Research

& Validation

PepsiConnect · ASIA & EUROPE · Dec 2025 - On going…

India was PepsiConnect's first WhatsApp-first market. No app. No precedent. Just a BRD, a tight deadline, and a lot of unknowns. This covers how research shaped the solution — from in-store visits to a 5-day solo design sprint during Christmas break.

Field research to FullStory analytics. Store visits, user interviews, digital questionnaires, shadowing sessions, post-launch funnels and heatmaps — across India and every live market I own.

Research Type

Field Research · Product Analytics

Research Type

EMEA · Saudi · Egypt · UEA

Field Research

India · 03 Phases

Field Research

DO · PM · CE · Prototype

Product Analytics

16+ Live Markets

Product Analytics

16+ Live Markets

Deliverables

Research Report · Empathy Map · Designs

Deliverables

Saudi Arabia · Egypt

What this covers

Taking PepsiConnect into India — a market like no other

01 Field research conducted in India across 3 phases — qualitative interviews, quantitative surveys, and product shadowing. Understanding real users before a single screen went live.

02 A 5-day solo design sprint during Christmas break. No leadership, no team. End-to-end WhatsApp ordering designs delivered to keep the project moving.

03 Research findings directly shaped design decisions — from channel strategy validation to UX trade-offs made under constraint, all documented and presented to leadership.

04 Post-launch behavioral analytics across 16+ live markets using FullStory — session replays, rage clicks, conversion funnels, and heatmaps to understand real usage patterns.

05 Findings fed back into the product cycle — identifying where users drop off, what they ignore, and what needs fixing. Data driving decisions, not opinions.

01 Field research conducted in India across 3 phases — qualitative interviews, quantitative surveys, and product shadowing. Understanding real users before a single screen went live.

02 A 5-day solo design sprint during Christmas break. No leadership, no team. End-to-end WhatsApp ordering designs delivered to keep the project moving.

03 Research findings directly shaped design decisions — from channel strategy validation to UX trade-offs made under constraint, all documented and presented to leadership.

04 Post-launch behavioral analytics across 16+ live markets using FullStory — session replays, rage clicks, conversion funnels, and heatmaps to understand real usage patterns.

05 Findings fed back into the product cycle — identifying where users drop off, what they ignore, and what needs fixing. Data driving decisions, not opinions.

Research Context

How it all started — and India came into the picture

01 ——

The brief landed

A BRD arrived for India — PepsiConnect's first WhatsApp-first market. No app. No precedent. Store owners ordering through phone calls and sales reps for years. The ask: design a digital ordering experience that fits into how they already live and work.

02 ——

Tight timeline, no blocker

Christmas break. Leadership and the team were off. The project timeline couldn't wait — other teams were depending on it. Stepped in to keep things moving. Started with Meta guidelines, platform constraints, and our own system capabilities — understanding what was even possible before touching Figma.

03 ——

Structure before screens

Mapped every user case and split the solution into two tracks — PM for onboarding and registration, DO for the ordering experience — divided into three stages: start, intermediate, and end state, with a dedicated section for edge cases. Collaborated with a team member to define the bot tone and language. Core flows and user journey built in Figma simultaneously.

04 ——

Designed with AI, validated by instinct

Used AI to refine UX copy, explore design directions, and run A/B testing to pressure-test decisions. Worked through it day and night. The output wasn't perfect — it was a strong, informed foundation the team could move forward with confidently.

05 ——

Time to go to the ground

But building on a BRD alone meant assumptions were still untested. Collaborated with the sales team to plan store visits across India. Shadowed distributors and store owners in their actual environment. What came back shaped the design in ways no BRD could have anticipated.

few Screens

Initial WhatsApp design

Market Research

Two phases. Real stores. Real users.

