Omnichannel Commerce: Why the Channel Map Isn’t the Strategy

Omnichannel commerce is the practice of selling across multiple channels, physical and digital, in a way that creates a consistent and connected experience for the customer at every touchpoint. Done well, it means a customer can discover a product on social, research it on your website, try it in-store, and complete the purchase wherever is most convenient, with no friction between any of those steps. Done poorly, it means you have five channels that each behave like separate businesses and wonder why conversion is soft.

Most brands are doing the latter and calling it the former.

I spent a significant part of my agency career managing performance budgets across retail, fashion, FMCG, and financial services clients. Early on, I was guilty of what most performance marketers are guilty of: I overvalued what I could measure. If a channel had a clean last-click attribution story, it looked like a hero. If it did not, it looked like waste. That framing shaped budget decisions that, in hindsight, were systematically wrong. The channels we defunded were often the ones doing the heaviest lifting at the top of the funnel. The channels we scaled were capturing demand that was already there, not creating new demand. Understanding that distinction changed how I think about omnichannel entirely.

What Most Brands Get Wrong About Omnichannel

The most common mistake is treating omnichannel as a channel expansion problem rather than a commercial integration problem. A brand adds Instagram Shopping, launches a mobile app, opens a few pop-up stores, and announces it has gone omnichannel. What it has actually done is added complexity without adding coherence.

True omnichannel commerce requires three things to be in place simultaneously: a unified view of the customer, a consistent commercial proposition across channels, and internal structures that allow channels to work together rather than compete. Most brands have none of these fully operational. Many have fragments of the first, a loose approximation of the second, and almost none of the third.

The internal structure problem is the one that gets talked about least and causes the most damage. When your e-commerce team has a separate P&L from your retail team, and both are measured on channel-specific revenue targets, you have created an incentive structure that actively works against omnichannel behaviour. The e-commerce team will resist anything that sends customers to stores. The retail team will resist anything that credits an online channel for an in-store sale. Both teams are behaving rationally within their own measurement frameworks. The problem is that those frameworks were designed for a world that no longer exists.

This is not a technology problem. It is a commercial design problem. And no amount of investment in a customer data platform will fix it if the business is still structured to reward channel competition over customer outcomes.

If you are thinking about how omnichannel fits into a broader go-to-market approach, the Go-To-Market and Growth Strategy hub covers the commercial frameworks that sit underneath decisions like these, including how to structure channel strategy around business objectives rather than platform availability.

Why Attribution Makes the Problem Worse

Attribution in an omnichannel environment is genuinely hard. A customer sees a display ad on Monday, searches the brand on Wednesday, visits a store on Friday, and completes the purchase on the brand website on Saturday. Which channel gets the credit? In most standard measurement setups, the answer is the paid search click on Saturday morning, which is almost certainly the least important touchpoint in that entire sequence.

I have sat in enough post-campaign reviews to know what happens when attribution is broken. The channels that look good in the data get more budget. The channels that drove the actual decision but did not capture the last click get cut. Over time, you end up with a marketing mix that is optimised for measurement rather than for growth. You are very efficiently capturing the demand you already have, while systematically underinvesting in the activities that would create new demand.

This is not a new insight. Market penetration research has long pointed to the importance of reaching new audiences, not just converting existing ones. The brands that grow market share are typically the ones that invest in reach, not just conversion. But when your attribution model makes reach look like waste, you will keep defunding it, and your growth will plateau.

The honest answer to the attribution problem in omnichannel is not a better attribution model. It is a more honest conversation about what each channel is actually for. Some channels create awareness. Some create consideration. Some close. If you try to measure an awareness channel on close rates, you will always conclude it is not working. That is not an insight. That is a measurement error.

The Role of Physical Retail in a Digital-First World

There is a version of the omnichannel conversation that treats physical retail as a legacy problem to be solved rather than a commercial asset to be deployed. That framing is wrong, and the brands that have acted on it have generally regretted it.

Physical retail, when the product category involves any kind of tactile evaluation, is one of the highest-converting touchpoints in the entire purchase experience. Think about the clothes shop: someone who tries something on is far more likely to buy than someone who only views it online. That conversion differential exists because the store has done something no digital channel can fully replicate. It has removed uncertainty. The customer knows how it fits, how it feels, whether the colour is what they expected. That is not a small thing. It is often the deciding thing.

