slice: Onboarding Teardown

A product teardown of getting a new user from install to a verified account, by Shivam Bajaj. Built only from public sources; I have no access to slice's internal data.

What's in it for slice?

slice began as a card and UPI app and merged with North East Small Finance Bank, effective 27 October 2024 (YourStory; Entrackr). Since then, a new user has to become a bank customer, and that means identity verification before the app can show its value.

  • Drop-off is the main risk. Digital account opening loses a large share of users at KYC; one UX study puts abandonment at up to 60% (Markswebb). Every verified account counts.
  • Trust is built early or not at all. Users are asked for PAN, selfie and video before they have seen the product.

Public information on slice's current flow is thin. Where I describe a step I could not confirm, I label it a hypothesis. Forum posts are individual reports, not measured rates.

Personas

Illustrative archetypes I made up, not real people. Hypotheses are labelled.

The first-jobber

Who: 23, just started a first salaried job in Bengaluru

Goals: A free, simple savings account with good interest

Pain points: Hypothesis: unsure what documents are needed and drops off to find them

Opportunity: A "keep these ready" checklist with time estimate

The existing slice user

Who: Already uses the slice card or UPI app

Goals: Move to the bank account without starting from scratch

Pain points: Post-merger, existing users move to the bank (YourStory); the migration steps are not clearly public

Opportunity: Pre-fill known details and explain what changes

The freelancer

Who: Irregular income, name spelled differently across documents

Goals: Open an account without a branch visit

Pain points: Name errors in CKYC data led to a rejected account in one reported case (DesiDime)

Opportunity: Show fetched name, allow corrections in place

User Journey Map

Reported issues come from forum posts and industry studies. Items marked Hypothesis are my guesses, not observed in the app. The deck I took the format from scores each step with a psych score; I leave that out because I have no data to back numbers.

Download & phoneOTPBasic detailsPAN & CKYC fetchVideo KYCActivation & first use
GoalUnderstand what slice offers and startProve the phone is mineGet through the easy part quicklyVerify identity without paperworkFinish verification in one goSee the account working
FeelingCuriousImpatientComfortableWary of sharing PANSelf-conscious, rushedRelieved, or let down if delayed
Pain pointHypothesis: users may not know the documents needed before startingFew, if any, since OTP is routineHypothesis: long forms before value is shownName mismatches in CKYC data led to a rejected account in one reported case (DesiDime)Selfie and liveness checks are a common source of retries across KYC vendors, and up to 60% abandon digital account opening overall (Markswebb)Some users report accounts showing as not serviceable after CKYC opening (DesiDime)
IdeaChecklist of what to keep ready, with time estimateAuto-read OTP and clear retry timerAsk name and email first, with reasonsShow the fetched name before submitting; let users correct itLighting tips, save-and-resume, booking a slotShow account status and next step immediately

Context: slice has used HyperVerge technology for identity checks (HyperVerge case study, vendor material). Forum users say account opening fetches CKYC data and uses video KYC with PAN, not Aadhaar, for many users (DesiDime).

UX Elements

Illustrative wireframes of proposed screens. They are not screenshots of slice's app, and all copy is placeholder.

Open your accountAbout 5 minutesKeep these ready✓ PAN card✓ Phone linked to Aadhaar✓ Good light for a selfieYour details are encrypted andused only to verify your identity.Start

1. Welcome: documents to keep ready and time needed, before any input

←About youStep 1 of 4Full nameAs on your PANEmailname@email.comWhy we askTo match your name with your PAN.Continue

2. Basic details first, with a progress bar and a reason for each ask

←Verify identityStep 2 of 4Fetch my details (recommended)Uses your PAN. About 1 minute.No documents to upload.Verify with Aadhaar OTPAbout 2 minutes.Needs your Aadhaar-linked phone.If a name mismatch appears, you can fixit here without starting over.Continue

3. KYC choice: time and effort shown, mismatch can be fixed in place

←Video verificationStep 3 of 4Camera previewTips• Face the light • Hold your PAN steady• Stay in frame for 2 minutesNot a good time?Save progress and finish later.Start video

4. Video KYC: tips, camera preview, save and resume

Good & Bad

What works (reported)

  • Fetching data from CKYC and using video KYC means no document upload for many users (DesiDime).
  • Fully digital account opening saves a branch visit, which is the stated promise of the bank.

What hurts (reported or likely)

  • CKYC name errors can cause a rejection with no easy fix (DesiDime).
  • Accounts opened via CKYC showing "not serviceable" in the app (DesiDime).
  • Hypothesis: the checklist of documents is not shown up front.

Suggested Improvements

My suggestions, not claims about slice's roadmap.

1. Tell users what to keep ready

Open with a checklist and a time estimate before any field. Why: Users who leave to find a document often do not return. Risk: A long checklist can scare people off, so keep it to three items.

2. Easy questions first

Ask name and email before PAN and the selfie, with a progress bar and a reason for each ask. Why: Small wins build investment before the sensitive steps. Risk: Longer flow if the extra steps add no information.

3. Show fetched data and allow fixes

After the CKYC fetch, show the name and let the user correct or switch method without restarting. Why: Mismatch rejections are a reported cause of failure. Risk: Needs a safe correction path to satisfy compliance.

4. Save and resume

Keep progress after KYC steps and send a reminder with a deep link to the exact step. Why: Users who drop off mid-way are the cheapest to win back. Risk: Reminders must respect consent and frequency limits.

Key Metrics

North Star

Verified accounts opened per day.

  • Counts only completed, activated accounts.

L1: Funnel completion

% of installs that finish video KYC.

  • Break down by step to find the biggest drop.

L2: Recovery

% of users who leave mid-flow and later finish.

  • Measures save-and-resume and reminders.

Guardrails

  • KYC rejection and manual-review rates, so easier flows do not weaken compliance.
  • Fraud and mismatch rates by method.
  • Time to complete and support contacts per 1,000 onboardings.