Case Study · FinTech · 2023

DSP Asset Managers

SARTHICASE STUDY

An AI-guided fund discovery tool that helped first-time investors find the right mutual fund for their goals — in under 5 minutes, without needing any prior investment knowledge.

Client

DSP Asset Managers

Platform

Mobile & Web

My Role

Lead UX & UI Designer

Year

2023

Scroll to explore

65%

Faster to decide
40 min → 14 min

6.75×

More first-time investors
4% → 27%

−48%

Fewer people leaving
before picking a fund

82%

Investors understood
their own risk level

The Problem

People wanted to invest.
The app made it too hard.

DSP's platform was built for experienced investors who already knew what they wanted. But most new users — salaried professionals, first-time SIP starters — had no idea where to begin. The platform didn't meet them where they were.

😰

What users experienced

They opened the app and saw dozens of funds listed by Sharpe ratios, NAV charts, and expense ratios — numbers that meant nothing to them. There was no "I'm new here" path. The app assumed you already knew what you wanted.

📉

What the business saw

Nearly half of all new users were leaving within 3–5 minutes — not at checkout, but at the very first step of browsing funds. Only 4 in 100 new users made their first investment. Every rupee spent on marketing was being wasted.

Design Challenge

How do we help someone who has never invested before find the right fund for their goal — without needing to know anything about finance first?

What We Found

The numbers were clear.
Users weren't lost — they were stuck.

We looked at 6 months of data from 40,000+ sessions, spoke to 12 real users in depth, and studied how other investing apps around the world handle first-timers — including Groww, ET Money, and Betterment.

63%

of users on the fund listing page left without tapping a single fund

48%

of new users gave up within the first 3–5 minutes — before even using a filter

completion rate — only 4 in 100 new users made their first investment

of new users tried the search bar — most didn't even know what to search for

4 Things We Learned — That Shaped Every Decision

Each insight wasn't just a finding — it became a design requirement.

Start with the goal, not the product

Every user thought about investing in terms of a life goal — a home, retirement, their child's education. None of them thought in fund categories. So we had to start with their goal, not DSP's catalogue.

Risk needs to make sense, not just exist

Telling someone their "risk score is 4/7" means nothing. Users needed to understand risk in plain terms — what it means for their money, not just a label on a scale.

Explain the recommendation, don't just show it

Users wouldn't commit to a fund unless they understood why it was right for them. The "why" had to be shown right next to the recommendation — not in a separate FAQ buried elsewhere.

3 options beats 7 every time

In early concept testing, showing 3 relevant funds converted at nearly twice the rate of showing 7. Too many choices created the same paralysis we were trying to solve.

The Experience — Welcome & Onboarding

Before Sarthi says anything,
it asks what you need.

The very first screen after login doesn't push users into a fund list. It asks a simple question: "Where do you want to start?" — giving users a choice between getting fund recommendations or learning how to invest first. This one decision removes the assumption that everyone arrives ready to invest.

Step-by-Step: From Welcome to First Question

The flow is designed so that no screen demands financial knowledge. Users are guided from "where do I start?" to "here's your recommendation" entirely on their own terms.

Welcome screen
"Where do you want to start?"

Sarthi landing — where strategy meets investments

Sarthi intro
"Get personalised recommendations in 5 min"

Q1/10 — Dependents
Human photo, plain language

Q4/10 — Market knowledge
"I do not understand" is a valid answer

Q8/10 — Education
Progress bar keeps users moving

🚪

Choice before commitment

The welcome screen offers two paths — "understand your risk profile" or "learn how to invest." This is the most important UX decision in the whole flow. It removes the assumption that everyone is ready to invest, and immediately reduces bounce from users who feel unprepared.

📸

Contextual photos, not generic icons

Each question uses a full-bleed photo that reflects the topic — a family for dependents, a classroom for education. This makes abstract financial questions feel personal and relatable, reducing cognitive load before users even read the question.

📊

"Questions: 1/10" — transparency builds trust

Showing exactly how many questions remain was one of the biggest drop-off fixes in testing. Without it, users abandoned mid-flow because they didn't know if it would ever end. The progress bar transforms an unknown commitment into a manageable one.

Design Process — Risk Score

The screen we redesigned
three times.

The risk score reveal was the most iterated screen in the project. Getting it wrong meant users either didn't understand their profile or felt judged by it. Getting it right meant 82% of users walked away knowing exactly what kind of investor they were.

