The Two Questions Ecommerce Brands Never Ask About Data

How Core Competency Thinking Frees Fast Growing Brands From The Infrastructure Trap

The story of two companies that reflects the why some e-commerce brands become thriving businesses while their counterparts fail to survive.

One asks: "What businesses should we be in?"

The other asks: "What should we be uniquely great at?"

That single question would determine which company would soar and which would stumble. E commerce brands face the same fork on the road with data. Most don't even realize the choice exists.

And it traces back to two questions that most e-commerce brands missed asking.

The Questions Nobody Asks

Walk into any fast growing ecommerce brand and ask about their data strategy. You'll hear about dashboards, data analysts, maybe a business intelligence tool.

But ask these two questions, and you'll get blank stares:

Question 1: "What is your core data competency as an ecommerce brand?"

Most founders have never articulated this. They've confused "having data" with "having data capability." They've hired smart analysts and assumed that it covers it.

If pressed, the real answer should be: "Understanding our customers and optimizing our business for profit."

That's it. That's the core competency.

Not building databases. Not maintaining ETL pipelines. Not debugging dashboard infrastructure.

Understanding which products drive lifetime value. Knowing which customer segments are most profitable. Optimizing contribution margin across 200 SKUs. Making faster, better decisions.

That's the competency that makes an ecommerce brand successful.

Question 2: "Are we organizing our data capabilities around our core competency, or around people?"

This is where it gets uncomfortable.

Most ecommerce brands have organized data around a person, their "data analyst" and expected that person to:

• Build and maintain all infrastructure (engineering)

• Answer ad hoc questions from every department (support)

• Perform strategic analysis (the actual analyst work)

• Document everything (knowledge management)

That's not a job. That's four jobs.

And here's the trap: When you organize around a person instead of around a competency, you've created a single point of failure that looks like a strategic asset.

Until that person gives notice.

How A 1982 Decision Explains Today's Crisis

In 1982, two telecommunications giants faced identical challenges. Both wanted to dominate the emerging information technology industry.

GTE asked: "What businesses should we be in?" They organized into business units. Telecommunications. Lighting. Defense. Each unit operated independently, hoarding resources and capabilities.

NEC asked: "What should we be uniquely great at?" They identified core competencies semiconductor technology, digital integration, systems coordination and made these capabilities accessible across the entire organization.

Eight years later, NEC was a top five global player in telecommunications, semiconductors, and mainframes. GTE had become... a phone company.

By 2000, GTE was acquired and absorbed into Verizon. NEC had defined an industry era.

The difference wasn't intelligence or capital. It was one question about organizational philosophy.

In 1990, C.K. Prahalad and Gary Hamel published what became Harvard Business Review's best selling article ever, analyzing exactly what happened. They called it "Core Competency Theory."

The insight: Companies must organize around what they're uniquely great at, not around siloed units that imprison capabilities.

GTE's business units became kingdoms. When the telecommunications division needed semiconductor expertise, they couldn't easily access it from other divisions. Each unit tried to own its entire value chain. The result? Mediocrity everywhere, excellence nowhere.

NEC's competencies were shared resources. Every division could tap into the company's collective expertise. When they entered new markets, they had 80% of what they needed through existing competencies.

Thirty five years later, ecommerce brands are making GTE's mistake with data. They just don't realize it yet.

The Solo Owner Model vs. The Shared Capability Model

Here's what organizing data around people instead of competencies looks like:

Month 1-6: The Heroic Phase

You hire a brilliant data analyst. They start building from scratch. Shopify data flows into a database. Dashboards appear. The CEO is thrilled. "We're finally becoming data driven!"

Month 7-12: The Maintenance Phase

The infrastructure exists but needs constant care. APIs change. Dashboards break. Data doesn't reconcile between systems. Your analyst now spends 60% of time maintaining what they built, 30% answering ad hoc questions ("Why did sales drop last week?"), and 10% on actual strategic analysis.

The CEO starts asking: "Why does it take three weeks to get a contribution margin analysis?"

Month 13-18: The Bottleneck Phase

Everything data related flows through one person. Marketing needs campaign attribution. Finance needs revenue reconciliation. Operations needs inventory forecasting. Your analyst is drowning.

No one else can help because no one understands the systems. They were custom built by one person, for one person.

Another challenge that is touched upon later is that a solo analyst cannot both solve technical issues to maintain APIs and understand business requirements. They are just too varied skills for one person to have. Even within "technical" issues, building or fixing efficient pipelines is significantly different from building intuitive dashboards. It is impossible for one person to have both skills. non-tech founders tend to undermine this.

Month 19+: The Breaking Point

Your analyst burns out. Or gets recruited away. Usually both.

Everything breaks. You spend 6 12 months rebuilding from scratch.

This is exactly what happened with GTE's business unit model. Each unit became a single point of failure, hoarding knowledge and capabilities. When key people left, entire capabilities vanished.

Ecommerce brands do the same with data treating their analyst like a self contained business unit and hoping it all works out.

This is what we'll call The Solo Owner Model: organizing around individuals who own entire domains.

The Three Tests: Is Data Analysis Really Your Core Competency?

Before we discuss organizational models, let's apply Prahalad and Hamel's three definitive tests to determine if something is truly a core competency:

Test 1: Does it provide access to multiple markets or opportunities?

For ecommerce brands, strategic data analysis opens doors to:

• New customer segments (identifying high LTV cohorts)

• New product categories (understanding cross sell patterns)

• New distribution channels (data driven expansion decisions)

• New geographies (market specific insights)

It also helps in understanding the current business performance. It helps in understanding what to fix before it snowballs into a major threat or running the current business efficiently which gets highlighted in dashboards.

✓ Data analysis passes this test. It unlocks multiple growth avenues.

Test 2: Does it make a significant contribution to perceived customer value?

Strategic data analysis directly drives:

• Better product recommendations (personalization)

• Optimal inventory (fewer stockouts, faster shipping)

• Fair pricing (contribution margin optimization)

• Product quality improvements (analyzing return patterns)

✓ Data analysis passes this test. It directly improves what customers experience.

Test 3: Is it difficult for competitors to imitate?

Your specific understanding of YOUR customers, YOUR products, YOUR market is impossible to replicate. But data infrastructure? That's table stakes every ecommerce brand can buy the same tools, use the same platforms, hire similar engineers.

Before you take a decision to hire a new ‘Head of Data’ or ‘CTO’ who would come in and try and build sophisticated/complicated/expensive infrastructure in-house you need assess your core competency. Setting up the team and tech for data may not be your core competency. You should rather outsource it and spend more time with business teams to understand business data requirements and leverage data for insights and problem solving. That adds to core competency.

✓ Data analysis (insight) is hard to imitate. Data infrastructure is not.

The verdict: Strategic data analysis and business insight IS your core competency. Data infrastructure is not it's critical enabling technology, but not core.

This distinction changes everything.

What Core Competency Thinking Actually Means For Data

Let's be explicit about the difference.

The Solo Owner Model (What Most Brands Do):

• Hire a data analyst as a self contained "data department"

• Expect them to build and maintain all infrastructure

• Hope they also have time for strategic analysis

• Treat data infrastructure as something you "own" through a person

• Cross your fingers that person never leaves

The Shared Capability Model (What Winners Do):

• Identify your core data competency: business insight and decision making about products, customers, and profitability

• Recognize that data infrastructure is critical but not core like AWS for servers or Shopify for ecommerce platform

• Partner with specialists for infrastructure: pipelines, connectors, data warehouses, maintained platforms

• Let your internal analyst focus 80%+ of time on analysis, not maintenance

• Build institutional knowledge that survives turnover

Your core competency as an ecommerce brand is building great products and distributing them profitably. Not managing data infrastructure.

Just like:

• Your core competency isn't managing servers (that's why you use Shopify or AWS)

• Your core competency isn't payment processing (that's why you use Stripe)

• Your core competency isn't shipping logistics (that's why you use ShipBob or a 3PL)

• Your core competency isn't email infrastructure (that's why you use Klaviyo)

So why would your core competency be building and maintaining data infrastructure?

It's not. Your core competency is using data to make better decisions about your products and customers.

Infrastructure is the means. Business insight is the competency.