The problem is that most retail measurement does not capture this well. If a customer researches online, visits a store to try the product, then buys online later that day, the store gets no credit in most attribution setups. Over time, that undervaluation leads to underinvestment in retail, which degrades the very touchpoint that was driving conversion. It is a slow, self-inflicted wound.

Smart omnichannel brands are doing the opposite. They are treating stores as experience infrastructure, not just transaction points. The store’s job is not to close every sale on the spot. Its job is to remove uncertainty, build confidence, and create the kind of brand experience that digital cannot replicate. Sales that follow, whether they happen in-store or online, are downstream of that experience.

Understanding how physical and digital channels interact at different stages of the purchase experience is a core part of commercial growth strategy. The brands getting this right are not treating retail and e-commerce as competing priorities. They are treating them as complementary stages in the same customer experience.

How to Think About Channel Roles in an Omnichannel System

The most useful reframe for omnichannel strategy is to stop thinking about channels as revenue generators and start thinking about them as job performers. Every channel in your commercial ecosystem has a specific job to do. The question is whether it is doing that job, and whether you are measuring it against the right criteria.

A rough but useful framework breaks channels into three functional categories: channels that build awareness and create demand, channels that support consideration and evaluation, and channels that convert. Most brands have channels in all three categories but measure all of them on conversion metrics. That is where the distortion starts.

Paid social, for most categories, is primarily a demand creation and awareness channel. Its job is to reach people who were not already looking for you, introduce the brand or product, and create enough interest that they will look further. Measuring it on direct return on ad spend will almost always make it look inefficient, because most of the value it creates shows up in other channels later. That does not mean it is not working. It means you are measuring the wrong thing.

Organic search and content are primarily consideration channels. A customer who finds your product through a detailed comparison article or a well-structured product page is further along in their decision process than someone who saw a social ad. The job here is not to create demand but to support the evaluation that demand creation has already triggered. Measuring content on the same metrics as paid acquisition will produce equally distorted conclusions.

Paid search, email to existing customers, and in-store conversion are primarily close channels. They work best when there is already intent or familiarity. Pouring budget into these channels without investing in the channels upstream is like opening more checkout lanes in a store that has no customers browsing. You will be very efficient at processing the small number of people who arrive, but you will not grow.

When I was growing the agency from around 20 people to over 100, one of the most important commercial decisions we made was to stop selling individual channel services and start selling integrated programmes. Not because integration was a trend, but because single-channel briefs almost always produced single-channel thinking, and single-channel thinking consistently underperformed against what an integrated approach could deliver. The clients who trusted us with a broader remit grew faster. That is not coincidence.

The Data Infrastructure That Actually Matters

Every conversation about omnichannel eventually arrives at data. Specifically, at the idea that a customer data platform, or a better CRM, or a more sophisticated analytics stack will solve the integration problem. Sometimes it will. More often, it will not, because the data problem in omnichannel is not primarily a technology problem. It is a data quality and data governance problem.

The most common data issues I see in omnichannel programmes are: customer identities that cannot be matched across channels because different systems use different identifiers, offline purchase data that never makes it into the digital analytics environment, and behavioural data that is collected but never actioned because no one owns the process of turning it into decisions.

A new platform will not fix any of these. What fixes them is deciding who is responsible for data quality, creating the processes that keep data clean and current, and building the internal capability to actually use the data once it is there. That is less exciting than buying new technology, but it is where the actual value comes from.

The brands that have genuinely useful omnichannel data infrastructure share a few characteristics. They have a single definition of what a customer is, and that definition is consistent across every system. They have invested in connecting offline and online data, even imperfectly, because imperfect connection is far more useful than no connection. And they have someone, or a team, whose job it is to turn that data into commercial decisions rather than just into reports.

This is also where agile operating models can make a real difference. The ability to take a signal from the data and act on it quickly, adjusting channel mix, creative, or pricing in response to what you are seeing, is a genuine competitive advantage. But it requires both the data infrastructure to see the signal and the operating model to act on it. Most brands have neither fully in place.

Inventory and Fulfilment: The Operational Layer That Kills Omnichannel

There is a part of omnichannel commerce that marketing rarely owns but that marketing always gets blamed for when it goes wrong: inventory and fulfilment. If a customer has a smooth discovery and consideration experience across channels but then finds that the product is out of stock online, available in a store 40 miles away but not transferable, and the customer service team cannot see what the store team sees, the omnichannel experience has failed. Not because the marketing was wrong, but because the operational infrastructure was not built to support the promise the marketing made.

