惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

TaoSecurity Blog
TaoSecurity Blog
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
F
Fortinet All Blogs
Cisco Talos Blog
Cisco Talos Blog
D
Darknet – Hacking Tools, Hacker News & Cyber Security
S
Secure Thoughts
美团技术团队
雷峰网
雷峰网
Hugging Face - Blog
Hugging Face - Blog
博客园_首页
C
CXSECURITY Database RSS Feed - CXSecurity.com
Engineering at Meta
Engineering at Meta
人人都是产品经理
人人都是产品经理
月光博客
月光博客
T
Tor Project blog
P
Privacy & Cybersecurity Law Blog
Recorded Future
Recorded Future
I
Intezer
博客园 - 【当耐特】
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
GbyAI
GbyAI
罗磊的独立博客
V
V2EX
Google DeepMind News
Google DeepMind News
D
DataBreaches.Net
Last Week in AI
Last Week in AI
T
Tailwind CSS Blog
www.infosecurity-magazine.com
www.infosecurity-magazine.com
A
About on SuperTechFans
Scott Helme
Scott Helme
Vercel News
Vercel News
Spread Privacy
Spread Privacy
T
Threat Research - Cisco Blogs
Recent Announcements
Recent Announcements
Hacker News: Ask HN
Hacker News: Ask HN
C
CERT Recently Published Vulnerability Notes
G
Google Developers Blog
B
Blog
博客园 - 叶小钗
WordPress大学
WordPress大学
博客园 - 聂微东
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Jina AI
Jina AI
IT之家
IT之家
C
Cybersecurity and Infrastructure Security Agency CISA
P
Palo Alto Networks Blog
小众软件
小众软件
博客园 - Franky
Microsoft Azure Blog
Microsoft Azure Blog
AWS News Blog
AWS News Blog

Amplitude

What Makes a Good vs Bad North Star Metric The Role of Feature Management in Successful Product Development Cohort Retention Analysis: Reduce Churn Using Customer Data 7 Steps to Measuring the Success of a Feature 14 Best Product Management Tools for 2026 (Plus Tips from Senior PMs) Putting A Number On AI Quality Meet the Winners of the 2026 Amplitude AI Impact Awards Beyond Last-Touch Attribution: Find Out Which Interactions Really Matter Agent Connectors Are Better Together Agents That Act on What Actually Happened How Square Used Amplitude to Enhance the Seller Experience and Power Growth Migrating Analytics Platforms Without The Chaos Wanted Lab Grows Sign-Ups by 150% & Builds Experimentation Culture How to Balance Inference Cost and User Experience for Agents Introducing Zoning Insights: Web Intelligence at a Glance Five best practices for getting started with AI agents 24 Quarters at #1. Here’s What’s Next. How We Built a Product That Tells Us What To Build Next: Inside Amplitude Wave Looking Beyond Campaign Metrics: 7 Marketing Success Stories AI Evals for Product Managers: A Beginner’s Guide to Getting Started The Builder Skills Library Introducing Agent Connectors in Amplitude Understand How AI Thinks, Get Better Results How We Redesigned Amplitude Docs for Agents and Made Everyone an Author AI Broke Your Experimentation Program. Here’s How to Fix It. Every Stuck User Is a Support Ticket Waiting to Happen Tracing the Sale: Connect Behavior to Conversions with Persisted Properties Building CLI Agents: It’s What You Don’t Give Them That Counts Three Tips for Better Prompts in Amplitude Global Agent How AI Took the Data Analyst’s Job, and Created a Better One Default Prompts Are Tanking Your Agent’s Retention Optimizing Core Web Vitals with Amplitude’s Global Agent Don’t Ask Global Agent Anything, Ask These Three Things How We Built a Design Agent at Amplitude with Claude Managed Agents and Cloudflare The Problem with Chasing Churn How Hostinger Achieved a 20%+ Conversion Lift Through Experimentation How STAGE Streams Smarter by Putting Data at the Center Building the Validation Stack for AI Product Development Making AI Analytics Safe for Financial Services Teams Amplitude Heatmaps Update: More Reliable Screenshots and Accurate Placement Most Teams Ship Agent Personalities by Accident. We Didn’t. What I Learned Pointing a Ralph Loop at My Product for a Week How Mercado Libre Scales Decision Making with AI Claude Cowork for PMs: 5 Playbooks to Get Started Agents Just Made Your Feature Launch Channel Smarter Homegrown FinOps Tools: How AI “Build” Beat “Buy” for Us in <1 Year Introducing The Amplitude Quickstart Series Rebuilding Session Replay’s Delivery Layer to Be Lighter on Your Page The Eval Signal That Predicts 3x Agent Retention Agents Write Code. Fixing It Is Still On You. Amplitude and Statsig Partnership 5 Agent Skills to Automate Your Weekly Product Review Amplitude Plug and Play: New AI Plugin in Claude and Cursor Marketplaces Introducing Amplitude Wizard CLI: Set Up Amplitude from Your Codebase Making AI Search Count (and Convert) How VEED Evolved Its AI Search Strategy What’s New with Amplitude Agents Effortless Support at Scale: Making Human Support More Human AI Week 2026: Upleveling All Together Amplitude AI Builders: Paul Hultgren Chats about AI Assistant Dashboard Dread to AI-Driven Decisions: How Tira Rebuilt Its Analytics Workflow Your Product Deserves a Better Support Agent How Cisco Systems Accelerated Adoption by 20% Through Data Innovation
How ACKO Drove 13% More Conversions & 50% Drop in Calls with GenAI
Agastya Sanjai · 2026-05-12 · via Amplitude

Insights/Action/Outcome: ACKO sought to streamline the high-pressure insurance-buying experience by creating a seamless, asynchronous chat journey. They used Amplitude’s funnel and cohort analysis to deploy a GenAI-powered chatbot, resulting in a 13% increase in health insurance conversions, a 50% reduction in inbound call volume, and an 88% reduction in answer latency.


Buying insurance can be a battle of nerves. Cost-conscious consumers flit from one provider to another, but go out of their way to avoid pressure-laden follow-up calls from aggressive sales reps.

ACKO wanted to do something different.

I’m the Lead AI Product Manager at ACKO, a digital-first insurance company in India that provides a comprehensive range of insurance products to nearly 80 million customers. We use analytics and AI for instant policy approvals, fast claim settlements, and paperless convenience. My job is to listen to our customers and leverage AI to help them enjoy a seamless user experience. And that meant moving away from stressful phone calls.

Solving one problem creates another

Our research showed that three out of four ACKO customers preferred a relaxed online chat over a high-pressure phone call, so we turned their anxiety into an opportunity.

What started as a bold hackathon winning idea soon evolved into a fully homegrown, enterprise-scale GPT-powered conversational agent, a first-of-its-kind for the Indian insurance space. Rather than a simple rule-based bot, we developed an AI agent capable of understanding free-form questions through goal-oriented dialogue.

Inspired by real sales agents, the asynchronous agent empowered customers to find answers at their own pace, instead of feeling chased. It was a big win, but in solving one challenge, we uncovered another: a wealth of insurance FAQs and prior customer queries remained locked within our chat interface and calls.

At its heart, this was a discovery problem. Users who didn’t proactively initiate a chat missed out on the very information they needed to feel confident enough to convert. How could we surface relevant questions and answers to spark their curiosity, counter their hesitancy, and lead them further along their conversion journey?

Unlocking conversational insights at scale

We knew providing personalized, timely, and pertinent assistance would boost the percentage of users who ask for help, leading to more conversions.

To meet our customers where they are, we needed a ‘conversation debugger’ that could mine user intent and help us answer their questions in real time. Amplitude was key to building that solution.

To meet our customers where they are, we needed a way to mine user intent and help us answer their questions in real time. Amplitude was key to building that solution.

