Concept: Designed and written by Nitin Bhasin
Concept: Designed and written by Nitin Bhasin
UX Case Study: research, product architecture, and 20 documented design decisions
UX Case Study: research, product architecture, and 20 documented design decisions
Trace
Trace
Trace
Trace
A personal financial intelligence layer for the Indian market
A personal financial intelligence layer for the Indian market
A personal financial intelligence layer for the Indian market
A personal financial intelligence layer for the Indian market
A ground-up UX research and design initiative for a personal financial intelligence layer, covering individual users, families, and certified financial planners across India's Account Aggregator ecosystem.
A ground-up UX research and design initiative for a personal financial intelligence layer, covering individual users, families, and certified financial planners across India's Account Aggregator ecosystem.
A ground-up UX research and design initiative for a personal financial intelligence layer, covering individual users, families, and certified financial planners across India's Account Aggregator ecosystem.
Act 1: The Problem World
Act 1: The Problem World
Act 1: The Problem World
Act 1: The Problem World
The Hook
The Hook
The Hook
The Hook
I have four payment apps, three bank accounts, and four credit cards across Visa, Mastercard, and RuPay. By most definitions, I'd call myself financially literate. And yet, more than once, I've had to open every single one of them just to answer a question that should take five seconds: can I afford this? If I can't answer that confidently, with everything I supposedly know about money, I started wondering how anyone else does. That question became Trace.
I have four payment apps, three bank accounts, and four credit cards across Visa, Mastercard, and RuPay. By most definitions, I'd call myself financially literate. And yet, more than once, I've had to open every single one of them just to answer a question that should take five seconds: can I afford this? If I can't answer that confidently, with everything I supposedly know about money, I started wondering how anyone else does. That question became Trace.
I have four payment apps, three bank accounts, and four credit cards across Visa, Mastercard, and RuPay. By most definitions, I'd call myself financially literate. And yet, more than once, I've had to open every single one of them just to answer a question that should take five seconds: can I afford this? If I can't answer that confidently, with everything I supposedly know about money, I started wondering how anyone else does. That question became Trace.
Product Vision
Product Vision
Product Vision
Product Vision
Before the research and the twenty decisions, one sentence anchors all of them:
What Trace is for, stated as plainly as the product itself tries to be.
Before the research and the twenty decisions, one sentence anchors all of them:
What Trace is for, stated as plainly as the product itself tries to be.
Before the research and the twenty decisions, one sentence anchors all of them:
What Trace is for, stated as plainly as the product itself tries to be.
Act 2: Understanding the User
Act 2: Understanding the User
Act 2: Understanding the User
Act 2: Understanding the User
Research Approach
Research Approach
Research Approach
Research Approach
8 in-depth interviews were conducted across deliberately different financial life stages: a salaried senior product manager, a cash-heavy business owner, a high-earning program manager in a multi-generational household, and a pre-earning college student. Each interview anchored on one question: do you feel in control of your financial future, and why or why not? The goal was not statistical significance; it was to map the distinct shapes that "not in control" takes.
8 in-depth interviews were conducted across deliberately different financial life stages: a salaried senior product manager, a cash-heavy business owner, a high-earning program manager in a multi-generational household, and a pre-earning college student. Each interview anchored on one question: do you feel in control of your financial future, and why or why not? The goal was not statistical significance; it was to map the distinct shapes that "not in control" takes.
8 in-depth interviews were conducted across deliberately different financial life stages: a salaried senior product manager, a cash-heavy business owner, a high-earning program manager in a multi-generational household, and a pre-earning college student. Each interview anchored on one question: do you feel in control of your financial future, and why or why not? The goal was not statistical significance; it was to map the distinct shapes that "not in control" takes.
Personas
Personas
Personas
Personas
Each persona below captures four things: how they actually spend, the behavior pattern that defines their relationship with money, how accurately they assess their own financial awareness, and the specific thing that would unlock them. The gap between believed awareness and actual awareness turned out to be the single most consistent finding: three of four participants were confident they knew their finances, and all three were measurably wrong.
