In this episode of DilSe Omni, Ankur Khaitan, Principal at Fireside Ventures, shares how India's leading venture capital firms evaluate consumer brands, why omnichannel has become non-negotiable, and what founders must do to build iconic businesses in the next decade.
One part of our conversation brought the opportunity into sharp focus: how much we can now understand about what happens inside a physical store.
For years, online businesses have studied customer journeys through traffic, product interactions, cart additions, and checkout conversion. Offline retail often had a much narrower picture: what sold, what remained, and what customers reported afterwards.
That gap is closing. As Ankur explained, structured retail data is making it possible to understand not just the transaction, but the journey leading up to it. For me, that changes the conversation about customer retention, store performance, and what being truly omnichannel means.
Table of Contents
1. The Real Shift Is From Channel First to Consumer First
2. Customer Retention Now Has Three Layers
3. NPS Is About Advocacy, Not Just Satisfaction
4. The Store Has a Funnel Too
5. Better Questions Make Store Data Useful
6. Fast Fashion Makes the Need More Urgent
7. Climb the Retail Analytics Pyramid
1. The Real Shift Is From Channel First to Consumer First
Ankur began with two observations: the speed at which retail can expand, and the amount of analytics now possible in an offline setup.
The second is especially important. Analytics has been central to online growth. Now, physical retail can generate structured signals about where customers go, what attracts them, which products they try, and where they abandon a purchase.
The benefit is not simply a better store dashboard. It is a different way of thinking.
When we organise everything around channels, we ask how the website is performing or whether a particular store is delivering sales. Those questions matter, but they are incomplete. A consumer-first approach asks what the customer is experiencing, irrespective of the channel.
My takeaway from Ankur: omnichannel becomes more meaningful when the unit of understanding is the consumer, not the channel.
That means using data to identify needs and friction, rather than treating online and offline as separate businesses with separate explanations for customer behaviour.
2. Customer Retention Now Has Three Layers
During our discussion, I shared how the meaning of retention has evolved. We still use the same word, but it can describe very different challenges.
Layer one: Retaining the relationship
The classical definition is repeat purchase. A customer buys a product, returns for another, and continues buying. Before digital became central to retail, loyalty programmes were one of the familiar ways to support that relationship.
This remains important. But repeat purchase is no longer the only retention problem a brand must solve.
Layer two: Retaining attention through the buying journey
A person reaches your website or installs your app. What happens next? Can you keep them engaged long enough to experience the product and make a purchase?
The same question applies inside a store. Entry does not guarantee engagement. A customer may need to test a product, sit on a sofa, lie down, or simply spend enough time with it to understand whether it is right for them.
Getting someone through the door is acquisition. Helping them stay engaged with the product is another challenge entirely.
Layer three: Retaining attention in the content feed
Social media has made retention even more immediate. People swipe quickly, and creators examine retention graphs to understand where attention drops.
I do this with my own podcast too. Where are people dropping off? At one minute, five minutes, or ten minutes? Those patterns tell us something about whether we are sustaining interest.
A useful framework is therefore relationship, journey, and attention. Before discussing retention, define which layer you mean. A repeat-purchase problem and a content-attention problem will not be solved in the same way.
3. NPS Is About Advocacy, Not Just Satisfaction
Another part of our conversation concerned a metric that is widely used and often misunderstood: Net Promoter Score.
The discussion used the terms NPS and eNPS loosely, but the customer recommendation question we examined belongs to customer NPS. Employee NPS concerns employees recommending their workplace.
My point was simple: asking whether someone would recommend your brand to friends and family goes beyond asking whether they had an acceptable experience. You are asking whether they are willing to become an advocate.
“It’s not about repeating. I’m going to the next step.”
Saurabh Agrawal, on the difference between returning and recommending.
I see that as a growth signal. A customer may return without actively recommending you. Advocacy asks for something more.
However, the way we manage the metric can damage its usefulness. Imagine giving a store manager a target of 70 or above and tying incentives directly to it. The pressure can move from improving the experience to persuading customers to submit a nine or ten.
