Onboarding Experience for Merchants

Disclaimer: Due to NDA's, I will not be able to share any final insights and areas of application and

will be focusing on process & high-level findings + recommendations.

 What is Razorpay?

Razorpay is a payments solution in India that enables businesses to accept, process, and disburse payments through its product suite. It gives access to all payment modes, including credit card, debit card, netbanking, UPI, and popular wallets.

One can manage their marketplace, automate bank transfers, collect recurring payments, share invoices with customers, and avail working capital loans - all from a single platform.

Why study the Onboarding Experience?

Onboarding experience is the first user interaction that any merchant encounters.

As per data analysis, we do have extensive drop-offs in the flow even amongst people who have mid-high intent, owing to various experience-led reasons.

So the Study focuses on uncovering the problem areas and suggests areas of improvement to make the overall complex setup experience seamless

1. Understand and improve our funnel drop-off numbers as mentioned:

   1. 84% drop in Sign Up to MTU Funnel (4.79 to 0.75%) in mKYC flow

   2. Sign Up to KYC Submit conversions dropped by 35% (from 22% to almost 14%)

   3. KYC Submit to Payments Enabled (Activated/AMP) funnel further dropped by 75% (from 62% to 16%)

2. Without understanding why merchants abandon signup, we're losing potential customers to a subpar experience

   1. Competitive positioning of the Onboarding flow

   2. Understand the Design impact contributing to Onboarding drop-offs and subsequently reduce them

3. Support costs increase when merchants struggle during onboarding so fixing experience/information gaps is crucial.

   1. Tickets, Chats, and Calls raised by merchants during KYC completion or post activation till 30 days or 5 settlements contribute to 56.36% of tickets

   
   
   
   

So, the Research Objectives?

1. What makes a business decide to sign up with Razorpay

   1. Intent & Motivation formation

      
      

2. What changes in their thinking, expectations, or effort make them stop the signup process?

   1. Cost vs Benefit changes

   2. Perceived value vs Perceived effort

      
      

3. How much help do businesses expect when signing up, and where do they need it the most? What kind of help do they prefer?

   1. Hand Holding & Emotional Responses

Phase 1

Approach and Methodology

I Passive Primary research - 

Method 1: Focus Group Discussions

Customer support agents Experience - 60 mins

Sampling: 7 participants = 3 Makers, 4 Checkers

Studied the following: 

• Merchant Behaviour responses to Onboarding problems 

• Emotions around Needs Clarification given

• Digital Literacy vs Business Category Differences

Method 2: Stakeholder Discussions

Merchant calls discussion with Agents - 120 mins

Sampling: 7 stakeholders

• 2 Customer Support Leaders (Listening to live/recorded calls followed by discussions),

• 2 Product managers

• 1 Designer

• 2 Risk & Compliance team members.

Studied the following: 

• Merchants lingo on onboarding processes

• Processes outside of the Design flow

• Ideal state of what processes could be part of the form instead with the Support

• Past data experiments to understand onboarding flow drop-offs

Method 3: Clarity Analysis

Observed Screen analysis of Merchants onboarding sessions. 

Date range Filter: 12.1.25- 27.1.25 (master-kyc flow)

Sessions Analysed: 

• 50 master-kyc flows

• 40 Rize modular flows

Deliverables:

i) Existing Pain points and blockers for the Onboarding experience

ii) Hypothesis building for Primary research

The infamous question: "So, Now What...?

• Unclogged: What are stakeholder assumptions vs facts on the hypothesis of user behaviours

• Mapped obvious experience breakers based on clarity analysis for immediate actionables for teams to work on

Learnings and Next Steps: Reality Check!!!!

• Surprisingly, customer support agents had an intuitive understanding of how they estimate user feelings and what kind of people want 'what'. This helped me to get more user context at the grassroots level.

  
  

• To analyse User session recordings without bias, I needed the support of stakeholders to share their analysis of the observations. However, the mobilisation of the common workshop proved to be hard.

---

Phase 2

II Active Primary research - 

Method 1: Open Card Sorting

• 45-minute hybrid card sort.

• Activity: Participants sorted the onboarding steps given as cards into their own categories.

  • Tagging: Each step was also tagged with an emotion, the help needed, and the time required as per the user

• Interview: A follow-up interview was conducted to understand key decisions and reasoning behind the sort style.

