Plotline | 2025

Conversational AI builder for consumer apps

Conversational AI for consumer apps //
Can AI actually help users in-context?

Every customer has a unique set of aspirations, sensibilities and expectations from consumer apps. Capturing the intent and assisting users at the right time is what every app aims for, but as B2C apps grow more complex, users struggle to complete key actions.

Team

1 Designer, 1 PM, 3 Engg

From scoping to launch

4-5 weeks

Improvement in discovery

~40%

Reduction in time-to-value

~20%

Reduction in drop-offs

~15-20%

Avg reduction in support tickets

~30-40%

// CONTEXT //

What does Plotline do?
and what it couldn't do yet

What does Plotline do? and what it couldn't do yet

Plotline helps growth teams at consumer apps build in-app experiences — stories, nudges, walkthroughs, scratch cards — without writing code. The platform sits between a marketer's intent and their end user's behaviour. Our customers are product and growth teams at B2C apps, mostly in fintech, e-commerce, and gaming.

But a nudge is one-way. It can prompt. It can inform. It can't listen.

"I'm using Plotline's nudges. They've helped. But users still have questions I can't anticipate, and when they don't get answers, they leave."

Initial problem space

Users are dropping off from my core journeys. How can I plug this gap?

Who is getting affected

End users - not able to complete key actions and derive value from the product

Product owners are not able to convert their users and plug holes in business growth

Why does it matter

For a user applying for a loan or making a first investment, that wait is the conversion killer.

Dropped-off users tend to raise support tickets leading to overhead on the business team and delayed clarity

// DISCOVERY //

What are some of these core user journeys?

For a finance app (60% of Plotline's user base)

Lending money

Products such as personal, housing, auto loans and credit against investments, P2P lending

Investments

Securities and fund investments

Add/Withdraw money from wallets

Adding money and retrieving money for use across the platform

Why are users dropping off?

To probe further

How is the hidden cost and conditions data currently communicated? Where does the source of truth sit?

The fall-out: What are users doing after dropping off?

How this impacts

Adds to the operational overhead of not only costs but also time it takes to resolve queries and nudge user to resume the journey

To summarise, impactful problem areas are

1. App usage has a learning curve - FAQs and delayed support is inadequate

2. Current In-app interventions are static - one-dimensional. They just communicate what product owners feel the doubt is, not covering every organic query

3. Missed cross-sell opportunities due to delay in capturing user intent

“I am using Plotline's nudges, they are impactful and have led to an uplift of 10-20% in key journey completions, but there is still a lot of information not getting passed through to end users, causing delay in completion”

Rishabh

Growth team, Dream11

Let's take an actual user journey - applying for a loan in a fintech app.

1. User starts the loan application process

2. Has queries about the loan and doesn't build enough confidence

3. Raises support ticket or sales team reaches out - delay of ~1-2 hours on avg for resolution

4. Users drop-off and only those who feel confident enough apply for loan

This same journey with a real-time intent recognition and communication system

1. User starts the loan application process

2. Has queries about the loan and doesn't build enough confidence

3. Discovers missing information through context-aware live conversations

4. User feels confident enough and applies for loan without delay

// APPROACH TO PROBLEM DISCOVERY //

What we didn't know

Before any research, I mapped the specific unknowns we needed to resolve - not methods, actual questions:

Where do drop-offs actually happen, and which of those moments could a conversation genuinely help?
What would make a marketer trust an AI agent enough to deploy it to their users?
What do end users expect from an AI inside a financial app - and where does trust break down?
What does "good" look like for a conversational interaction in a high-stakes context (lending, investments)?

How we found the answers

We ran 1:1 semi-structured interviews with product and growth leads at fintech apps in our customer base. Semi-structured because we wanted to follow threads, we had a guide, but the most valuable findings came from places we didn't expect. We talked to teams at Kredivo, Dream11, and several others across lending, investing, and wallets.

We also did a journey audit, mapped the core user flows (loan application, investment, wallet top-up/withdrawal) against where support tickets were being raised. This gave us a quantitative layer to ground the qualitative interviews.

We didn't do end-user research upfront. That was a deliberate tradeoff given the timeline, we'd validate with real users during the pilot.

// PRIMARY RESEARCH //

Understanding existing workflows, uncovering any latent aspirations and concerns for a conversational AI agent inside their app

User interviews

To understand their challenges better, we started 1:1 semi-structured interviews which gave us insights into user needs and pain points, expected experiences from end users and what is the current experience provided to them.

