Beyond Platform ROAS: How Ecommerce Brands Should Measure Omnichannel Growth

A Meta dashboard can say performance is getting worse while the business is still growing.

That sounds contradictory until you look at how ecommerce customers actually buy today.

A shopper might discover a product on Instagram, search for the brand on Google, compare it on Amazon, and finally buy through Blinkit or Swiggy Instamart. The platform that created demand may never record the conversion.

That was one of the central questions raised during AI Native E-Commerce · Edition 01. Leaders from Perfora, Phool, and Swiggy Instamart discussed what happens when awareness starts on one platform and conversion ends somewhere else.

For ecommerce teams, omnichannel measurement is no longer just an analytics problem.

It influences where the next marketing rupee should go.

Platform ROAS Is Only One Part of the Story

For years, performance marketing measurement looked relatively clean:

Meta ad → Shopify → Checkout

That journey still exists. It is simply no longer the whole journey.

As we explored in From Meta to Quick Commerce: The New Ecommerce Customer Journey, discovery and conversion increasingly happen on different platforms. The customer may encounter the brand on Meta but choose a marketplace, quick-commerce platform, or brand website when it is time to buy.

Perfora described this exact behaviour during Edition 01. Meta remained one of its most cost-efficient ways to reach new users, but customers ultimately decided where they wanted to complete the purchase.

The measurement problem is straightforward:

The platform that records the conversion is not always the only platform that influenced it.

A lower Meta ROAS therefore does not automatically mean Meta is creating less value.

This is also where a connected operating model becomes useful. ShopOS brings specialised ecommerce agents and human experts into the same broader environment, spanning creative, performance marketing, and GEO & SEO workflows.

The objective is not another dashboard. It is better context for the decisions teams make from the signals they already have.

Perfora Shows Why Business-Level Metrics Matter

Perfora‘s panel discussion made the cross-channel problem tangible.

The team looks at MER, Blended ROAS, total advertising cost, and channel-level results. But it also watches what happens elsewhere when marketing investment increases.

Are searches increasing on Google?

Is branded demand rising on Amazon?

Are more customers actively searching for the brand on Blinkit?

Jatin shared that Perfora’s branded searches on Blinkit had grown from roughly 15,000 to nearly one lakh over around 18 months, despite limited brand-building spend inside Blinkit itself.

That does not prove every additional search came from Meta.

It does show why performance marketing measurement cannot stop at a platform dashboard.

The better question is:

What happened to total demand and revenue while we were investing?

Which Ecommerce Marketing Metrics Should Teams Track?

No single metric solves omnichannel measurement.

Brands need a small set of platform and business-level signals that can be read together.

Metric	What It Helps Answer
Platform ROAS	How does the platform attribute its own performance?
Blended ROAS / MER	What revenue are we generating relative to total marketing spend?
CAC	What does customer acquisition cost at business level?
Contribution margin	Is growth economically sustainable?
Branded search growth	Is active demand increasing elsewhere?
Incremental revenue	What additional business did marketing create?
Marginal ROI	Is the next unit of spend still productive?

These ecommerce marketing metrics should not replace platform reporting.

They should prevent platform reporting from becoming the entire decision system.

Channel teams can still use platform signals for day-to-day optimisation. Growth leaders can then look at blended revenue, CAC, margin, search movement, and incremental growth to determine whether the wider business is becoming more efficient.

When those two views disagree, that disagreement itself becomes a signal worth investigating.

For ShopOS teams, the same principle extends beyond media. Creative, paid performance, and AI-search visibility are handled by specialised agents within one wider ecommerce system instead of being treated as unrelated AI workflows.

Blended ROAS Gives a Wider Business View

Blended ROAS asks a broader question than platform ROAS:

How much total revenue did the business generate relative to total advertising spend?

It will not tell you which Meta ad caused a specific Blinkit purchase.

Its value is keeping the overall business view in focus.

Suppose Meta-reported ROAS declines, but Amazon revenue grows, Blinkit branded searches rise, total revenue increases, and contribution margin remains healthy.

Cutting Meta immediately could remove demand that is being captured elsewhere.

Perfora described this tension directly. When Meta dashboard ROAS weakens, the team also looks at whether broader business indicators and other channel revenues are holding up before deciding whether to reduce investment.

That is why Blended ROAS works best as part of a broader measurement model rather than as another isolated KPI.

Perfect Attribution Is the Wrong Goal

The panel’s view on attribution was pragmatic:

Attribution will never be perfect.

Customers move between platforms, devices, and purchase environments. Better technology may improve the picture, but reconstructing every journey perfectly is unrealistic.

The more useful question is:

What additional growth did this marketing create?

That moves the conversation from attribution toward incrementality.

How Do You Measure Marketing Incrementality?

Abhishek discussed one practical approach during the panel: reduce or stop activity on selected platforms for a defined period and study what changes.

If a brand believes Meta is creating demand that later converts on Amazon or Blinkit, the business can test that belief rather than endlessly debating attribution.

A controlled test might look at changes in:

• Total sales
• Branded search
• Marketplace and quick-commerce revenue
• New-customer growth
• Contribution margin

If those outcomes weaken meaningfully when activity is removed, the channel may be contributing more value than its own reported conversions suggest.

The panel referred to the idea of going media dark for a defined period to better understand this incrementality.

For ecommerce leaders deciding where to increase or cut spend, this is far more useful than perfect historical credit assignment.

Marginal ROI Tells You Where the Next Rupee Should Go

Historical ROAS tells you what previous spend produced.

Marginal ROI asks:

Where should the next rupee go?