The Questions You Can't Answer (And What They're Costing You)

Here's the test of whether you've organized around competency or around people.

Try answering these five questions right now:

1. What's your contribution margin by product after factoring in actual shipping costs, discount rates, and return rates?

2. What's the LTV difference between customers acquired through Meta vs. Google vs. email?

3. Which products have negative true margins once you include payment processing and returns?

4. What's your repeat purchase rate by cohort, and how is it trending month over month?

5. Which products, when purchased first, drive the highest customer lifetime value?

If you can't answer at least three of these in under 60 seconds, you have an infrastructure problem disguised as an analyst problem.

These aren't nice to know metrics. These are the questions that determine:

• Where would you allocate $500K+ in annual ad spend?

• Which of your 200 SKUs is destroying value?

• Is your discount strategy eroding the margin on products that would sell anyway?

• Which customer segments drive profitability?

• How should you sequence new product launches?

For a $30M brand, getting these wrong costs $500K $1M annually in suboptimal decisions.

Now here's what actually happens when you ask your solo analyst:

You: "What's our true contribution margin by product?"

Your analyst: "I'm working on it. The dashboard is 80% done, but I'm still debugging why return data from Loop Returns isn't syncing correctly with Shopify refunds. Also, the shipping cost allocation logic breaks for bundle orders. Give me another two weeks."

Translation: "I'm spending 90% of my time on infrastructure, not on answering the business question."

This isn't the analyst's fault. This is an organizational design problem.

You've organized around a person, not around competencies. You've asked them to be both the infrastructure engineer AND the strategic analyst. That's like asking your head of product to also manage your AWS servers.

They could probably do it. But should they?

The Real Cost: A $40M Brand's Blind Spot

Let me show you exactly what the wrong organizational model costs.

Scenario: $40M ecommerce brand with 180 SKUs

Your analyst has spent 4 months building a contribution margin dashboard. Here's where the time went:

• Month 1: Building connections to 7 data sources (Shopify, ShipBob, Stripe, Loop Returns, Meta Ads, Google Ads, inventory system)

• Month 2: Debugging why revenue totals don't match between Shopify and Stripe, building reconciliation logic

• Month 3: Figuring out how to properly attribute returns to original orders and handle bundle SKUs

• Month 4: Optimizing queries because the dashboard keeps timing out with 2 years of data

Result after 4 months: Dashboard "mostly works" but has edge cases, performance issues, and requires weekly maintenance.

The strategic insight work "Which products should we promote aggressively and which should we kill?" still hasn't started.

Meanwhile, during those 4 months, you've almost certainly been, but at the end of 4 months, there could be a newer problem statement that you need to solve for.

• Spending 20% of ad budget on low/negative margin products: $200K+ wasted annually

• Carrying 30 40 SKUs that destroy value: $100K+ in holding costs and clearance losses

• Offering discounts on high margin items that would sell anyway: $150K+ in unnecessary margin erosion

• Missing upsell opportunities from high margin products: $250K+ in lost revenue

Conservative annual cost: $700K+ in profit sitting on the table.

All because your analyst spent 4 months building infrastructure instead of 2 weeks analyzing clean, reliable data.

This is what Prahalad and Hamel called "imprisoning capabilities."

Your analyst's “strategic thinking ability to find insights that drive $700K in profit improvement” is imprisoned by infrastructure work that should be handled by specialists.

What Winners Do Differently: The Partnership Model

The pattern among successful ecommerce brands is clear: They separate infrastructure from competency. They separate infrastructure from competency. They know that the secret to speed and scalability lies in the right partnerships.

Rather than spending engineering bandwidth on maintaining fragile data connectors, brands like Ridge, Momentous partner with Saras Pulse to automate data movement and performance tracking.

When Mamaearth expanded across multiple online marketplaces in 2023, their key challenge was fragmented reporting. Instead of building custom scripts in-house, they integrated Saras Pulse. Overnight, sales and campaign data from Amazon, Flipkart, Nykaa, and their own web store were unified into a single analytics view. The internal marketing team didn’t spend months cleaning CSVs or debugging APIs, they spent that time optimizing campaigns, improving conversion rates, and analyzing purchase patterns to identify their highest LTV customers.