Each persona below captures four things: how they actually spend, the behavior pattern that defines their relationship with money, how accurately they assess their own financial awareness, and the specific thing that would unlock them. The gap between believed awareness and actual awareness turned out to be the single most consistent finding: three of four participants were confident they knew their finances, and all three were measurably wrong.
Each persona below captures four things: how they actually spend, the behavior pattern that defines their relationship with money, how accurately they assess their own financial awareness, and the specific thing that would unlock them. The gap between believed awareness and actual awareness turned out to be the single most consistent finding: three of four participants were confident they knew their finances, and all three were measurably wrong.
Journey Map
Journey Map
Journey Map
Journey Map
Mapping the emotional arc behind the reframe in Act 1, from impulse to reinforced trust:
Mapping the emotional arc behind the reframe in Act 1, from impulse to reinforced trust:
Mapping the emotional arc behind the reframe in Act 1, from impulse to reinforced trust:
The six-stage journey every persona described in some form, and where Trace intervenes.
The six-stage journey every persona described in some form, and where Trace intervenes.
The six-stage journey every persona described in some form, and where Trace intervenes.
Cross-User Patterns
Cross-User Patterns
Cross-User Patterns
Cross-User Patterns
• Insurance authority concentrates upward. The primary earner manages and pays for dependents' insurance across every household type, including for pre-earning dependents. This produced the per-member insurance admin view (Decision 17).
• Insurance authority concentrates upward. The primary earner manages and pays for dependents' insurance across every household type, including for pre-earning dependents. This produced the per-member insurance admin view (Decision 17).
• Insurance authority concentrates upward. The primary earner manages and pays for dependents' insurance across every household type, including for pre-earning dependents. This produced the per-member insurance admin view (Decision 17).
• Gold is a universal blind spot with two distinct shapes. Category exclusion (Aayush, Priyanka: gold not counted as wealth at all) versus valuation blindness (Mukesh: counted, but worth unknown) are different failure modes needing different fixes (Decision 19).
• Gold is a universal blind spot with two distinct shapes. Category exclusion (Aayush, Priyanka: gold not counted as wealth at all) versus valuation blindness (Mukesh: counted, but worth unknown) are different failure modes needing different fixes (Decision 19).
• Gold is a universal blind spot with two distinct shapes. Category exclusion (Aayush, Priyanka: gold not counted as wealth at all) versus valuation blindness (Mukesh: counted, but worth unknown) are different failure modes needing different fixes (Decision 19).
• CFP receptivity is consistently positive but not uniform in motivation. Status-seeking, second-opinion-seeking, and aggregation-gap-closing all look like "interest in a CFP" but respond to different pitches (Decisions 15 through 17).
• CFP receptivity is consistently positive but not uniform in motivation. Status-seeking, second-opinion-seeking, and aggregation-gap-closing all look like "interest in a CFP" but respond to different pitches (Decisions 15 through 17).
• CFP receptivity is consistently positive but not uniform in motivation. Status-seeking, second-opinion-seeking, and aggregation-gap-closing all look like "interest in a CFP" but respond to different pitches (Decisions 15 through 17).
Competitive Landscape
Competitive Landscape
Competitive Landscape
Competitive Landscape
A 22-player audit was conducted in June 2026 spanning aggregators, neobanks, credit platforms, lending apps, and investment platforms. The matrix below shows the eight capabilities that define Trace's territory across the seven closest players, plus three shaded rows for capabilities Trace deliberately does not build. The read-only boundary is a strategic choice, not a missing feature: every player that moves money inherits a structural conflict of interest about what to show the user.
A 22-player audit was conducted in June 2026 spanning aggregators, neobanks, credit platforms, lending apps, and investment platforms. The matrix below shows the eight capabilities that define Trace's territory across the seven closest players, plus three shaded rows for capabilities Trace deliberately does not build. The read-only boundary is a strategic choice, not a missing feature: every player that moves money inherits a structural conflict of interest about what to show the user.
