What if a meaningful part of your acquisition budget is going towards customers you have already acquired?
That was the uncomfortable question behind my conversation with Sidharth Gopalakrishnan on Dilse Omni Talks. I wanted to unpack “ad waste,” a term we hear increasingly often, and understand what it actually means for a brand’s growth economics.
About the Guest: Sidharth Gopalakrishnan is the COO of Netcore, a leading customer engagement and marketing technology platform, where he drives strategy on AI-led marketing transformation and enterprise partnerships, including Netcore’s collaboration with Google Cloud.
What followed felt like a masterclass in customer retention. We moved from wasted advertising spend to customer drift, from campaign execution to cohort analysis, and finally to the role of AI in creating more relevant customer experiences.
My biggest learning was simple: the cost of losing a customer does not end with the lost transaction. It can return as another acquisition bill.
Table of Contents
1. Ad Waste: Paying Again for a Relationship You Already Built
2. Customer Drift Happens Before Customer Dormancy
3. A Fashion Brand Experiment That Made the Economics Real
4. First-Party Data Must Become Useful Action
5. The Missing Capability Is Not Another Campaign Tool
6. The Three Real Jobs of MarTech
7. AI Personalization: From a Name to a Meaningful Moment
Customer Retention Is Now a Business Leadership Question
1. Ad Waste: Paying Again for a Relationship You Already Built
Sidharth began with the basic journey. Brands invest in performance marketing to bring people to their websites, apps, and stores. Some of those people purchase. A customer relationship begins, and the brand now has first-party information that can support future engagement.
But three months, six months, or a year later, some customers stop buying.
They move into a dormant segment. Eventually, the brand uploads those customers as custom audiences to advertising platforms and pays to bring them back into the funnel.
The customer has not become a stranger. Yet the business is spending as though the relationship must be acquired again.
“The amount of dollars that brands are spending today to acquire the same customer whom you may have acquired in the past.”
Sidharth Gopalakrishnan, explaining ad waste
Sidharth described an analysis across six e-commerce brands that led his team to estimate that about 30% of the spend they examined was wasteful, because some existing customers could have been reactivated through owned channels.
That is not a universal claim that every brand can cut its entire advertising budget by 30%. It is a diagnostic insight: how much paid acquisition is actually reacquisition, and how much of that could happen more efficiently?
Performance marketing is attractive because it is relatively easy to activate and scale. Retention has a finite customer base to work with. But that difference should not excuse mixing genuinely new customers with returning customers when evaluating acquisition economics.
2. Customer Drift Happens Before Customer Dormancy
The most important word in our conversation was not acquisition. It was drift.
A customer does not necessarily move straight from loyal to lost. Engagement weakens over time. Someone who was highly active becomes less active, then less responsive, and eventually dormant.
Sidharth noted that dormant customers can represent as much as 80% of the accumulated base for brands he was describing. He also referred to the familiar practice of using a period such as 180 days to classify inactivity.
The problem is not simply the label. It is what happens before the label appears.
Are we identifying declining engagement while there is still an opportunity to respond? Or are we waiting until the customer has effectively disappeared?
“I don’t think we’ve done a great job of understanding drift.”
Sidharth Gopalakrishnan
What I appreciated was his willingness to include retention platforms in that criticism. This was not just a complaint about brands. Platforms must become better at capturing signals, building cohorts, and triggering action when engagement starts falling.
A Simple Customer Drift Map
I would represent the journey Sidharth described like this. This is a visual summary of our discussion, not a named Netcore framework.
Active relationship | Early drift | Dormancy | Paid reacquisition |
|---|---|---|---|
Customer buys and engages | Engagement begins declining | Customer stops transacting for a sustained period | Brand pays an advertising platform to bring the customer back |
Understand preferences | Detect the change and intervene | Run a focused reactivation effort | Separate this cost from genuinely new acquisition |
The opportunity is to act earlier in this journey. A dormant-customer campaign matters, but preventing avoidable dormancy is an equally important capability.
3. A Fashion Brand Experiment That Made the Economics Real
I wanted to understand whether this was just a persuasive argument or something that could produce business results.
Sidharth shared an experiment with a fast-growing, profitable fashion brand known for quirky garments and designs featuring franchises such as Marvel and cartoons.
The team ran a dedicated three-month effort to revive dormant customers. He reported two striking outcomes:
About 30% to 40% of the dormant customers were revived.
Their average order value was approximately 1.5 times the brand’s prevailing average.