Sampling Split

Total sample = 12

• 7 Razorpay Heavy Users 

• 3 Competitors [contrasts]

• 2 Developers [contrasts]

  • 1 In-house Dev

  • 1 Out-sourced Dev

    
    

Participant Profiles

• Owners/Decision makers only 

• 3-6 months of joining the Payment Gateway

• 50:50 SME: MM

Recruitment style

• Hotjar + Cold Calling

• Marketing Emailers

Method 2: In-depth Interviews

• Interview session with open-ended questions and prototype testing

  • Duration: 60-90 mins

Studies the following: 

• Intent to start PG

• Reasons for Drop-offs (value trade-offs), both PGs

  • Scenario-based decision-making

• Hand Holding Expectation

• Modular Design Flow prototype testing

Sampling Split

Total sample = 12

• 6 Competitor Heavy PG users

• 3 In-house developers

• 3 Outsourced developers

Participant Profiles

• Owners/Decision makers only 

• 3-6 months of joining PG 

• 50:50 SME:MM

• Should have experience of using Razorpay PG (churn, backup, dormant <3mts)

Recruitment style

• External Agency - Median Research

Incremental show of cards to identify decision-value changes

---

Top-level Insights/Findings

Insight 1: Intent to join PG

The growth stage of the company determines its motivation or intent to join Razorpay and start the Onboarding process.

• L1 - Survival mode

  • If the company is in Runway mode, speed to onboard is the primary motivator to start the onboarding journey.

• L2 - Security

  • If the company is in Security mode, looking at Razorpay as a backup option to their existing PG, it is sluggish to onboard. Trigger point is “good support” 

• L3 - Belonging

  • If the company is above level 3, it would require customised care to onboard and is highly likely to go via the AM route with custom solutions to be built

The infamous question: "So, now what,...?

[Insights that turned into Business Impact]

1. Problem to Tackle: Emphasising on increasing the speed to start onboarding for first-time transactions

   
   

   Actions --> Impact:

   1. Automated form-filling by enabling auto checks documents in 1 go instead of a progressive form format

   2. Form Design to show a sense of urgency and speed to fill by reducing cognitive load. The prominent side-loading state was improved to have a directly actionable progress bar.

   
   

2. Problem to Tackle: Security-focused merchants need a lot of nudging and proactive support to fasten their onboarding process.

   
   

   Actions --> Impact:

   a. Enabled AI Conversational Onboarding, where a voice agent assists users during the onboarding process, especially on documentation support. People can ask onboarding-related queries directly to the voice agent.

News Coverage

Insight 2: Exploration is Key to High Drop-off Rate

Form lacks a clear roadmap of required info, 1st time onboardees creates dummy accounts to de-risk the commitment and explore the true effort of onboarding

• Perceived Usefulness gauging

  Analyze expectations around business: 

  Merchants view onboarding as a way to achieve business goals (eg: reaching younger audience etc). They explore the form to see if value could be achieved easily or not

• Perceived Effort gauging

  Analyse effort to achieve goal: 

  Explores the form to see what docs/business info is required and would go ahead only if they have all info handy. Otherwise, will make a switch to competitors

The infamous question: "So, now what,...?

[Insights that turned into Business Impact]

1. Problem to Tackle: Having a clear “Exploration” journey for users to set expectations on the form-filling and be assured

Actions --> Impact:

a. Milestones pages through the form + Visual assets on Expectations setting on Onboarding Form to prevent users from going through the form just to understand the process.

Insight 3: Human Support vs Self-Serve

Users prefer self-service for its low mental effort, but a blocker triggers cognitive overload, making the offloading of the problem to a human the most efficient path to relief

• Bots vs Humans

  A merchant’s goal is to reduce cognitive load as fast as possible.

  • A human is seen as capable of absorbing the complexity, whereas a bot is perceived as demanding even more diagnostic effort from an already overloaded user.

• Competence problem

  • Merchants with less competency (less tech savvy), their first instinct is not to trust themselves and rely on someone else’s competence. 

  • They require complete hand-holding to finish the process. 

• Risk Aversion

  • Merchants, when they are faced with confusion in the process, as simple as which PAN to use. 

  • They assume it’s risky to continue (errors can’t be corrected - heuristic mismatch) further without validation/support from Razorpay

The infamous question: "So, now what,...?

[Insights that turned into Business Impact]

1. Problem to Tackle: Improving the quality of Information Assets in the form of comprehensive tutorial assets and making them available on request by the user in the journey

   
   

Actions --> Impact:

a. Design assets to have picture-based explanations, templates of documentation, and motion graphic explanations for tasks like a cancelled cheque. 

1st design cut on assets