A 29-player audit was conducted in June 2026 spanning aggregators, neobanks, credit platforms, lending apps, and investment platforms. The matrix below shows the eight capabilities that define Trace's territory across the seven closest players, plus three shaded rows for capabilities Trace deliberately does not build. The read-only boundary is a strategic choice, not a missing feature: every player that moves money inherits a structural conflict of interest about what to show the user.
Capability matrix. Shaded rows are ceded territory; the read-only boundary is the moat.
Capability matrix. Shaded rows are ceded territory; the read-only boundary is the moat.
Capability matrix. Shaded rows are ceded territory; the read-only boundary is the moat.
Three findings that shaped the product
Three findings that shaped the product
Three findings that shaped the product
Three findings that shaped the product
• Single-partner aggregation dies with the partner. Fi Money shut down banking services in March 2026 after Federal Bank ended the partnership. Lesson: aggregation built on a single banking partnership is structurally fragile; Trace connects via the RBI Account Aggregator rail, which no single institution can revoke.
• Single-partner aggregation dies with the partner. Fi Money shut down banking services in March 2026 after Federal Bank ended the partnership. Lesson: aggregation built on a single banking partnership is structurally fragile; Trace connects via the RBI Account Aggregator rail, which no single institution can revoke.
• Single-partner aggregation dies with the partner. Fi Money shut down banking services in March 2026 after Federal Bank ended the partnership. Lesson: aggregation built on a single banking partnership is structurally fragile; Trace connects via the RBI Account Aggregator rail, which no single institution can revoke.
• The AA rail is being used against the user. Lending platforms like Zype already use the AA framework, but only to verify a borrower's income faster for the lender. No player uses it to show the borrower their own total debt exposure. That inversion, same rail, opposite beneficiary, is Trace's core positioning.
• The AA rail is being used against the user. Lending platforms like Zype already use the AA framework, but only to verify a borrower's income faster for the lender. No player uses it to show the borrower their own total debt exposure. That inversion, same rail, opposite beneficiary, is Trace's core positioning.
• The AA rail is being used against the user. Lending platforms like Zype already use the AA framework, but only to verify a borrower's income faster for the lender. No player uses it to show the borrower their own total debt exposure. That inversion, same rail, opposite beneficiary, is Trace's core positioning.
• Everyone reacts; nobody intervenes. Qubera, the closest conceptual competitor, is reactive: it answers questions on demand. INDmoney's nudges read as alerts. Nobody intervenes before a discretionary purchase. The Pause occupies that empty moment.
• Everyone reacts; nobody intervenes. Qubera, the closest conceptual competitor, is reactive: it answers questions on demand. INDmoney's nudges read as alerts. Nobody intervenes before a discretionary purchase. The Pause occupies that empty moment.
• Everyone reacts; nobody intervenes. Qubera, the closest conceptual competitor, is reactive: it answers questions on demand. INDmoney's nudges read as alerts. Nobody intervenes before a discretionary purchase. The Pause occupies that empty moment.
DETAILED AUDIT STRUCTURE (FULL 29-PLAYER APPENDIX)
DETAILED AUDIT STRUCTURE (FULL 29-PLAYER APPENDIX)
DETAILED AUDIT STRUCTURE (FULL 29-PLAYER APPENDIX)
The complete audit is maintained as a separate appendix, one page per player, structured as: category and business model; data connectivity method (AA, email parsing, scraping, manual); monetization mechanics and where the conflict of interest sits; nudge philosophy (reactive, alert-based, or interventionist); and what Trace borrows or explicitly rejects from each. Players covered: Qubera, INDmoney, CRED, Fi Money, Jupiter, Zype, ET Money, Paytm, PhonePe, Google Pay, Groww, Zerodha, Angel One, Bajaj Finserv, Fibe, LazyPay, OneScore, CheQ, BharatNXT, Walnut legacy, Axio, and MoneyView.
