Chat History

1. What is Fini’s Chat History feature and how does it work?
Chat History gives you a centralized view of all past customer conversations across sources like Zendesk, Front, and Intercom. It displays the question asked, the AI-generated response, ticket ID, categories, and metadata in one dashboard, enabling transparency and faster issue diagnosis.

2. How does Fini categorize conversations in Chat History?
Fini uses AI to auto-categorize each chat based on the topic, intent, or issue type. These categories help in organizing conversations for analysis, feedback, and content improvement without manual tagging.

3. Can I filter chats by platform in Fini's Chat History?
Yes, filters allow you to isolate conversations by source—Zendesk, Front, Intercom, etc.—making it easy to audit channel-specific performance or dive into certain team workflows.

4. What insights can I extract from the Chat History feature?
You can analyze conversation quality, trace faulty responses to their knowledge source, identify repetitive questions, and decide which content areas need refinement or escalation paths.

5. How does the action column in Chat History improve support operations?
The action column allows you to quickly inspect conversation flow, identify what information the AI used, and take corrective actions on knowledge items or configuration settings without toggling between tools.

Analytics

6. What metrics does Fini’s Analytics dashboard provide?
You get resolution rate (AI-handled vs escalated), usage patterns, agent transfer rate, and knowledge coverage per category—offering both macro and granular views of your AI agent’s performance.

7. How does Fini calculate content coverage in the knowledge base?
Fini maps questions asked by users to knowledge categories and highlights areas where the AI lacks content, enabling you to plug gaps and ensure complete topic readiness.

8. Why is the agent transfer rate an important metric to track?
It reveals how often conversations require human intervention. A high transfer rate may indicate poor knowledge coverage, complex workflows, or insufficient escalation training for the AI.

9. Can analytics help improve the AI model's accuracy?
Absolutely. By identifying categories with low resolution rates or high escalation volume, teams can refine knowledge items, retrain the model, or use flows to handle complex cases.

10. How do Fini’s analytics differ from traditional ticketing dashboards?
While traditional dashboards focus on ticket volume and time-to-close, Fini’s analytics measure AI performance, knowledge health, and resolution effectiveness across entire conversations.

GPT-4o Integration

11. What is GPT-4o and how does Fini leverage it?
GPT-4o is OpenAI’s latest multimodal language model optimized for speed and accuracy. Fini integrates it to power its AI agents, enabling faster replies, better understanding, and higher resolution rates.

12. How does GPT-4o affect AI response times in Fini?
It delivers up to 2x faster responses compared to previous models, which drastically reduces user wait times and improves CSAT in real-time support environments.

13. What accuracy improvements can customers expect with GPT-4o?
Fini AI agents now achieve over 95% accuracy in supported categories, particularly when resolving billing, account, and transactional queries with enhanced contextual awareness.

14. Is GPT-4o available across all Fini integrations?
Yes, the GPT-4o upgrade applies universally across all channels and integrations—Zendesk, Intercom, Front, Slack, Discord, and more—ensuring consistent quality.

15. Does GPT-4o impact how flows or knowledge items are processed?
Yes, its deeper comprehension capabilities allow flows to branch more naturally and knowledge items to be interpreted more contextually, improving the overall support journey.

Front and LiveChat Integrations

16. What can I do with the new Front and LiveChat integrations?
You can deploy Fini directly into these platforms to handle email or chat queries. Fini supports internal notes, auto-categorization, and direct AI replies in both systems.

17. How do internal notes work in Front with Fini?
Fini can leave internal comments within Front tickets, helping agents understand why a response was made, what data was used, or what action is required next—without replying to the customer.

18. Can Fini resolve tickets autonomously in Front or LiveChat?
Yes, in many cases Fini resolves tickets without human intervention by replying with accurate, policy-aligned answers and flagging complex cases only when needed.

19. Is training needed to use Fini in LiveChat or Front?
No special training is required. Fini comes with no-code deployment options and a guided onboarding process for fast rollout and customization.