That second point matters. A customer who has drifted away is not automatically a low-value customer. The previous relationship may still contain meaningful purchase intent and brand affinity.
However, Sidharth was clear that the effort was neither instant nor free. Reaching customers through WhatsApp, email, RCS, or SMS carries costs. Planning and sustained execution also take work.
In this example, reactivation cost more than generating repeat purchases from active loyal customers, but less than bringing customers back through paid advertising.
The lesson is not that retention is free. It is that different customer states have different economics. Treating all returning revenue as one bucket hides those differences.
4. First-Party Data Must Become Useful Action
Why can owned-channel reactivation work? Because the customer is already identified.
The brand has first-party data. That information can be enriched and used to make communication more personalized and need-based, rather than sending the same message to everyone who has not purchased recently.
To turn Sidharth’s explanation into an operating framework, I would use four connected steps:
Identify: Establish which customers are already known to the brand, including those returning through paid campaigns.
Understand: Examine engagement, purchase history, and product or category affinity.
Intervene: Use relevant communication through owned channels, ideally before full dormancy.
Evaluate: Compare reactivation outcomes and costs with repeat purchasing and paid reacquisition.
This is where dashboards must change. Campaign activity alone does not tell us whether customer relationships are strengthening or weakening.
Our discussion pointed towards questions about three-month, nine-month, and twelve-month repeat rates. Are those rates improving? Which cohorts are weakening? Are we distinguishing a new customer from someone we have already paid to acquire?
5. The Missing Capability Is Not Another Campaign Tool
At this point, I pushed back slightly on the idea of fault.
Many retention marketers are set up as execution specialists. Their responsibility is to send the SMS campaign, launch the WhatsApp campaign, or deliver the email calendar. They are not necessarily equipped or supported to analyse cohorts and build a reactivation business case.
Meanwhile, quick results dominate decision-making. An immediate recovery campaign can look more attractive than a longer effort to revive dormant relationships.
I also raised a wider education problem. So much marketing education centres on Meta ads, Google ads, and influencer marketing. There is much less attention on the practical “how” of retention and omnichannel growth.
My concern is the promise that buying a tool will somehow create the outcome. I compared it to buying a car without knowing how to drive and expecting to reach the destination anyway.
Technology enables the journey. It does not replace the capability to make the journey.
I suggested that retention can sometimes deliver two or three times better ROI than paying to reacquire the same customer. But the important task is to establish that business case for the brand, rather than treat any multiplier as a guaranteed result.
6. The Three Real Jobs of MarTech
When I asked what marketing technology actually helps brands do, Sidharth brought the discussion back to three business outcomes.
MarTech role | Core question | Business purpose |
|---|---|---|
Acquisition and conversion | How do we help someone already in the funnel convert across touchpoints? | Turn interest into a transaction |
Customer experience | How do we make the interaction relevant to this person and their context? | Improve the quality of engagement |
Loyalty | How do we encourage the next relevant purchase? | Build repeat cycles and customer lifetime value |
This is a useful way to assess MarTech because it shifts the conversation away from feature lists. A platform should contribute to conversion, experience, and loyalty, not merely increase the number of messages a team can send.
7. AI Personalization: From a Name to a Meaningful Moment
Sidharth’s coffee example made contextual personalization easy to understand.
Putting someone’s name into a message is one level of personalization. Understanding their relationship with coffee, the monsoon season, and the relevance of that moment is another.
Imagine a rainy day, a cup of coffee, and communication that reflects the person’s preferences and situation. The experience becomes more meaningful because it connects behaviour with context.
He described AI and context graphs as ways to bring those connections together. He also acknowledged the dependency: richer contextual experiences require different sets of data.
For loyalty, the same principle applies to product, category, and subcategory affinity. The objective is a relevant nudge towards a repeat purchase, not simply another reminder that the brand exists.
Customer Retention Is Now a Business Leadership Question
Our conversation ended beyond the marketing department. Sidharth is seeing discussions involve business leaders, chief digital officers, and AI and data leadership alongside CMOs.
That makes buying and decision-making more complex, but it also makes the potential role of MarTech more meaningful.
My takeaway is that reducing ad waste requires more than a budget adjustment. It requires a better understanding of customer drift, stronger analytical capability, useful first-party data, and accountability for the entire relationship.
The central question remains: are we growing by acquiring new customers, or repeatedly paying to recover relationships we allowed to weaken?
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.