The complete audit is maintained as a separate appendix, one page per player, structured as: category and business model; data connectivity method (AA, email parsing, scraping, manual); monetization mechanics and where the conflict of interest sits; nudge philosophy (reactive, alert-based, or interventionist); and what Trace borrows or explicitly rejects from each. Players covered: Qubera, INDmoney, CRED, Fi Money, Jupiter, Zype, ET Money, Paytm, PhonePe, Google Pay, Groww, Zerodha, Angel One, Bajaj Finserv, Fibe, LazyPay, OneScore, CheQ, BharatNXT, Walnut legacy, Axio, and MoneyView.
The complete audit is maintained as a separate appendix, one page per player, structured as: category and business model; data connectivity method (AA, email parsing, scraping, manual); monetization mechanics and where the conflict of interest sits; nudge philosophy (reactive, alert-based, or interventionist); and what Trace borrows or explicitly rejects from each. Players covered: Qubera, INDmoney, CRED, Fi Money, Jupiter, Zype, ET Money, Paytm, PhonePe, Google Pay, Groww, Zerodha, Angel One, Bajaj Finserv, Fibe, LazyPay, OneScore, CheQ, BharatNXT, Walnut legacy, Axio, and MoneyView.
Scope of the research and build that Act 3 and Act 4 unpack in detail.
Scope of the research and build that Act 3 and Act 4 unpack in detail.
Scope of the research and build that Act 3 and Act 4 unpack in detail.
Act 3: Product Architecture
Act 3: Product Architecture
Act 3: Product Architecture
Act 3: Product Architecture
Before individual decisions, the two structural views that hold the product together: how a user moves through Trace, and how all the screens hang off four questions.
Before individual decisions, the two structural views that hold the product together: how a user moves through Trace, and how all the screens hang off four questions.
Before individual decisions, the two structural views that hold the product together: how a user moves through Trace, and how all the screens hang off four questions.
How firsthand friction became a 20-decision, multi screen prototype.
How firsthand friction became a 20-decision, multi screen prototype.
How firsthand friction became a 20-decision, multi screen prototype.
Core User Flow
Core User Flow
Core User Flow
Core User Flow
Three loops matter. The onboarding path is engineered to end in an insight, never an empty state. The daily loop is the behavioral core: it is where the Pause, the Score, and the Reflection compound into the feeling of control. The value branches are the only places money changes hands, and each one is reached from inside the daily loop rather than pushed.
Three loops matter. The onboarding path is engineered to end in an insight, never an empty state. The daily loop is the behavioral core: it is where the Pause, the Score, and the Reflection compound into the feeling of control. The value branches are the only places money changes hands, and each one is reached from inside the daily loop rather than pushed.
Three loops matter. The onboarding path is engineered to end in an insight, never an empty state. The daily loop is the behavioral core: it is where the Pause, the Score, and the Reflection compound into the feeling of control. The value branches are the only places money changes hands, and each one is reached from inside the daily loop rather than pushed.
Onboarding path, daily behavioral loop, and the three value branches.
Onboarding path, daily behavioral loop, and the three value branches.
Onboarding path, daily behavioral loop, and the three value branches.
Onboarding path, daily behavioral loop, and the three value branches.
Onboarding path, daily behavioral loop, and the three value branches.
Onboarding path, daily behavioral loop, and the three value branches.
Information Architecture
Information Architecture
Information Architecture
Information Architecture
Trace is organized by time horizon and trust boundary, not by feature. Home answers "how am I doing" across Now, Month, and Future. Sources answers "what does Trace see and how," with per-category effort labeling (instant AA, email OAuth, manual). Family and Profile hold the two trust boundaries: who inside the household sees what, and who outside it does. The CFP portal is deliberately a separate surface: an advisor never touches the consumer app's data plane.