A channel can have strong average ROAS and still be a poor destination for additional budget because returns often diminish as spend rises.

The panel recommended looking at marginal ROI alongside diminishing returns.

Instead of only asking:

Which channel had the best ROAS last month?

Ask:

Which channel can absorb more budget without destroying efficiency?

That makes omnichannel measurement a budget-allocation discipline, not just a reporting framework.

Brand and Performance Marketing Are Converging

The panel also challenged the traditional separation between brand and performance marketing.

Marketers use those labels.

Customers do not.

A shopper may see a Meta ad, remember the brand, search for it later, and convert through a marketplace.

Phool argued that performance communication can itself contribute to mental availability before someone enters a commerce platform.

The journey may look like:

create awareness → strengthen memory → increase branded search → influence a later purchase

without fitting neatly inside one platform funnel.

That is another reason omnichannel measurement needs to look at business movement, not only attributed conversions.

AI Should Mine Signals, Not Just Produce More Ads

One of the strongest AI points from the panel was signal mining.

AI can help teams identify which markets are improving, which cohorts are responding, which experiments are working, and which customer behaviours may indicate new intent.

But there is an important connection to the production side.

In How Ecommerce Brands Can Build a Human-AI Creative Production System, we explored how brands such as The Bear House, Phool, and Perfora are using AI to increase creative production while keeping human direction and review in the workflow.

Once production becomes faster, the bottleneck moves.

The question is no longer only:

Can we make another creative?

It becomes:

Which creative deserves more spend?

That is where performance marketing analytics becomes valuable.

ShopOS supports different parts of that decision loop through specialised agents. Monica handles creative direction and campaign asset workflows, while Gavin works across performance-marketing workflows and paid-channel execution. Big Head handles AI visibility, including prompt and citation tracking across AI search.

The value is not claiming perfect omnichannel attribution.

It is giving teams more connected context for deciding what to scale, pause, investigate, or test next.

More Creative Is Not the Same as More Growth

The panel repeatedly challenged the idea that producing more automatically produces better growth.

When everyone can create hundreds or thousands of assets, marketing judgment becomes more valuable.

That connects back to the first article in this series, When Content Is Easy to Create, Brand Taste Becomes the Advantage.

That article explored why AI-generated content does not remove the need for consumer insight, curation, and a distinct brand point of view.

Measurement adds the commercial layer to that argument.

A team should know:

• Which creative moved demand?
• Which audience deserves more investment?
• Which channel is reaching diminishing returns?
• What should happen next?

Those are growth decisions, not content-volume metrics.

Where ShopOS Fits Into Omnichannel Growth Decisions

The problem described throughout the Edition 01 panel is bigger than reporting.

Creative sits in one workflow.

Paid-media activity sits in another.

Search and discovery happen elsewhere.

Commerce happens across brand.com, marketplaces, and quick commerce.

ShopOS brings Monica, Gavin, Big Head, and human experts into one broader AI-team model for ecommerce brands.

Each agent owns a different function:

Monica → creative

Gavin → performance marketing

Big Head → GEO & SEO

Together, the useful operating loop becomes:

creative → performance → discovery → commerce → next action

The objective is not to pretend every conversion can be perfectly traced.

It is to give ecommerce teams stronger context for deciding where to invest, what to improve, and which signals deserve action.

Conclusion: Better Judgment Beats Perfect Attribution

The Edition 01 panel made the shift clear.

Customer journeys are fragmented. Attribution is incomplete. Marketplaces and quick commerce can capture demand created elsewhere. AI gives teams more signals than they can manually evaluate.

In that environment, platform ROAS alone is not enough.

Brands need omnichannel measurement that combines platform performance with ecommerce marketing metrics, Blended ROAS, incrementality, marginal ROI, branded-search behaviour, and business-level outcomes.

But the competitive advantage is not the metric itself.

It is knowing what to scale, what to stop, where to invest next, and which signals deserve another experiment.

As ecommerce becomes more AI-driven and more fragmented across channels, the brands that grow efficiently will not be the ones with the most dashboards.

They will be the ones that turn better signals into better decisions.

That is the real value of omnichannel measurement: not perfect attribution, but better judgment about what drives growth.

Connect the Signals Behind Ecommerce Growth

Creative, performance marketing, discovery, and commerce should not operate as separate decision-making silos.

See how ShopOS brings specialised ecommerce agents and human experts into one operating environment so teams can turn fragmented signals into better growth decisions.

Book a demo to see ShopOS in action.

Frequently Asked Questions

What is omnichannel measurement in ecommerce?

Omnichannel measurement evaluates marketing and business performance across channels rather than relying on one platform’s attribution. It combines platform data with broader signals such as revenue, CAC, branded search, margin, and incremental growth.

How can ecommerce brands measure performance beyond platform ROAS?

Brands can combine platform ROAS with Blended ROAS, CAC, contribution margin, branded-search growth, total revenue, marginal ROI, and incrementality tests to understand wider business impact.

What is Blended ROAS?

Blended ROAS compares total business revenue with total advertising spend. It provides a broader performance view without trying to assign every individual purchase to a single platform.

How do you measure marketing incrementality?

Brands can use controlled holdouts, geo experiments, media-dark periods, and similar tests to compare business outcomes with and without selected marketing activity.

How can AI improve performance marketing analytics?

AI can help surface patterns across markets, cohorts, experiments, creative performance, search visibility, and customer behaviour so teams can decide what deserves further investment or investigation.

Why is omnichannel measurement important for ecommerce brands?

Customers frequently discover products on one platform and purchase on another. Omnichannel measurement helps teams account for those cross-channel effects and make investment decisions using a wider view of business growth.