Metrics are things that can be displayed in a chart. Each metric has filters you can apply to drill deeper into what you want to see (ex. total messages via live chat only). Additionally, you can use the Formula metric to do math comparisons of base metrics (ex. AI Spend ÷ AI Reply = Cost Per Message).

If you need a refresher on what reports and dashboards are, or what each type of chart is, see this article here. See the full description of metrics and examples below.

Every message that went out to your leads… AI and human, combined. This is helpful if you're trying to identify the total number of messages per contact your team is sending through your source.

Below shows a simple chart of total replies per day over 1 week.

Every reply your AI sent entirely on its own because the source matched Job Flow filters.

You may want to know what times of day and which days of the week are getting the highest AI engagement... do that with the AI Response Count metric and the Heatmap chart. See below.

Every message your leads sent back. Similar to how we count all outbound messages as a metric, we also give you a count of total inbound messages. See when your leads are messaging you, regardless of whether or not you have an AI replying.

See the chart below that shows Inbound Message Count per day over the past 7 days compared to the previous 7 days.

How many total contacts there are running through your system even counting ones your bots don’t reply to.

See below where we show the total number of leads added per month over the past 12 weeks easily compared in a pie chart.

How many real people your AI personally engaged. Useful if you want a clear line into when and how your Agents are replying to leads.

See the chart below that shows a line graph view of the number of contacts AI Replied to per Source over 4 weeks.

Contacts who sent their very first message to you even ones AI is not replying to. Not a count of the unique contacts that sent you messages during that time, but a count of those who sent you their first ever message during that time.

The average quality of the leads coming in. Watch it climb as your targeting and conversations sharpen. You set up your own lead scoring metrics within the source settings of each source.

You may want to see the average lead score over time, hoping that this number is going up with your improvements in lead quality and AI Agent capabilities.

Every point your bot awarded as leads showed signals you set up within your source settings Lead Scoring options.

See below where we show the daily trend of points gained by our leads compared to the previous 7 days.

How many real people ended up on your calendar. One contact that books multiple appointments counts once.

See the chart below that shows daily contacts booked over the past several weeks (in which only 3 days had booked contacts).

Every appointment booked and re-booked by your AI, where one contact that books twice is counted twice.

See the chart below where we break down the bookings per day by channel.

Exactly what your AI conversations cost you. Available per source, so you always know your raw cost and where it's being spent.

See the chart below where we look at AI Spend per source.

What your sources were actually charged… matching your rebilling rules.

See the chart below that reflects the total account's billed costs over time as a table.

Your consumption in billable segments. One segment always equals one message your AI sends... unless you unlock your agent node. Learn more about that Here.

Below shows AI Response Segments over the past week in a Heat Graph. We can see that most of our segments happen between 6am and 2pm with the lowest volume days overall being Friday and Saturday.

AI spend but with different filters that allow you to break down by type of spend (thinking, tool context, global prompt, etc.)

AI segments but with different filters that allow you to break down by type (thinking, tool context, global prompt, etc.). Remember 1 message is always equal to 1 segment unless you have unlocked an agent node within a job flow.

Credit remaining at each moment. On agency plans it covers your clients' wallets too; view by source to see each ones own wallet balance.

The chart below is a table sorted by balance, showing the amount of money each source under your agency account has within their wallet. It looks like 3 sources are out of money and need topped up. Maybe their card is expired!

Every nudge your AI sent to a lead who went quiet. Followups are set up within your job flow settings. Learn more about setting up follow-ups here.

This example below shows follow-up count segmented by job flow across the past 4 weeks. We can see the purple job flow is sending a lot of follow-ups, but it has recently decreased. Alternatively, the orange job flow has had a large spike in followups.

This second example shows AI Spend on the same chart as follow-ups generated. You can see the correlation between the follow-up count and AI Spend over time.

How your leads actually feel in conversation, scored 0–100 (50 is neutral). Catch a dip before it costs you the deal.

See the chart below where we show average contact sentiment over time. This shows that leads have been getting happier over the past 4 weeks.

Every time one of your agent nodes took action with a tool… booking, tagging, updating your CRM. Break it down by tool if you want to see which tools are used more and when.

See the chart below where we enumerate the counts of all tools used by agents over the past 3 days. Check availability happens a lot followed by no reply suppressed (our update that allows you to tell the AI within the prompt itself not to reply in certain scenarios).

A special chart type that shows discussed topics, shaded by sentiment so you know which topics result in unhappy vs. happy leads and when those topics are discussed.

See the chart below that shows topics discussed over a timeframe and the general happiness score (sentiment) when the leads are chatting about each of the topics.

Questions your bot couldn't answer with the info you gave it. Your Agents create Smart FAQs for you to fill out so it knows it next time. Decreases in FAQs added over time means your Agents are getting closer to having their knowledge gaps fully filled.

Below shows a bar chart of FAQs added over time week over week. It looks like there was a 2 month period in this example where the AI had more gaps in knowledge. It may be worth looking into why that is.

How often your bot reached into your knowledge base to answer a question. Even if 3 documents are referenced to reply to 1 message, it's just counted once.

The pie chart below shows an example of a pie chart that shows the amount of knowledge base references per source over the past 6 months.

Which of your documents your bot’s using earning its keep and what needs rewriting.

See the example below where we show the total amount of times documents are referenced segmented by the type of document over the past 3 days. It looks like Google Drive was a popular reference item yesterday!

Site Visitors who visited a site with your pixel installed for the first time.

Site visitor sessions, known crawlers are excluded

Pages visited on sites running your sales pixel, known crawlers excluded

Visitors who gave their contact information through the chat widget

There's no other platform that gives you such depth into the analytics of the conversations your Humans and AI Agents are having with leads. Use any of the metrics above or the formulat metric to display anything you could want depending on what's important to you and your clients.