Trace is organized by time horizon and trust boundary, not by feature. Home answers "how am I doing" across Now, Month, and Future. Sources answers "what does Trace see and how," with per-category effort labeling (instant AA, email OAuth, manual). Family and Profile hold the two trust boundaries: who inside the household sees what, and who outside it does. The CFP portal is deliberately a separate surface: an advisor never touches the consumer app's data plane.
Trace is organized by time horizon and trust boundary, not by feature. Home answers "how am I doing" across Now, Month, and Future. Sources answers "what does Trace see and how," with per-category effort labeling (instant AA, email OAuth, manual). Family and Profile hold the two trust boundaries: who inside the household sees what, and who outside it does. The CFP portal is deliberately a separate surface: an advisor never touches the consumer app's data plane.
Onboarding path, daily behavioral loop, and the three value branches.
Onboarding path, daily behavioral loop, and the three value branches.
Onboarding path, daily behavioral loop, and the three value branches.
Four pillars, one separate advisor surface, across screens.
Four pillars, one separate advisor surface, across screens.
Four pillars, one separate advisor surface, across screens.
Act 4: Twenty Design Decisions
Act 4: Twenty Design Decisions
Act 4: Twenty Design Decisions
Act 4: Twenty Design Decisions
Every product choice is documented as tension, decision, and hypothesis. No decision enters the prototype without a log entry. The first eight define the product's shape; nine through fourteen extend it into connectivity and monetization; fifteen through twenty were produced directly by user research.
Every product choice is documented as tension, decision, and hypothesis. No decision enters the prototype without a log entry. The first eight define the product's shape; nine through fourteen extend it into connectivity and monetization; fifteen through twenty were produced directly by user research.
Every product choice is documented as tension, decision, and hypothesis. No decision enters the prototype without a log entry. The first eight define the product's shape; nine through fourteen extend it into connectivity and monetization; fifteen through twenty were produced directly by user research.
Decision 1: Read-only vs. transactional
Decision 1: Read-only vs. transactional
Decision 1: Read-only vs. transactional
TENSION
TENSION
TENSION
Should Trace move money, or only observe it?
Should Trace move money, or only observe it?
Should Trace move money, or only observe it?
DECISION
DECISION
DECISION
Read-only core with a deliberate handoff layer at the point of action. Trace never executes a transaction itself.
Read-only core with a deliberate handoff layer at the point of action. Trace never executes a transaction itself.
Read-only core with a deliberate handoff layer at the point of action. Trace never executes a transaction itself.
HYPOTHESIS
HYPOTHESIS
HYPOTHESIS
Keeping the product free of transactional conflict of interest lets monetization stay honest: referral fees only when a recommendation genuinely helps the user.
Keeping the product free of transactional conflict of interest lets monetization stay honest: referral fees only when a recommendation genuinely helps the user.
Keeping the product free of transactional conflict of interest lets monetization stay honest: referral fees only when a recommendation genuinely helps the user.
Home / Now: the read-only daily view.
Home / Now: the read-only daily view.
Home / Now: the read-only daily view.
Decision 2: Completeness vs. confidence (the cash gap)
Decision 2: Completeness vs. confidence (the cash gap)
Decision 2: Completeness vs. confidence (the cash gap)
TENSION
TENSION
TENSION
Cash and informal spend can't be tracked with certainty. Hide the gap, disclaim it, or show it honestly?
Cash and informal spend can't be tracked with certainty. Hide the gap, disclaim it, or show it honestly?
Cash and informal spend can't be tracked with certainty. Hide the gap, disclaim it, or show it honestly?
DECISION
DECISION
DECISION
A two-tier data confidence system: Confirmed (Account Aggregator or parsed email) versus Estimated (inferred from ATM withdrawal patterns). Carried through surface weight rather than hue, so the single-accent rule holds.
A two-tier data confidence system: Confirmed (Account Aggregator or parsed email) versus Estimated (inferred from ATM withdrawal patterns). Carried through surface weight rather than hue, so the single-accent rule holds.
