In today's increasingly digitized economy, the customer experience is a decisive factor in the success of any company. When the entire user journey takes place online, including all transactions, the quality of the experience and the resulting relationship with the customer are crucial to making the business model work. This is particularly true in e-commerce.
Customer expectations are constantly rising. Every seamless purchase, easy payment, or quickly resolved service request they experience with another company broadens their understanding of what is possible. The possible becomes their standard and what they expect from you as well.
Whether on your website or through channels like Facebook Messenger and WhatsApp, messaging is one of the most promising additions to the customer experience playbook. While live messaging has consistently shown excellent results in the past, it is Conversational AI in the form of AI chatbots that has unlocked this channel for large-scale business.
The direct and personal messaging experience in the app that your customers are already using creates proximity, individual relevance, and seamless accessibility without forcing them to change channels. But there is more to it than just the familiar mantra of "be where your customers are."
To leverage the benefits of messaging throughout the customer journey, a blend of human live chat and AI automation will deliver the best results. However, most implementations fail when it comes to the collaboration between human support agents and AI chatbots and combining their strengths.
While the conversational capabilities of chatbots are a topic for another post, we want to highlight a fundamental issue with most messaging and chatbot software providers that is often overlooked. This not only heavily impacts the quality of conversations but also your ROI for messaging in general.
To understand the problem, we need to look at the business models of SaaS platforms offering messaging and chatbot services, as well as their value propositions. As is common with license-based software, there is typically a "consumption" metric used for pricing. The more of a variable unit you consume, the higher your monthly price. Broadly speaking, there are two types of such metrics:
Billing by number of customer contacts: This approach is prevalent among platforms originating from the marketing automation space or those that primarily sell to marketers. The idea here is that user data is the key output of the tool, and the value lies in generated leads. The more contacts—meaning the more value—the higher the fee. This approach makes sense for lead generation chatbots, and that is where these platforms are most commonly used. Consequently, however, these tools offer limited features to address needs that arise later in the customer journey. The actual value there is not created by adding more users, but by nurturing relationships with existing customers through engaging customer service.
Billing by the number of support agents: This pricing approach is typical for CRM, helpdesk, and customer service tools. The core concept is the same as with other tools used by a professional to complete a specific task. It focuses on the number of people doing the work and then sells the tool "per seat". In customer service, this means that given your regular volume of service requests, you need a certain number of live chat agents. By paying per agent, you are able to give all these agents the tool they need to handle the volume.
Since this pricing model is built on the number of agents manually handling the support volume, it is interesting to consider what happens when chatbots become part of the equation. Let's look at what chatbots typically do in messaging platforms where you pay per agent: they either perform low-level FAQ automation that reduces inquiry volume just enough to keep your agent team growing steadily, or they perform pre-qualification and agent routing to make better use of your available agent resources.
As a rule, however, they do not fully automate service requests and thus cannot decisively reduce the volume that hits the agents. And why should they? It reduces "consumption" and therefore the provider's revenue. When we now think about how important it is to combine human live chat agents and AI chatbots to provide a personalized customer experience along the entire customer journey, we see a fundamental flaw in this approach.
So, how can we price a messaging platform that does not discourage effective automation?
At Mercury.ai, we have chosen an approach that creates positive incentives for ideal collaboration between live agents and AI automation, maximizing chatbot ROI. This approach follows two basic principles.
First: The number of agents is unlimited. We do not charge fees for live agents.
And we mean exactly that: you do not pay a fee per agent seat or per active user.
We believe that live agents drive quality through their work, not just plain quantity. Therefore, they should not be a cost factor.
Furthermore, the introduction of messaging as a customer communication channel should not be hindered by considerations of how many of your agents "really need access". Instead, the focus should be on identifying inquiries that can be automated. Then, you can quickly and seamlessly insert AI components to handle these tasks.
Second: You pay for the range of functions. So, the price increases with the value delivered.
We believe that the value of AI assistants lies in the tasks they can perform. An AI assistant that can answer questions, create a ticket, and schedule appointments is more productive than one that can only answer questions. Accordingly, functionality is our "consumption" metric.
The foundation for this pricing approach lies in the modular nature of our AI chatbots. It allows for simple, functional upgrades of the bots, expanding their feature set. Many features already exist in our library as modules and can be added to a bot without much effort.
This means that as a customer, you only pay per module. So, if analysis shows that your live chat agents are spending time on a series of recurring return inquiries, you enable the "Returns" module. Instead of investing your agents' time, you pay a flat fee for the module.
The more functions you add to a bot, the more it will be able to automate, and the more manual service volume will be reduced. The difference is that you do not pay extra to get the desired results from your AI chatbot.
So, don't pay for live agent seats - pay for the value of automation instead.
Sounds too good to be true? Talk to us and we will run the numbers for your use case.