What are the main customer interactions within your app that could benefit from AI-powered conversations (e.g., customer support, product recommendations, order tracking)?
How do you currently gather customer feedback, troubleshoot issues, and upsell products? Would a conversational agent be suitable for any of these?
How do you estimate the agent's impact in your app? (e.g., multilingual support, personalization, deep product knowledge)?
How important is AI-human handover in complex cases? What is your expectation of bot vs. human interactions?
What are your top concerns about integrating conversational AI agents? (Options: technical complexity, security and privacy, customer trust, handling edge cases, impact on brand, regulatory compliance)

“Over the past 2 years, we have seen that the time taken to resolve support tickets is inversely proportional to lifetime value of our customers"

ALETHIA TAN

SVP, Growth, Kredivo Indonesia

“An agentic experience inside my app should aid the overall discoverabilty and usage. It should intelligently understand when it is needed and what it should help with”

Rishabh

Growth team, Dream11

Key findings from interviews

We expected the dominant concern to be "Will the AI give wrong answers?" That was a concern, but it wasn't the primary one.

Training: "How do I make it know what it needs to know?"

Marketers were anxious about knowledge gaps - stale information, missing context, wrong answers. But what surprised us was how they wanted to solve this. They wanted to teach the system from conversations they'd already had.

This insight directly shaped the benchmark system.

Testing: "How do I know it'll behave the right way before I push it live?"

There was near-universal anxiety about deploying something they couldn't fully preview. Teams wanted to simulate conversations - not just check settings. This wasn't about technical QA. It was about confidence. A marketer needs to be able to say "I have talked to this thing and it makes sense" before they trust it with their users.

Deployment: "What happens when it doesn't know something or gets it wrong?"

The question of escalation - when does the bot hand off to a human, and how - came up in every single interview. More importantly, several people raised brand risk: "If my AI agent says something incorrect about a loan product, I'm liable." This wasn't paranoia. It was valid. It changed how we thought about the autonomy spectrum.

How might we

Design a system that enables marketers and product owners to understand user intent more closely and give generally available information to them in real-time

Why LLMs, not rule-based flows

The first fork in the road was the underlying technology. We considered three approaches:

Decision trees / scripted flows

Deterministic, auditable, but brittle. Every new query type requires a new branch. Can't handle the open-ended, organic questions users actually ask ("wait, how does the lock-in period work if I withdraw early?")

FAQ overlays / static knowledge surfaces

Already existed in some form via Plotline's nudges. The research told us users had already moved past this.

LLM-based responses

Flexible, can synthesise context from multiple sources, can hold a conversation across turns, can adapt to tone and intent.

The risks were real: hallucination, inconsistency, latency, brand voice drift. But these were design problems, not reasons to abandon the approach. Every structural decision in the product - the knowledge base architecture, the benchmark system, the testing simulator, the deployment controls - exists to constrain and direct the LLM, not to replace it.

Configuring an agent - snapshot of the core UX journey

Setting up your agents

Since the whole concept of having an AI agent take care of your users' needs, aspirations and frustrations was new, I built a few pre-configured templates to help the marketers get started and explore in a low friction way.

Measuring success for this flow

It is very important to track the usability of a completely new product added to our core dashboard. Thus, we are closely tracking the performance.

Task success rate - Creation

Currently ~ 57%

Usability support ticket ratio

Currently ~ 45%

Time to first value

Currently ~ 8 minutes

How will I train my system?

Knowledge base

Collection of data points that the agent can retrieve as required such as FAQs, policy documents

Benchmark conversations

Marketers add benchmarks for the agent to learn and give addition feedback in blind testing

A dynamic, highly relevant knowledge base is essential for any agentic system to train and learn.

"Can I tag the sources of information precisely to reduce bloat"

"But, I can instantly recognise an AI system is talking to me"

How will I train my system?
Let marketers simulate real conversations and build trust on the agent

Testing each agent

Here’s a snapshot of how we can simulate the entire conversation experience, rate previous conversations and help the agent learn exactly how it is supposed to communicate with your users.

"How will I simulate my user's conversation"

"Can I test agent's responses at scale"

"What is causing latency in replies"

How will I deploy the system with confidence?
Giving enough context and situation handling directions to the agent

Context broken down into communication styles, conversation guidance & escalation and hand-overs

How will the system detect when to intervene?