20. What makes Fini’s integrations more powerful than chatbots?
Unlike basic bots, Fini acts as an agentic AI that takes actions, understands business logic, categorizes conversations, updates workflows, and scales across channels.

Coming Soon Features

21. What is the “Fini Internal Knowledge” feature?
It lets you create dynamic macros within Fini, either manually or automatically from past chats and feedback. This feature helps build private, internal content that guides AI replies without being customer-facing.

22. How does Flows work in upcoming updates?
Flows are conversational micro-agents that guide users through structured journeys (like cancellations or refunds). They support multiple decision branches and integrate with attributes and APIs.

23. What are user attributes and why do they matter?
User attributes are custom metadata (e.g., plan type, churn risk) that personalize how Fini replies. With this, the same question could receive different answers depending on the customer’s profile.

24. Will Fini’s flows be able to take direct actions?
Yes, the next version of Flows will include support for API-triggered actions like canceling subscriptions, updating payment methods, or providing tailored product suggestions in real time.

25. What is table querying and how is it different from RAG?
Table querying allows AI to fetch structured data using SQL-like queries (e.g., “top 5 orders under $50”). It goes beyond RAG by interacting directly with databases to return accurate, real-time answers.

Platform Management

26. What are the new user roles introduced in Fini?
Fini now supports three roles:

• Agent – Can only use Ask Fini

• Member – Has access to Ask Fini, Chat History, Analytics, and QA

• Admin – Full access to the platform including integrations, flows, and configurations

27. Why are role-based permissions important in support tools?
They provide security and reduce clutter by giving users access only to what they need, improving productivity and minimizing the risk of accidental misconfigurations.

28. Can Admins customize what Members can see?
Not yet, but future updates may include granular permission controls beyond the current Agent/Member/Admin structure.

29. Will roles affect how users interact with Chat History or Analytics?
Yes, only Members and Admins can access performance data or past conversations, ensuring sensitive data is available only to those who need it.

30. How can I assign or update roles in Fini?
Roles can be assigned during user onboarding or updated anytime from the Admin dashboard. Changes take effect immediately.

Getting Started

31. Do I need engineering support to use these features?
Most features, including Chat History, Analytics, Flows, and integrations, are fully no-code. Fini is built to be self-service for CX and ops teams.

32. How long does it take to set up GPT-4o-powered Fini agents?
Setup can be completed in under a week, depending on data availability and integrations. Fini offers guided onboarding to accelerate deployment.

33. Can Fini work alongside human agents or replace them?
Fini is designed to complement human agents by handling repetitive or simple tickets autonomously and escalating edge cases for human review.

34. What data does Fini use to train its AI?
Fini uses anonymized historical conversations, knowledge base content, and structured attributes. It does not train on customer data by default unless explicitly configured to do so.

35. Is Fini safe to use in regulated industries like fintech?
Yes, Fini includes over 40 guardrails for compliance, including PII detection, tone control, hallucination suppression, and customizable validation filters.

Results and ROI

36. What kind of resolution rate can I expect from Fini?
Many customers see 70–90% resolution rates within weeks of deployment. For example, Column Tax resolved 70%+ of tickets with over 90% accuracy.

37. How does Fini impact SLA metrics?
By resolving queries instantly and reducing backlog, Fini improves SLA adherence dramatically. For some clients, SLA improvement exceeded 120%.

38. Can Fini reduce support costs?
Yes. With high automation and low escalation volume, Fini reduces ticket load, labor costs, and operational overhead by up to 50% or more.

39. How do I measure Fini’s impact after launch?
Use Fini’s built-in analytics to track resolution rate, deflection, CSAT trends, knowledge coverage, and escalation volumes—along with feedback tags from users.

40. Where can I learn more or book a demo of these features?
You can explore these updates in your Fini dashboard or book a personalized demo tailored to your team’s needs.