This mirrors how Glossier worked with Fivetran to automate their data infrastructure. In both cases, specialized partners managed the plumbing, while internal teams focused on driving growth through insights.

Similarly, Saras Pulse supported the omnichannel transformation of Borosil, integrating POS data from over 100 retail touchpoints with online performance dashboards. The result was a clear view of how local in-store promotions influenced online sales—a finding that redefined their product placement and pricing strategies. Like Warby Parker’s discovery that hybrid customers had 70% higher lifetime value, Borosil’s data-driven insights reshaped their omnichannel investments.

The pattern is consistent across winners:

• Partner for infrastructure: data pipelines, connectors, warehouses, and maintenance.

• Own the analysis: strategy, insight generation, and decision-making.

• Free your teams to do what makes your business unique: creating great products and growing customer value.

This is the Shared Capability Model in practice here pooled data expertise fuels sharper, faster business decisions.

The Infrastructure vs. Competency Test

Here's the simplest way to know if you've organized correctly:

Would you let your data analyst manage your AWS servers?

No. That's infrastructure work requiring specialized expertise. You partner with AWS because:

• They're specialists in cloud infrastructure

• They maintain and optimize continuously

• They handle security and compliance

• They scale as you grow

• Your engineering team can build products instead of managing servers

So why do you let your data analyst build and maintain data infrastructure from scratch?

The competencies required are completely different:

Data Infrastructure (Engineering):

• Database architecture and optimization

• ETL pipeline reliability and monitoring

• Data modeling for performance at scale

• Security, governance, compliance

• Platform maintenance and upgrades

Data Analysis (Strategy):

• Business question framing

• Statistical reasoning about customer behavior

• Contribution margin optimization

• Product and customer cohort analysis

• Communicating insights that drive executive decisions

One person can learn both. But they can't excel at both simultaneously. And they definitely can't do both while also answering ad hoc questions from every department.

This is the core competency distinction: Infrastructure is critical but not core. Analysis and decision making is core.

You partner for the first to excel at the second.

The $600K Annual Cost of the Wrong Question

Let's return to the brand that lost their data analyst.

Here's what organizing around a person instead of competencies costs them:

Immediate costs:

• 6 months to rebuild infrastructure: $60K in lost analyst productivity

• 3 4 months for new analyst to ramp: $40K+

• Opportunity cost of decisions without data: $200K+ (conservative)

• Total immediate impact: $300K+

Ongoing annual costs:

• Analyst spending 60% of time on maintenance vs. analysis: $70K+ in misallocated talent

• Slower time to insight on strategic questions: $300K+ in suboptimal decisions

• Reduced agility to test and optimize: Incalculable

• Risk of this repeating when next analyst leaves: All of above, again

• Total ongoing annual cost: $600K+

All because they never asked: "What's our core data competency?"

If they'd organized around competencies partnering for infrastructure, focusing internal talent on analysis the analyst's departure would have been a minor setback, not a crisis.

The infrastructure would have remained stable and maintained. The institutional knowledge would live in documented systems, not one person's head. The new analyst could contribute value in week one, not month six.

Most importantly: The first analyst would never have been buried in infrastructure work. They would have spent 4 years doing high value analysis, not maintenance.

What This Means For Your Business

Let's be direct about what core competency thinking means for your ecommerce brand's data strategy.

Your core competency is NOT:

• Building Postgres databases

• Maintaining ETL pipelines

• Debugging Shopify API connections

• Optimizing data warehouse queries

• Managing dashboard infrastructure

Your core competency is:

• Understanding which products and customers drive profitability

• Optimizing contribution margin across your catalog

• Making faster, better decisions about inventory, marketing, and product development

• Using data to build competitive advantage in YOUR market

• Scaling your business profitably through insight driven decisions

The first list is infrastructure. Critical, but not core.

The second list is competency. Core to your success as an ecommerce business.

When you organize around competency:

✓ Strategic questions get answered in days, not months

✓ You make better annual decisions about margin, customers, and products  

✓ Your team focuses on what makes YOUR business successful, not on maintaining pipes

✓ When people leave, capabilities remain because they live in systems, not heads  

✓ You can scale analytical capacity by hiring analysts who contribute immediately

This is the Shared Capability Model: pooled infrastructure expertise, focused business insight.

