A personal reflection from MetaRouter CEO, Nikhil Raj.

Most technology stories start with a breakthrough. Most breakthroughs happen because of constraints.

When I started the advertising business for Walmart within Walmart Labs in early 2012, we were faced with a paradox - a significant constraint that led to a breakthrough.

Walmart had enormous customer traffic, online and in-store, and had become the world’s largest company without a traditional loyalty program. Its philosophy was Every Day Low Price. Low Price was the loyalty program. The paradox - the largest company was built without a loyalty program - was also a constraint.

The intentional lack of a loyalty program meant that there were no stored personal identifiers, even in eCommerce. Most customers checked out as guests, with email used only for transactional communication. And underpinning all of it was a deeply held belief that customer trust mattered more than short-term marketing gain.

At the same time, the suppliers participating in the launch of the advertising business wanted something reasonable that all advertisers wanted: to understand who their advertising was reaching, whether it worked, and how to invest better.

So the question I faced as we launched WMX, the advertising business at Walmart was: how do we provide advertisers the targeting and measurement they need, without a loyalty program? How do we do this without compromising privacy, performance, or trust?

Solving the problem at the first mile

The solution my team built wasn’t flashy. It was a deep infrastructure architecture. They designed a system that treated customers as anonymous, privacy-safe identities, not email addresses or cookies.

Anonymous IDs connected browsing behavior, purchases, and engagement, but always under Walmart’s control, inside Walmart’s data infrastructure.

Most importantly, the data was captured the moment it was created, and distributed in real time; not batch-processed later or shipped off to dozens of third parties. The two platforms that did this at Walmart were called Hubble and Pulse.

That approach delivered multiple benefits simultaneously:

• Faster walmart.com with no 3rd party tracking pixels
• Compliant data practices - fully consented under Walmart’s complete control
• and meaningful commercial impact with higher conversion rates on walmart.com, higher return on ad spend for marketing and higher ad revenue for the WMX, the ads business that kicked off Walmart’s efforts in advertising

It became the foundation of Walmart’s early advertising and marketing technology stack. And at the time, it solved the constraint and paradox we needed to solve.

We documented the thinking in a technical presentation: remove third-party pixels, centralize first-party data, and activate audiences without giving away the keys to the business.

Then I moved on.

A decade later, the same problem reappeared

Fast forward nearly ten years. I started seeing organizations wrestling with the exact same challenges, only at a much bigger scale:

• websites slowed by hundreds of tags
• data scattered across vendors
• rising privacy and compliance risk
• AI layered on top of data that wasn’t fit for purpose.

What surprised me most wasn’t the problem. It was how familiar the failed approaches felt.

Then I came across MetaRouter.

And I had one of those rare professional moments where you stop mid-conversation and say:

“You’ve built the system we designed at Walmart, but as a product.”

The MetaRouter team had independently arrived at the same architectural conclusion:
get the data right at the first mile, anonymous and privacy-safe before anything else touches it. They had taken what was once a bespoke, Walmart-only piece of technology and turned it into a platform that any company could use. The coincidence was that they were inspired by the same technical presentation that the team at Walmart had published. That was the full-circle moment for me.

That’s why I joined MetaRouter

I’d already seen this approach work, at extreme scale, inside the world’s largest retailer. What MetaRouter has done is make that capability accessible to everyone else.

Companies shouldn’t need:

• 20 engineers
• 18 months
• and a tolerance for architectural debt to control their own data properly.

MetaRouter sits quietly at the start of the journey, ensuring data is:

• clean
• compliant
• real-time
• and owned by the business.

Exactly how Hubble / Pulse did it for Walmart.

Everything downstream, from retail media to AI, benefits.

Why the first mile matters now more than ever

What’s changed since my Walmart days isn’t the challenge, it’s the context. AI has arrived and enabled real-time action like no other time before. Much of what we call “AI-driven marketing” today still operates too late in the process:

• analysing data after it’s been copied,
• trying to clean it downstream,
• or drawing insights from yesterday’s data.

This is like archaeology, digging through the past and hoping it tells you what to do next.

The real opportunity is to operate in the present moment, in real time.

When data is captured and governed at the first mile, the instant a customer interacts, entirely new things become possible:

• real-time personalisation,
• real-time suppression,
• real-time audience creation,
• real-time attribution,
• and real-time AI decisioning.

All without exporting data, duplicating it, or slowing the experience. This isn’t about replacing CDPs, ad platforms, or analytics tools. It’s about making everything else work better by fixing the foundation.

Looking ahead

We’re entering a period where:

• retail media is maturing
• AI is accelerating
• and regulators and customers alike are raising the bar.

In that environment, the winners won’t be the companies with the most data. They’ll be the companies with the best-governed data at the moment it’s created. I’ve built this system before. What’s different now is that the world is finally ready for it.

And that’s why the first mile matters.

‍