Use case: Product & Customer Support
Goal
Decrease the time customers spend solving issues when they arise by offering 24/7 multi-language automated AI front-line support
How
Quickly build self-service conversational AI chat agents that help users and customers find the right technical information to solve their issues. If they can’t find that information they easily escalate their issue to traditional methods where helpdesk staff see the entire conversation transcript to help them respond quickly to resolve these problems.
Listen, personalize, measure...
Get started...
First, create the agent: We make it very easy to embed USEFUL product and lead generation agents into your websites. We take your current support knowledge base documents or URLs and add it to our database in ways that make it searchable. Then you copy a simple iFrame embed code and put it into your WordPress or other website (ask anyone under 25 for help).
Then, respond to interactions: All conversations are captured and tracked, so you can see what customers are asking, where they might be confused, and what they have tried. If those customers click a call to action on your agent, you get notified by both email and in-app flags.
Is it expensive? No, our plans start at $149 per month. Compared to a multi-lingual, 24/7 available, trained product expert…well, we think it pays for itself.
Frequently Asked Questions...
- Web Pages via their URL (we discover pages based on a website sitemap.xml)
- Text files (great for video transcripts, or other reference material)
- PDF Files
- Markdown files
- All Major Code language files (lots, email us if you have something exotic)
- XML files
Note: We are always adding more types of documents.
The built in Agent Panel is fully customizable via CSS (all colors and layout). Custom CSS files can be linked in the control panel, full documentation can be read here. On the rare occasion complete interactive control of the Agent Interface is needed, our panel is Open Source on GitHub and can be forked and customized.
No, and we can help get you started. Good agents take time and actual use to answer the wide variety of questions the way you would have. It starts with general instructions and reference material that is uploaded and ingested. After that, you start asking the agent questions you think customers will ask and assess the results, and refine the gaps.
What are the instructions for help.llmasaservice.io (shown above):
“You are a helpful agent answering users questions using the context provided by our knowledge base. Do not answer any questions unrelated to how to use or integrate llmasaservice.io product into website and applications.”
And we added all of the pages on our knowledge base.
If you currently use Zendesk, Intercom, JIRA Service Desk, you have a very full featured product for service agents. They are strong at the legacy way of managing helpdesks. Our value is the self-service AI driven experience for customers. The major vendors are of course implementing AI features, but we think ours is more cost effective and in most cases responds equal or better (and we don’t charge $1 per conversation).
Our strength is the customer relationship and conversation analytics. Our summarization and Conversation Intelligence features help improve the self-service experience and filling the gaps in knowledge base material.