When you organize around people:

✗ Brilliant analysts spend 60 80% of time on maintenance

✗ Strategic questions take weeks or months while infrastructure gets debugged  

✗ Single points of failure disguise themselves as strategic assets

✗ Turnover means starting over from scratch  

✗ Your core competency business insight is imprisoned by infrastructure work

This is the Solo Owner Model: individuals trying to own entire value chains, capabilities imprisoned in isolated domains.

The Transformation: What Changes When You Shift Models

Here's what changes when ecommerce brands move from the Solo Owner Model to the Shared Capability Model:

ROI Comparison (Including Data Partner Costs)

1. Annual Cost Structure

Insight:

Even with partner costs, the Shared Model does not cost more.
In many brands, it is equal or cheaper, because the solo model burns most of its value in maintenance.

2. Annual Strategic Output Value

3. Net ROI After Costs

The throughput transformation is dramatic:

With the Solo Owner Model, a $120K/year analyst delivers roughly $50-100K in strategic value because only 10% of their time goes to high impact analysis.

With the Shared Capability Model, that same $120K/year analyst delivers $300-500K+ in strategic value because 70% of their time goes to high impact work. Add $60-80K in partnership costs, and you still get 3-4x the return on investment.

But the real transformation isn't just throughput it's capability:

• Speed: Strategic questions answered in days instead of weeks

• Quality: Reliable infrastructure means trust in data, faster decisions

• Agility: Test and optimize in real time, not quarterly

• Resilience: Business doesn't halt when people leave

• Growth: Scale analytical capacity without rebuilding

The math is clear: focusing on core competency while leveraging partner expertise doesn't just improve efficiency; it transforms what's possible.

The Fork in the Road

Every fast-growing ecommerce brand reaches this moment.

You're at $10M, $30M, or $50M in revenue. You know you need data capabilities. You've hired a smart analyst or are about to.

Here's where the fork appears:

Path 1: The Solo Owner Model

Hire a data analyst. Expect them to build everything. Hope they can also deliver insights. Cross your fingers, they don't leave. When they do, you have to start over.

This path organizes around siloed ownership; each person tries to own their entire domain.

Path 2: The Shared Capability Model

Define your core data competency: business insight and profitable decision making. Recognize that infrastructure is critical but not core just like your servers, payment processing, and logistics. Partner with specialists for infrastructure. Let your analysts spend 80%+ of time on what makes your business succeed: understanding customers, optimizing products, driving profitability.

This path organizes around shared capabilities infrastructure expertise is pooled, business insight is focused.

The evidence from 35 years of business strategy research is unambiguous: The Shared Capability Model organizing around core competencies and partnering for critical but not core infrastructure outperforms the Solo Owner Model of siloed domains.

The evidence from ecommerce is equally clear: Brands that separate infrastructure from competency make better, faster decisions and scale more sustainably.

The Question That Determines Everything

So here it is, the question for your ecommerce brand:

Are you organizing your data capabilities around your core competency business insight or around a person you're hoping can do everything?

If you're asking one analyst to build infrastructure, maintain pipelines, answer ad hoc questions, document systems, AND deliver strategic insights that drive millions in value...

You're not building a data capability. You're building a single point of failure.

Your core competency is building great products and distributing them profitably.

Data infrastructure? Critical, but not core. Partner for it.

Data analysis and business insight? Core to success. Own it, focus on it, excel at it.

The question isn't whether your analyst can build the infrastructure.

The question is: Should they?

Because every hour they spend on infrastructure is an hour they're not spending on the insights that could drive $500K in annual profit improvement.

Thirty five years ago, NEC and GTE answered different versions of this question.

One organized around core competencies. One organized around business units.

The results weren't even close.

Today, ecommerce brands face the same choice with data.

Which path are you choosing?

The brands that figure this out don't just get better data. They get better businesses.

Stop overloading your team and start unlocking the insights that drive real growth. Let our experts help you focus on what matters—building a profitable, insight-driven business. Talk to our data consultants today and take the first step toward smarter decisions.