A two-tier data confidence system: Confirmed (Account Aggregator or parsed email) versus Estimated (inferred from ATM withdrawal patterns). Carried through surface weight rather than hue, so the single-accent rule holds.
HYPOTHESIS
HYPOTHESIS
HYPOTHESIS
Users trust a system more when it's honest about the limits of what it knows than one that presents a polished number quietly built on guesses.
Users trust a system more when it's honest about the limits of what it knows than one that presents a polished number quietly built on guesses.
Users trust a system more when it's honest about the limits of what it knows than one that presents a polished number quietly built on guesses.
Month view: confirmed and estimated spend, separated at a glance.
Month view: confirmed and estimated spend, separated at a glance.
Month view: confirmed and estimated spend, separated at a glance.
Decision 3: Nudge timing: proactive vs. reactive
Decision 3: Nudge timing: proactive vs. reactive
Decision 3: Nudge timing: proactive vs. reactive
TENSION
TENSION
TENSION
When should Trace speak up, before spend, during it, or after?
When should Trace speak up, before spend, during it, or after?
When should Trace speak up, before spend, during it, or after?
DECISION
DECISION
DECISION
A three-moment model.
• The Pause: a calm, single data point before a discretionary purchase.
• The Signal: a forward-looking pattern warning mid-month.
• The Reflection: a warm check-in after an unusual transaction, always ending in a question.
A three-moment model.
• The Pause: a calm, single data point before a discretionary purchase.
• The Signal: a forward-looking pattern warning mid-month.
• The Reflection: a warm check-in after an unusual transaction, always ending in a question.
A three-moment model.
• The Pause: a calm, single data point before a discretionary purchase.
• The Signal: a forward-looking pattern warning mid-month.
• The Reflection: a warm check-in after an unusual transaction, always ending in a question.
HYPOTHESIS
HYPOTHESIS
HYPOTHESIS
Timing the intervention before the moment of regret, not after it, is what separates coaching from an alert feed.
Timing the intervention before the moment of regret, not after it, is what separates coaching from an alert feed.
Timing the intervention before the moment of regret, not after it, is what separates coaching from an alert feed.
Month view: confirmed and estimated spend, separated at a glance.
The Pause, rendered: overspend context plus goal impact, then a binary call.
Month view: confirmed and estimated spend, separated at a glance.
Month view: confirmed and estimated spend, separated at a glance.
Decision 4: The decisiveness layer
Decision 4: The decisiveness layer
Decision 4: The decisiveness layer
TENSION
TENSION
TENSION
How do you reward good financial behavior without infantilizing a professional audience or shaming a bad month?
How do you reward good financial behavior without infantilizing a professional audience or shaming a bad month?
How do you reward good financial behavior without infantilizing a professional audience or shaming a bad month?
DECISION
DECISION
DECISION
The Trace Score: a private, self-comparative 0 to 100 score, never compared against other users. Pause decisions add two points each; a streak multiplies gains; pattern divergence holds the score flat rather than dropping it, a deliberate no-shame-mechanics rule.
The Trace Score: a private, self-comparative 0 to 100 score, never compared against other users. Pause decisions add two points each; a streak multiplies gains; pattern divergence holds the score flat rather than dropping it, a deliberate no-shame-mechanics rule.
HYPOTHESIS
HYPOTHESIS
HYPOTHESIS
Behavioral-science framing (aligned, diverging) rather than game language (points, badges) keeps it a serious tool for a financially literate audience.
Behavioral-science framing (aligned, diverging) rather than game language (points, badges) keeps it a serious tool for a financially literate audience.
Behavioral-science framing (aligned, diverging) rather than game language (points, badges) keeps it a serious tool for a financially literate audience.
Month view: confirmed and estimated spend, separated at a glance.
The Pause, rendered: overspend context plus goal impact, then a binary call.
Month view: confirmed and estimated spend, separated at a glance.
Month view: confirmed and estimated spend, separated at a glance.