01 —

Explore & Observe

6-7 stores. Printed forms. No rigid agenda. Visited small and traditional trade stores. No rigid agenda — watched live ordering conversations between distributors and store owners before asking a single question. What we found: - WhatsApp exists but as a backup, not primary - Phone calls dominate — fast, human, trusted - Past app experiences left real damage — stockouts, broken trust - Ideal user flow unprompted: voice note / list photo → result → confirm → done - Biggest risk: bot failure = instant drop-off. Users won't try twice.

02 —

Structure & Validate

20-25 stores. Digital questionnaire. Leadership-ready data. Returned with a structured digital form — ordering ownership, trust levels, channel preference, app barriers, and what would actually make them switch. What it confirmed: - Sales reps own the process end to end - Digital adoption isn't an access problem — most have smartphones - Real blockers: stock reliability, delivery transparency, margin visibility - Sales rep involvement in early onboarding is non-negotiable

03 —

Outcome

Two phases didn't guarantee a solution — they reduced the guesswork. Key risks surfaced early shaped every design decision. Budget constraints meant Meta catalogue templates weren't viable — so the solution became an input-based WhatsApp flow as MVP, with a browser companion accessible at every stage. Research-informed, not assumption-driven. Whether it succeeds gets answered post-launch.

03 —

Outcome

Two phases didn't guarantee a solution: - they reduced the guesswork. - Key risks surfaced early shaped every design decision. - Budget constraints meant Meta catalogue templates weren't viable — so the solution became an input-based WhatsApp flow as MVP, with a browser companion accessible at every stage. - Research-informed, not assumption-driven. Whether it succeeds gets answered post-launch.

Cinematic

Grid

No items

Nutty Delight Marketing Campaign Thumbnail

SunnyCo Summer Essentials Campaign Thumbnail

Refreshingly Effortless Campaign Thumbnail

No items

Product Analysis

16+ markets. Every interaction tracked. Nothing assumed.

FullStory runs across every live market the team owns — Poland, Turkey, New Zealand, Thailand, Mexico, Egypt, Saudi Arabia, and other POD markets. For each market, dedicated dashboards are built with custom funnels and events tailored to key user flows, tracking what actually happens post-launch. Session replays show where users hesitate. Heatmaps show what gets ignored. Funnels show exactly where they drop off. All of it feeds back into the next design decision — not as a one-time audit, but as an ongoing practice every sprint cycle. Data over opinion, every time.

Learnings

What stayed with me.

01 Designing without talking to users is educated guessing. Ground reality and digital data don't always match — and assuming they do is where products quietly fail. Meeting users in their actual environment changes the quality of every decision that follows.

02 Ownership under pressure builds instinct. Working solo, under a live deadline, with no precedent or safety net — you stop second-guessing and start trusting your judgement. That muscle only develops when there's no one else to defer to.

03 Research doesn't slow design down. It stops you from building the wrong thing fast. Having both field truth and post-launch behavioral data makes every design argument stronger and every product decision more defensible.

04 Delivery isn't visibility. Executing quietly isn't the same as leading. Strong output means nothing if the decisions behind it aren't legible to the people who need to trust them.

01 Designing without talking to users is educated guessing. Ground reality and digital data don't always match — and assuming they do is where products quietly fail. Meeting users in their actual environment changes the quality of every decision that follows.

02 Ownership under pressure builds instinct. Working solo, under a live deadline, with no precedent or safety net — you stop second-guessing and start trusting your judgement. That muscle only develops when there's no one else to defer to.

03 Research doesn't slow design down. It stops you from building the wrong thing fast. Having both field truth and post-launch behavioral data makes every design argument stronger and every product decision more defensible.

04 Delivery isn't visibility. Executing quietly isn't the same as leading. Strong output means nothing if the decisions behind it aren't legible to the people who need to trust them.

Recognitions

The work didn't go unnoticed

Open to Opportunities

Somewhere between the hero and here, you decided I might be worth it. I think so too.

Get in touch

Open to Opportunities

Somewhere between the hero and here, you decided I might be worth it. I think so too.

Get in touch

Open to Opportunities

Somewhere between the hero and here, you decided I might be worth it. I think so too.

Get in touch