What do you like best about Ada?
It is customizable and it is user-friendly.No-Code / Low-Code Design
You don’t need to be a developer to build or update chatflows.
Teams can easily set up responses, decision trees, and logic through a visual interface.
Powerful Integrations
Works well with systems like Zendesk, Salesforce, Shopify, and internal CRMs.
Lets customers get account info or submit tickets without switching platforms.
Omnichannel Capability
Ada can run on web, mobile apps, social media (Messenger, WhatsApp), or SMS — with a consistent user experience across all.
AI + Natural Language Understanding (NLU)
Can understand how users phrase questions differently (e.g., “block my account” vs. “deactivate my wallet”).
Learns over time from interactions, improving accuracy.
Scalable for High Volume
Can handle millions of interactions simultaneously, useful for companies like GCash with huge user bases.
Customizable User Journeys
Businesses can tailor flows — like ID verification, balance check, or ticket creation — to match their exact process and tone.
Real-Time Handoff to Human Agents
Smoothly transfers complex issues to a live agent in Zendesk or another CRM without losing chat history.
Analytics & Reporting
Offers dashboards showing metrics like resolution rate, deflection rate, and common queries — helping improve automation coverage.
Security & Compliance
Enterprise-grade encryption and data privacy features suitable for fintech and banking environments.
Multilingual & Localization Support
Can understand and reply in multiple languages, including Taglish, which is particularly useful for Philippine customers. Review collected by and hosted on G2.com.
What do you dislike about Ada?
1. Limited “True AI” Capabilities
Ada uses pattern-based understanding (NLU) rather than full generative AI.
It recognizes intent and keywords well, but it can’t fully “converse” freely like ChatGPT.
Complex or unexpected phrasing may still confuse it unless trained specifically.
2. Requires Strong Maintenance
To keep accuracy high, Ada needs continuous training and content updates.
If intents, flows, or keywords aren’t regularly reviewed, the bot can misroute or give outdated answers.
3. Dependent on Scripted Flows
Most of Ada’s responses are predefined, so it can’t reason dynamically.
It follows decision trees — if a user goes off-script, Ada may fail to recover or get “stuck.”
4. Limited Context Memory
Ada can recall the current conversation but doesn’t retain long-term user context (e.g., what the user said in past chats).
This limits personalization unless connected to a CRM or database.
5. Integration Complexity
While it integrates with Zendesk, Salesforce, etc., setup can be complex and time-consuming.
Errors in API configuration can cause ticket syncing or data handoff issues.
6. Analytics Depth
The built-in analytics dashboard gives general data (volume, resolution, escalation),
but advanced insights (like user sentiment or root cause analysis) require third-party tools or exports. Review collected by and hosted on G2.com.