Chatbot: definition, value and limits
Chatbot Updated
Definition
A chatbot is a text-based dialogue service that answers user messages in a channel such as a website, app or messenger. It answers scoped questions, guides through forms or hands over to a human. It is limited to text and usually to one channel and works either along fixed dialogue paths or with a language model.
Why the Term Matters
The term is often equated with any kind of AI surface, although a chatbot is narrower. For the mid-market the clear boundary matters, because a chatbot without a link to CRM or ticketing quickly becomes a dead end. Who defines the term cleanly plans the chatbot as a channel surface over a connected process, not as an isolated answer machine.
When It Is Relevant
- A channel receives many similar standard questions.
- Answers can be derived from checked knowledge.
- Requests should be pre-qualified and then handed over cleanly.
- A clear path to a human exists for complex cases.
When Not
When requests are highly individual and need professional judgement, a chatbot frustrates more than it helps. And when it is not connected to the systems, it can chat but cannot create a case, check a status or truly relieve a process.
Industries That Use It
- Online retail: answer delivery-status and returns questions right in the web shop.
- Municipal utilities: catch standard tariff and meter-reading requests in the customer portal.
- Education providers: guide prospects through course choice and enrolment.
- Trade businesses: take appointment requests and pass them structured to scheduling.
Typical Use Cases
- Answer frequent questions from checked knowledge and name sources.
- Pre-qualify a request and hand it to a human with a summary.
- Run an application dialogue and write the data structured into a system.
- Retrieve the processing status of a case from the connected system.
- Take requests outside service hours and forward them later.
Decision Guide
| Signal | Chatbot fits | Alternative |
|---|---|---|
| Many similar text questions | a chatbot in the channel | a static FAQ page |
| Case creation needed | a connected chatbot | a plain contact form |
| Cases highly individual | straight to a human | a chatbot frustrates |
| Several systems involved | an assistant, not a chatbot | a chatbot alone too narrow |
Example
An online retailer runs a chatbot in the web shop that answers delivery and returns questions from checked knowledge and, when needed, pulls the shipment status from the connected system. Complex complaints go to service with a conversation summary. Value is measured in relieved standard requests. This describes the method, not an asserted client metric.
Service Connection
A chatbot belongs on a connected process, so the entry is often process automation with AI and an RPA build, framed through the process automation consulting.
Sources
This framing describes common mid-market practice for chatbots. Concrete implementation, connections and data protection depend on the case, the systems and the tools and must be checked case by case.