Adding the right tools and knowledge bases - can we reduce the cognitive load here?

Setting up for success? How will I define it?

Making sense of it all
Setting up iteration and improvement for agents

Agent performance broken down into actionable intelligence

Started by focusing on core metrics such as conversation volume, goal completion rate, and human handoff rate (when users are escalated to live agents). Real-time conversation logs, knowledge and tools performance also help in targeting the agent better.

Agent performance broken down into actionable intelligence

"I want the system to learn from its mistakes and not repeat them"

Ensuring a system of record for diving deeper

Key learnings & next steps

Key UX decisions and learnings

Conceptualising how to build modular agentic experiences for platforms like Plotline, for marketers from leading consumer apps and visualising experience for their end users was a great opportunity to understand and design for

Building conversational agents in a modular way

Breaking the whole process into functions such as context, knowledge, tools & actions ensured a very gradual learning curve

Segregating what building blocks are global and what are at agent-specific

Centralising appearance, communication and brand guidelines reduces the potential for inconsistent experiences

Building trust and traceability into every AI decision

Breaking down every decision in every interaction solves for trust at a scale of millions
Building unbiased testing and learning flows for agent's training

Improvements & next steps

Working on agentic experiences opens a whole world of possibilities. For the next versions, ideations and concepts have already started!

Building 1:1 testing simulator

Will help simulate conversation with real time performance tracking for different users and use cases

Improving handover flow to manual agents

Giving more visibility into agent handovers and escalations.

Introducing new access points and interactions

Access points like floating buttons, pinned banners, gestures like long hold, bottom swipe can be introduced

Can users talk to my app in real-time and get their doubts and queries resolved? How should we guide users to complete key actions without relying on support?

Conversational AI builder for consumer apps

Conversational AI for consumer apps // Can AI actually help users in-context?

1 Designer, 1 PM, 3 Engg

1 Designer, 1 PM, 3 Engg

4-5 weeks

4-5 weeks

~40%

~40%

~20%

~20%

~15-20%

~15-20%

~30-40%

~30-40%

What does Plotline do? and what it couldn't do yet

What does Plotline do? and what it couldn't do yet

What does Plotline do? and what it couldn't do yet

Initial problem space

Users are dropping off from my core journeys. How can I plug this gap?

Who is getting affected

Who is getting affected

Why does it matter

Why does it matter

// DISCOVERY //

What are some of these core user journeys?

For a finance app (60% of Plotline's user base)

For a finance app (60% of Plotline's user base)

Lending money

Lending money

Investments

Investments

Add/Withdraw money from wallets

Add/Withdraw money from wallets

Why are users dropping off?

To probe further

To probe further

The fall-out: What are users doing after dropping off?

How this impacts

How this impacts

To summarise, impactful problem areas are

1.

1.

App usage has a learning curve - FAQs and delayed support is inadequate

App usage has a learning curve - FAQs and delayed support is inadequate

2.

2.

Current In-app interventions are static - one-dimensional. They just communicate what product owners feel the doubt is, not covering every organic query

Current In-app interventions are static - one-dimensional. They just communicate what product owners feel the doubt is, not covering every organic query

3.

3.

Missed cross-sell opportunities due to delay in capturing user intent

Missed cross-sell opportunities due to delay in capturing user intent

1.

1.

User starts the loan application process

User starts the loan application process

2.

2.

Has queries about the loan and doesn't build enough confidence

Has queries about the loan and doesn't build enough confidence

3.

3.

Raises support ticket or sales team reaches out - delay of ~1-2 hours on avg for resolution

Raises support ticket or sales team reaches out - delay of ~1-2 hours on avg for resolution

4.

4.

Users drop-off and only those who feel confident enough apply for loan

Users drop-off and only those who feel confident enough apply for loan

1.

1.

User starts the loan application process

User starts the loan application process

2.

2.

Has queries about the loan and doesn't build enough confidence

Has queries about the loan and doesn't build enough confidence

3.

3.

Discovers missing information through context-aware live conversations

Conversational AI for consumer apps //
Can AI actually help users in-context?

What does Plotline do?
and what it couldn't do yet

How will I train my system?
Let marketers simulate real conversations and build trust on the agent

How will I train my system?
Let marketers simulate real conversations and build trust on the agent