Risk score screen — Moderately High, score 65

Version 1 ✗ — Didn't work

A bar on a scale

A horizontal spectrum with a marker. Users anchored on the label ("Moderately High") without understanding what it actually meant for their money. It felt like a verdict handed down, not something they could act on.

Your Risk Profile

Moderately High

ConservativeModerateAggressive

Risk Score

65/100

Based on your answers

Version 2 — Getting closer

Visual metaphor (sea conditions)

Calm sea for conservative, choppy for moderate, stormy for aggressive. Users started saying "that makes sense for me" — but they still lacked a concrete financial expectation to anchor on.

Your Risk Profile

Moderately High

Choppy waters ahead 🌊

Your risk level means

Some ups and downs expected

Version 3 ✓ — Final

Arc gauge + plain-language summary

A clean gauge showing score 65/100, a bold label "Moderately High," and a "What is it?" link for users who want more. No jargon visible by default. The re-assess link gives users back control — they're not trapped by the algorithm.

↑ 82% of users understood their risk profile post-launch

Key UX principle: DSP's brand voice says "invest better" — not "let us decide for you." The risk screen honours that by always showing the re-assess option. Trust is built by giving control back, not by hiding it.

The Solution — Asset Allocation & Fund Picks

From risk profile
to a real investment plan.

The last two screens turn the user's risk profile into something they can actually act on. Every number shown traces back to their own answers — making the recommendation feel earned, not arbitrary.

Personalised fund recommendations screen

UX Decision 01

Show the split before the funds

Seeing "45% Equity / 30% Debt / 25% Liquid" before individual fund names gives users a mental model. They understand the logic of the recommendation before they see the specific products. This one screen reduced drop-off at the recommendation stage in testing.

UX Decision 02

Enter your amount first, then see the plan

Users pick their investment type and enter their amount before funds are shown. So instead of abstract fund names, they see "Invest ₹45,000 in DSP Healthcare Fund." Concrete rupee amounts turn a financial product into a personal decision.

UX Decision 03

Group funds by what they do, not what they're called

Funds are shown under "Equity funds", "Debt fund", "Liquid funds" — not by fund name or return percentage. This reinforces the asset split the user just saw and makes the list feel like a natural consequence of their profile, not a random selection.

UX Decision 04

FAQs at every moment of doubt

"What is the basis of recommendations?" and "What is risk score?" are shown inline on both screens — right where users need reassurance, not in a help centre they'd never find. Inline education, not a detour.

Results — 3 Months After Launch

It worked better
than the target.

Measured via platform analytics, post-session surveys, and a follow-up usability study with 6 new participants. The target was to reach 15% first-time conversion. We hit 27%.

was 35–40 min

14 min

Average time to pick a fund

was 4%

27%

New users who made their first investment

was 48% leaving early

−28%

Fewer people abandoning during fund browsing

post-launch survey

82%

Users who understood their risk level

54%

Started their first SIP without needing outside help

6.75×

Improvement in conversions vs the starting baseline

91%

Task completion rate in post-launch usability testing

+14

Point improvement in platform NPS score

"I've been meaning to start investing for two years. Every time I opened the app it felt like a subject I'd failed at. This time it just asked me what I wanted — and told me what to do. That was it."

— Priya, 28, Mumbai · Post-launch user interview

Looking Back

Things I learned
along the way.

Sarthi hit its targets and then some. But every project teaches you something — here's what I'd carry into the next one.

🧪

Validate structure with users, not just logic

The 10-question flow was shaped by internal alignment and worked well — but post-launch surveys surfaced an opportunity to tighten it further. I learned that even a well-reasoned structure benefits from a length A/B earlier in the process, before it's baked in.

🔄

Scope for the return journey from the start

The v1 focus was rightly on first-time users — that's where the biggest problem was. But it taught me to think about return journeys earlier in the design process, so the next phase has a natural foundation to build on rather than starting fresh.

⚖️

Bring cross-functional partners in earlier

Working closely with the compliance team in the later stages taught me just how much nuance sits at the intersection of UX and regulation. I'd now pull that conversation forward into research — not because it slows things down, but because it makes the design sharper from day one.

What Comes Next

Sarthi solved acquisition.
Retention is the next problem.

The guided questionnaire brought new investors in. The next version should keep them — by tracking progress toward their goal, adapting to what they already know, and evolving into a long-term investment partner, not just a one-time guide.

🎯 Adjust question depth based on how much the user already knows

📈 Show users how their investment is tracking against their original goal

🔁 Smarter recommendations for returning users who already have funds

💬 Explore a chat-style version for mobile — more natural, less form-like