  • Tracking intent in real time: Segment integration lets us feed Segment data into Amplitude and take deep dives into user interactions.
  • Visualizing the user journey: Funnel and cohort analyses help us visualize user journeys and identify drop-off points.
  • Spotting conversational friction: Pathfinder and Retention charts identify sticking points in our dialogue flows. One sticking point (in which users tapped multiple suggestion chips without progressing in terms of sales intent) informed our decision to introduce goal-oriented suggestion sequencing.
  • Iterating quickly: Amplitude Web Experimentation additionally gave us what we need to run A/B tests and iterate on the conversational experience.
  • Democratizing live insights: Amplitude notebooks are how we share insights, highlight key metrics, and generate charts and graphs that drive support and expansion.

With access to these features, we moved from week-long sprints to shipping knowledge & experience improvements in hours. Non-engineers can deploy end-to-end bot journeys on their, cutting our launch time by 6x and making experimentation our default operating mode.

Data tells us everything about the user experience

Digging into user behavior funnels and cohorts revealed critical insights.

For example, we discovered that nearly 50% of users walked away after exchanging three messages with our chatbot, whether or not they got the answer they needed. To alleviate this pain point, we added a Call Upgrade button that prompted them to request human assistance before they dropped off.

We also introduced Question Cards—tappable, curiosity-driven prompts that displayed popular queries—to steer passive browsers into active conversations. Our chatbot mirrors the guiding presence of a real salesperson on key pages, surfacing questions, providing recommendations, and seamlessly transitioning to human help without breaking the conversational flow. After launching, we additionally trained our model on trending user questions such as “comparing competitor plans alongside ACKO’s”, helping them find hospitals near them, and nuanced questions around existing diseases.

Amplitude confirmed that our Question Cards led to a 3x increase in chat initiations, and those who engaged with a card converted at a 40% higher rate. By lowering the barrier to entry through chat, we realized a 200%+ increase in engagement.

Additionally, we have implemented a warm handover process. For cases requiring human intervention, call center agents receive a summary of the chat interaction so they’re prepared for the conversation, and customers don’t have to repeat themselves.

50% lower call volume and millions more in premiums

Within a year, our GenAI-powered chatbot drove a 13% increase in health insurance sales conversions—compared to our previous callback experience. This uplift translates into a million-dollar increase in user premiums.

We also deployed the new chatbot on our recently launched automobile purchase platform, ACKO Drive. Customers who interacted with the chatbot were 2.5x more likely to proceed to payment and complete the checkout funnel than those who didn’t. In other words, driving conversational engagement leads to greater revenue gains.

Customers who interacted with the chatbot were 2.5x more likely to proceed to payment than those who didn’t. We found that conversational engagement leads to greater revenue gains.

Additionally, our GenAI-powered chatbot has significantly cut down our need for outbound sales calls. We’ve seen a nearly 50% reduction in inbound call volume, indicating that users who previously would have scheduled calls are happy to complete their purchases through our chat interface.

We segmented this cohort in Amplitude and found that customers who switched from voice calls to online chat are converting at the same or slightly higher rate as before. They are enjoying a more convenient experience while freeing our agents to focus on high-touch cases rather than fielding basic queries, further reducing operational costs.

Customer service that prizes empathy and data

With Amplitude, we’ve engineered a mindset shift at ACKO. We have gone from a strictly “Call Us” mentality to a more functional “Get Help” approach that favors our customers, creating a digital-only user experience that truly meets customers where they are and where they want to be.

Our chat CSAT (customer satisfaction) score has surged to 80%—an increase of more than 17% across our high volume health and auto segments. We’ve since launched across all our remaining D2C products (such as Travel, Life and Bike insurance) and have achieved a similar ~80% CSAT. The conversational flow, fast response times, and seamless handovers have created a friction-free journey, and we regularly hear users express delight with the chatbot and the overall customer experience.

We continue to iterate and make our chatbot faster and more accessible. So far, we’ve had success caching popular Q&A pairs and optimizing our mobile UI, reducing latency and cutting answer times by 88%.

Thanks to Amplitude, ACKO is proving that empathy, backed by data, is the ultimate expression of customer service and the greatest possible competitive advantage in the insurance space. No pressure required.

Thanks to Amplitude, we proved that empathy backed by data, is the ultimate expression of customer service and the greatest possible competitive advantage in the D2C space.