Many of us have experienced this after a car service: the feedback request becomes a plea to protect somebody’s incentive.
Now the score is no longer a clean signal.
My caution is to avoid incentive structures that encourage staff to influence the response. Digital feedback, including a survey sent through WhatsApp, can reduce direct interpersonal pressure. But the larger principle remains: protect the honesty of the feedback before celebrating the number.
4. The Store Has a Funnel Too
Ankur took the discussion beyond surveys. Retail analytics can now reveal where customers spend time, which products they pick up, and what they take into a trial room.
But activity is not the same as an outcome.
“What matters is how many products move out of the trial room into the shopping bag.”
Ankur Khaitan, on measuring meaningful conversion.
That distinction is powerful. A busy trial room may suggest interest. It does not tell you whether the product experience is translating into a purchase.
Here is a practical way to organise the offline funnel we discussed:
Entry: Who comes into the store, and where do they go first?
Attention: Which aisles or displays attract them, and where do they spend time?
Interaction: Which products do they pick up and examine?
Trial: What reaches the trial room, and how long does the experience take?
Purchase: What moves into the shopping bag and completes a transaction?
The purpose is to identify broken journeys. If customers repeatedly try products but do not buy them, the useful question is what happens between those two stages.
A survey can help explain the experience. Behavioural data can show where to investigate. Together, they provide a fuller picture than either one alone.
5. Better Questions Make Store Data Useful
I connected Ankur’s observations to an earlier conversation with Tambo AI about the store funnel. Online teams routinely examine landing-page conversion, cart additions, and checkout. Offline teams can apply a similar discipline to aisle and trial-room conversion.
There are also questions about who is shopping. Is it an individual, parents with a child, or a family group including grandparents? Who is interacting with the product, and who ultimately purchases it?
These details help us understand the shopping occasion rather than treating every store entry as identical.
Ankur also highlighted how quickly merchandising decisions can now be informed. Earlier, much of the analysis happened after the fact: what inventory was on the shelf, what was stored below it, and what was displayed on a mannequin.
With more timely data, teams can ask:
What is attracting attention as customers enter?
Where are they spending the most time?
Which products are being selected for trial?
Which products reach the cash counter but still do not convert?
That final question is particularly revealing. Even reaching the counter does not always mean the decision is complete.
The value of retail analytics is not collecting more events. It is making the next decision better informed.
6. Fast Fashion Makes the Need More Urgent
This reminded me of my conversation with Sumit from NEWME. In fast fashion, styles change quickly. Products cannot simply occupy shelf space indefinitely while the business waits to understand demand.
When the assortment moves rapidly, learning must move rapidly too. Understanding attention, trial, and purchase becomes especially useful because sales alone describe only the final outcome.
As Ankur noted, the availability of these capabilities is making it easier for founders and brands to build their retail playbooks. The opportunity is to turn those capabilities into a regular operating habit.
7. Climb the Retail Analytics Pyramid
Towards the end, I described a pattern I often see in how businesses prioritise analytics:
Sales analytics: What sold, and how much?
Inventory analytics: What stock is available, and what needs attention?
Customer and marketing analytics: Who is buying, and how are we reaching them?
Customer experience analytics: Where is the journey working, and where is it breaking?
Sales gets attention first. Inventory follows, especially in retail. Customer experience often gets whatever time and energy remain.
That is the imbalance worth correcting. Sales tells us the result. Understanding the experience helps us investigate how that result came about.
My learning from Ankur is not that every brand needs every possible metric. It is that we now have the opportunity to connect customer behaviour, feedback, and commercial outcomes much more thoughtfully.
Start with one broken journey. Understand it across the relevant touchpoints. Improve it, then learn again. That is a practical path from channel-first reporting to consumer-first retail.
I am Saurabh Agrawal and we come with a new episode on Dilse omni talks every fortnight and cover different aspect of omnichannel with amazing speakers.
