<!-- JSON-LD markup generated by Google Structured Data Markup Helper. --><script type="application/ld+json">{  "@context" : "http://schema.org",  "@type" : "Article",  "name" : "Metrics for chatbots: Conversational Analytics | Part 1",  "author" : {    "@type" : "Person",    "name" : "Neerajshikhar Joshi"  },  "image" : "https://global-uploads.webflow.com/5ef788f07804fb7d78a4127a/5ef853da7c7d225accc6e2d8_Metrics%20for%20Chatbot-%20Conversational%20Analytics%20_Part%201.jpg",  "articleSection" : "measuring bot performance",  "articleBody" : [ "User Metrics", "Total Users", "Active Users", "Engaged Users", "New Users" ],  "url" : "https://www.engati.com/blog/metrics-chatbots-conversational-analytics",  "publisher" : {    "@type" : "Organization",    "name" : "Engati"  }}</script>

Conversational analytics part 1: What bot metrics should you track?

Neerajshikhar Joshi
min read
Conversational analytics part 1: What bot metrics should you track?


Today, we decided to write about metrics for chatbot and its importance in conversational analytics. As we were going through our previous blog about crafting conversational UI for chatbots, we noticed that we had only touched upon the metrics. If you haven’t had a chance to read our previous blog, you can find it here.

How do you measure productivity delivered by a chatbot? After all, there is a reason chatbots are employed in the first place, and every organisation has a unique objective that they want to achieve. There is no doubt that chatbots are going to be a big part of our future. Studies inform that 80% of businesses are reported to get  on the chatbot bandwagon by 2020.

The funny thing is, not many companies know how to gauge the efficiency of chatbots in the first place. This, perhaps, explains why the chatbot ride has been so bumpy for a few. We say ‘a few’ because some companies have seen tremendous success implementing chatbot technology, while some others have fumbled or failed in their attempts. Some chatbots took off easily and some crashed and burned.

So, how does one ensure that the chatbot one invests in does not fail to perform?

Same way you ensure anything, you test it, analyse the results, tinker with it and repeat till you meet you benchmark for performance. Which brings to our topic, how do we measure a chatbot’s efficiency?

A metric is a quantifiable measure that is used to track and assess the status of a specific business process

Since many of the capabilities of a chatbot can be measured using metrics, it is important that you employ the right metrics to make most of the data. New chatbots can show erratic fluctuations when it comes to metrics. Companies should vigilantly monitor new chatbots after implementation.

Generally the main points of focus revolve around efficiency enhancement, greater conversion and faster responses, defining the right metrics plays an important role. This ensures that you can monitor chatbots for performance, efficiency, and improvement.When it comes to metrics for measuring bot performance, there are 3 main classifications that we will be looking at.

  • User Metrics
  • Message metrics
  • Bot metrics

User Metrics

These are the metrics that help you index the user related activities. They show how engagement is distributed amongst users.

User Metrics - Engati
Photo by Ben Kolde on Unsplash

User Metrics

Total Users

This is the most fundamental metric. It informs you about the number of people that interact with your chatbot. This metric is essential as it’s trend showcases the change in the number of users and consequently the amount of data you expose to your chatbot. Also, this provides crucial information about the market size and your chatbot’s overall effectiveness.

Active Users

The number of people that are reading messages from your chatbot within a given time frame are your active users. These users are also your potential target audience. You can estimate your promotional activity impact using these numbers. These numbers  show how many people have seen your promotional content. Just like in social media, you cannot guarantee engagement but the content is visible to users.

Engaged Users

These are the users that are exchanging messages with your chatbot, they are in communication with it. This sub-sample enables your chatbot to provide you with conversation statistics, hence it is an important metric to apply. This metric helps inform  your decisions about the effectiveness of your chatbot, since a chatbot that cannot start a conversation with users is almost irrelevant.

New Users

This metric will help capture the effectiveness of promotional efforts employed for your chatbot. New users keep the pool of active users fresh, older users tend to lose interest in interaction over time. Hence, you can make promotional decisions for the chatbot  informed by this metric.


Once you have an understanding of how to make informed decisions by using the right user metrics, you can layout strategies for your chatbot to be more effective in engagement. However, there are some more metrics that are crucial for making a chatbot more productive. We will be discussing message metrics and bot metrics in our next blog. Stay tuned for more information.

Meanwhile, if you want to experience a chatbot platform that saves you a lot of time and trouble and provides smooth interface for building your dream bot, check us out!

Photo by Oleg Magni on Unsplash

No items found.
About Engati

Engati powers 45,000+ chatbot & live chat solutions in 50+ languages across the world.

We aim to empower you to create the best customer experiences you could imagine. 

So, are you ready to create unbelievably smooth experiences?

Check us out!