, like Improve, Shorter, Longer, etc.
- Submitted auto-replies - the number of auto-replies approved and submitted by agents.
- Agent reply time - measures how quickly agents respond to reviews in AppFollow, excluding platform moderation.
- Platform reply time - measures the total time from when a review is submitted until the reply is published on the platform, including moderation.
- CSAT - the percentage of reviews with the positive reply effect (the reviews with the updated rating to more stars) from the total number of reviews with reply effect except for the 5-star reviews.
- Tags applied - the number of tags applied.
- Reply share - the share of reviews with an agent reply out of the total number of reviews replied.
- Replies per day - the rate of reviews replied to the number of days. Shows how many reviews are replied to daily. For this metric, we only count days when there was at least one reply published.
-
Platform 1st reply time - measures the total time from when a review is submitted until the first reply is published on the platform, including moderation.
- Agent 1st reply time - measures how quickly an agent sends the first reply to a review in AppFollow, excluding platform moderation.
- Tag share - the share of reviews tagged by an agent out of the total number of tagged reviews.
- Reviews reported - the number of reviews reported as a concern.
- Agent reply effect - the update of the review rating as a result of an agent reply.
- Reply effect reviews - shows the number of reviews whose ratings changed after receiving a reply.
- Reply effect reviews % - shows the percentage of replied reviews that resulted in a rating change.
CSAT
CSAT - the percentage of reviews with the positive reply effect (the reviews with the updated rating to more stars) from the total number of reviews with reply effect, except for the 5-star reviews.
To calculate CSAT:
- We take the reviews with reply effect for which the rating has increased after the reply.
Note! Reply effect is calculated for reviews that were updated within 3 months after the reply. - We take the total number of reviews with reply effect.
- From this amount (total reviews with reply effect), we deduct the number of reviews with reply effect that have 5 stars before and after the reply.
- The formula is applied:
(1) Number of reviews with positive reply effect
_________________________________________________________________ X 100% = CSAT
(2) Number of reviews with reply effect - (3) 5-star reviews
AI-powered summary
The AI-powered summary highlights the most important takeaways from the Agents distribution section. For the selected date range and filters, it analyzes core metrics: reply effect, CSAT, reply volume, first reply time, and AI usage, and generates concise positive and negative points for each agent performance.
Use it to spot who drives impact, where responsiveness needs attention, and which workflows benefit from AI assistance.
Reply Effect
The Reply Effect section helps evaluate how agents’ and the team’s replies influence changes in source review ratings.
Agent Reply Effect shows how many reviews were updated and how their ratings changed as a result of replies sent by each agent.
All AppFollow Agents Effect reflects the overall impact of all replies — including those from team members and auto-replies — made during the selected period.
No Reply Effect indicates reviews whose ratings changed independently or were not affected by any replies.
Reply Time
The Reply Time metrics help evaluate how quickly your team responds to user reviews, providing insight into internal performance (how fast agents respond within AppFollow) and external factors (how long platform moderation takes before the reply is published). Use them together to analyze your team’s efficiency and identify potential delays in the review management process.
- The Agent Reply Time metric reflects how quickly agents respond to reviews through AppFollow, excluding the platform’s moderation time. This allows Team Leads and Managers to accurately evaluate agent efficiency based on actual response speed within AppFollow.
- The Platform Reply Time includes both the agent’s response time and the time it takes for the platform to process and publish the reply. This process can take up to a few days in the App Store, for example.
On Free, Essential, and Team plans, reply time may be shown as categories (e.g., Fast or Slow) instead of exact time values.
Share of Replies to Tagged Reviews
The Share of Replies to Tagged Reviews section shows the percentage of replied reviews for the five most frequently used tags (e.g., only 34% of reviews tagged “Genie” have been replied to).
Agent Workload
The Agent Workload section shows how active the team was during the selected period. It includes data on:
- Replies submitted
- Reviews tagged (displayed daily or weekly)
- Auto-replies moderated by each agent
Currently, the Agent Performance page displays data for replies submitted from the Reply to Reviews page, Email and Zendesk agents, and replies sent via API. Data from messenger or platform integrations is not included.
Report a Concern
The Report a Concern section provides both a Summary and Distribution of all reported concerns. Regardless of workspace settings, it includes data for every report submitted, even those from other workspaces.
- The Summary tab shows the total number of concerns reported, the percentage of deleted reports, and a breakdown of each report’s status: Pending, Submitted, Error, Declined, Satisfied, and Deleted.
- The Distribution tab lets you filter and view reports by specific statuses or display all statuses together, either on a daily or weekly basis.
Automation Agents
The Automation agents tab shows how effective your review management automation is. You can track your main automation KPIs, see which rules are used most, and check your overall Automation score.
The tab uses your account's default timezone, which you can change on the Settings page.
Filters
Filters apply to the automation rules that match the selected conditions. You can select:
- Sources
- A time period
Summary
- Auto-replies created: the number of auto-replies created for the selected filters, both published and pending approval. This includes replies generated for reviews that were later deleted, so some of them are never published.
- Concern reported: the total number of concerns reported.
- Concerns satisfied: the share and number of reported concerns that were satisfied.
- Automation score: the share of replies created by automation out of all replies.
Auto-replies Created
This chart has two tabs:
- Overview: the number of auto-replies by Published and Pending status, daily or weekly.
- By rules: the number of auto-replies created by your most used rules.
Automation Summary
This section shows a pie chart with the shares of your most used auto-reply rules, plus a table of metrics.
Overview shows metrics for all auto-reply rules:
- Rule: the rule name.
- Replies created: the number of replies submitted by the rule.
- Pending approval: the number of the rule's replies in pending status.
- Last run: the last date the rule was applied.
- Reply/Rule share: the rule's share of all automated replies.
- Replies per day: the average number of replies the rule submits per day.
Report a Concern shows metrics for all Report a Concern rules:
- Rule: the concern type (Offensive, Spam, No sense).
- Reported: the total number of reports submitted.
- Reviews deleted/Concerns satisfied: the number of satisfied concerns.
- Declined: concerns the platform declined, meaning the review was not deleted and is still visible.
- Approval rate: the share of satisfied concerns out of all concerns reported.
- Last run: the last date the rule was applied.