Platform Strategy · 1→N

Turning Fragmented Storefronts into a Reusable D2C Platform

As Rebel Foods expanded its D2C footprint, every new storefront risked becoming another independent engineering project. Catalogue, cart, checkout, discounts, payments and tracking repeated across brands, but improvements in one stack did not compound across the portfolio.

I helped shape a modular, multi-tenant platform that standardized repeated commerce capabilities while preserving brand identity and merchandising flexibility. The platform supported approximately 2M MAUs and 450–500K monthly orders, enabled the Faasos app to launch in roughly one week, and became a major enabler of India’s D2C mix increasing from 8% to 21%.

I helped lead the product strategy, platform scope and phased adoption approach across brand, engineering, growth, operations and leadership teams.

8% → 21%

India D2C web and app mix

~2M

MAUs supported

~450–500K

Monthly orders

~1 week

Faasos app launch

Standardise what repeats. Preserve what differentiates.

01 · The Repeated Constraint

Repeated engineering was becoming the bottleneck.

Every brand needed differentiation, but rebuilding the same commerce stack independently destroyed speed, consistency and leverage.

01

Repeated commerce capabilities rebuilt per brand

02

Improvements did not propagate across storefronts

03

Experiences, analytics and release cycles diverged

04

Launches and migrations became slower

05

Engineering effort was spent, not compounded

Duplicated engineering and QA effortInconsistent customer experienceFragmented analyticsSlow brand launchesHigher maintenance burdenLimited platform leverage

Decision Sequence

Business Strategy · Increase direct-channel leverage
Highest Constraint · Repeated storefront engineering
Dependency · Shared commerce foundations
Proof · Migrate one meaningful brand safely
Scale · Expand only after guardrails held

The constraint was not the number of features we lacked. It was that the same capabilities had to be rebuilt repeatedly.

02 · Strategic Choice

The platform decision was a balance between reuse and autonomy.

How do we standardise what repeats without flattening what differentiates?

Option 01

Continue brand-specific stacks

Advantage: Maximum local control and minimal immediate migration risk.

Why rejected: Repeated work, inconsistent foundations and slow rollout remained structural problems.

Option 02

Build one monolithic shared storefront

Advantage: Maximum standardisation and simpler shared ownership.

Why rejected: Brand differentiation became expensive and local exceptions risked turning the platform into a central bottleneck.

Option 03 · Chosen

Build a modular multi-tenant platform

Advantage: Shared commerce infrastructure with brand identity, merchandising and commercial configuration separated into configurable layers.

Risk: A platform only creates leverage if brands trust and adopt it.

Standardisation without autonomy would have created a different scaling problem.

03 · Platform Boundary

Shared where repetition created leverage. Configurable where brands competed.

Brand teams were not resisting reuse of payments, checkout or tracking. They were resisting the loss of control over identity, merchandising and customer experience. Making that boundary explicit changed the adoption conversation.

Layer 01A · Shared Commerce Core

CatalogueSearch and navigationCartCheckoutDiscountsPaymentsOrder tracking

Layer 02B · Shared Operating Foundation

AnalyticsSupport integrationsReliabilityPerformance foundations

Layer 03C · Configurable Brand Layer

Identity and themeHomepage compositionMenu and category presentationMerchandisingCampaignsPricing rulesCommercial configurationBrand-specific content

A modular model allowed shared capabilities to evolve once while keeping brand differentiation configuration-led rather than engineering-led.

04 · Prove Before Scaling

Trust the platform with real revenue before asking others to adopt it.

The Behrouz migration became the operational proof that the platform could support a meaningful brand without compromising revenue or customer experience.

A · Why Behrouz

Behrouz had meaningful D2C traffic, strong brand identity and enough scale to expose migration, conversion and performance risks.

B · Controlled rollout

Small initial cohort
Progressive exposure
~3 months of validation
Seasonality considered
Expand only after stable results

C · Business + CX Guardrails

Revenue impactOpen-to-order conversionEngagementReliability and performanceCustomer-experience / support signals

The rollout was governed by predefined business and CX thresholds rather than technical completion alone.

D · Proof of Leverage

~1 week

Faasos app launch

Once the shared components were stable, the Faasos app launched in roughly one week — evidence that platform reuse had converted into materially faster execution.

05 · Adoption

The hardest platform problem was organisational, not architectural.

I did not try to win adoption through architecture arguments. The platform had to prove that reuse improved business outcomes without removing brand control.

Concern 01

Brand identity and roadmap control

Response

Separate shared infrastructure from configurable brand experience.

Concern 02

Revenue and conversion risk

Response

Use phased rollout, pre-agreed thresholds and progressive migration.

Concern 03

Cross-team ownership

Response

Create shared dashboards, common metrics and explicit migration accountability.

Concern 04

Delayed platform value

Response

Translate the platform into business outcomes — launch speed, reliability, customer ownership, first-party data and margin control.

Proof before persuasionShared dashboardsKPI guardrailsBusiness narrative

Standardisation

Creates reuse and consistency.

Autonomy

Protects brand differentiation and local commercial control.

The platform had to create leverage without becoming a centralized bottleneck.

06 · Outcomes

Platform leverage became visible at multiple levels.

Group A · Platform Leverage

~1 week

Faasos app launch

  • ·Shared storefront and commerce modules
  • ·Reusable infrastructure across brands
  • ·Common reliability, analytics and experimentation foundations

Group B · Scale Supported

~2M

MAUs supported

~450–500K

Monthly orders supported

Group C · Business Outcome Enabled

8% → 21%

India D2C web + app mix

  • ·Stronger first-party customer ownership
  • ·Faster launch of new brand experiences
  • ·Foundation for a broader partner-brand D2C model

Group D · Operating-Model Change

  • ·Shared health and rollout guardrails
  • ·Common dashboards and metrics
  • ·Explicit migration accountability
  • ·Configuration-led differentiation
  • ·Evidence-based platform adoption

The D2C mix shift was a company-level outcome influenced by multiple product, growth, brand and operational initiatives. The platform was a major enabler, not the sole cause.

Before → After

Brand stacks

FragmentedShared modular platform

Feature development

Repeated per brandReusable capabilities

Differentiation

Engineering-ledConfiguration-led

Platform health

Local analytics / reliabilityCommon guardrails

07 · Reflection

The biggest challenge was not architectural. It was organisational.

Platforms create leverage only when teams adopt them because business outcomes improve, not because the architecture is cleaner.

01 · What worked

Separating shared infrastructure from brand identity made the platform strategically useful without making brands feel interchangeable.

02 · What I would change

I would define the platform operating model earlier — decision rights, migration ownership, adoption metrics and a cross-team review cadence — rather than focusing first on architecture and delivery.

03 · What success should have included from day one

Platform success should have been measured through reuse, adoption, launch speed, reliability and incremental business leverage, not migration completion alone.

04 · What I took forward

Platform work compounds only when architecture, incentives and operating ownership reinforce one another.

I did not prioritise features. I prioritised the repeated constraint.

Next case study

Rebuilding Foundit's discovery-to-apply engine