This is a structural problem that requires executive-level commitment to fix. Unified inventory visibility, flexible fulfilment options including ship from store, buy online and pick up in-store, and consistent stock availability data across all customer-facing touchpoints are not marketing decisions. They are supply chain and operations decisions. But they directly determine whether the omnichannel experience the marketing team is building actually works in practice.

The brands that do omnichannel well have typically made this a board-level priority rather than a marketing project. They have invested in the operational infrastructure that makes the commercial promise deliverable. The ones that struggle have usually tried to build the customer experience layer on top of an operational model that was never designed to support it.

Understanding where omnichannel sits within a broader commercial transformation is worth exploring in more depth. The BCG commercial transformation framework is a useful reference for how to think about the organisational changes that make programmes like this work at scale.

What Good Omnichannel Commerce Actually Looks Like

The brands that execute omnichannel well do not necessarily have the most channels or the most sophisticated technology. What they have is clarity about what they are trying to achieve commercially, and an operating model that is aligned to that objective rather than to individual channel metrics.

They have made deliberate decisions about which channels serve which functions in the customer experience, and they measure those channels accordingly. They have invested in connecting their data, even imperfectly, so that they can see the customer experience across touchpoints rather than just within individual channels. They have built internal structures that reward customer outcomes rather than channel-specific revenue, which means their teams are incentivised to collaborate rather than compete.

They also tend to be honest about what they do not know. Attribution in an omnichannel environment will never be perfect. The best brands accept that and make decisions based on honest approximation rather than waiting for measurement certainty that will never arrive. They know, roughly, what each part of the system is contributing, and they fund it accordingly.

Early in my career, when I was handed the whiteboard pen at a brainstorm and told to run with it, the instinct was to reach for the most sophisticated idea in the room. What I learned, over time, is that the most commercially effective ideas are usually the clearest ones. The ones that have a direct line from the customer insight to the commercial action, with no unnecessary complexity in between. Omnichannel strategy is no different. The goal is not to have the most elaborate channel architecture. The goal is to make it as easy as possible for the right customers to buy from you, wherever they are and however they want to shop.

Campaigns that reach new audiences through the right channel mix, rather than just recapturing existing intent, are the ones that show up in growth numbers rather than just efficiency metrics. That distinction matters whether you are managing a retail portfolio or a direct-to-consumer brand, and it is one of the core ideas running through the Go-To-Market and Growth Strategy section of this site.

The Measurement Framework That Fits Omnichannel Reality

Given everything above, what does a sensible measurement framework for omnichannel commerce actually look like? Not perfect attribution. Not a single dashboard that magically unifies all channel data. Something more honest than that.

Start with commercial outcomes at the top: total revenue, customer acquisition, retention rates, and average order value across the business, not within individual channels. These are the numbers that tell you whether the omnichannel programme is working commercially. Everything below them is context, not conclusion.

Below the commercial outcomes, measure each channel against the job it is supposed to do. Reach and frequency for awareness channels. Engagement quality and consideration metrics for mid-funnel channels. Conversion rate and cost per acquisition for close channels. This gives you a picture of whether each part of the system is performing its function, without forcing every channel to justify its existence on conversion metrics alone.

Use incrementality testing where you can. Controlled experiments that isolate the effect of specific channels or tactics are more reliable than attribution models in most omnichannel environments, because they do not require you to solve the identity matching problem across channels. They just require you to measure what happens when you do something versus when you do not. That is a much simpler question to answer.

And invest in qualitative data alongside quantitative. Customer surveys, store staff feedback, and post-purchase research will tell you things about the purchase experience that your analytics platform never will. The customer who says they came into the store because they saw an ad three weeks ago is giving you information that no attribution model would have captured. That information is commercially valuable. It should inform budget decisions, not just sit in a research report.

The reason go-to-market feels harder than it used to is partly because the measurement environment has become more complex as channels have multiplied. But the underlying commercial logic has not changed. Reach the right people, give them a compelling reason to buy, make it easy for them to do so, and measure whether it worked at the business level, not just the channel level. Omnichannel commerce is just that logic applied across a more complex channel architecture.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions