Commerce Velocity: The new currency in e-commerce

“I skate to where the puck is going to be, not where it has been.”
— Wayne Gretzky

Most e-commerce dashboards are quite good at showing us where the puck has been. Revenue, conversion rate, average order value, customer acquisition cost, ROAS, and return rates are now part of every e-commerce manager's standard repertoire. These metrics are important because they show how efficiently an existing business model is performing.

But commerce is changing faster than many of these metrics can capture. New marketplaces are emerging, product ranges and sales channels are becoming more international, AI search is transforming product discovery, and agentic commerce is already signaling the next stage of development.

In this world, it is no longer enough to just know how well a company is selling today. The more critical question is:

How quickly can it tap into new revenue opportunities?

This is exactly where a new generation of e-commerce KPIs begins.

The old KPIs aren't wrong – they just tell half the story

Revenue, conversion rate, ROAS, and CAC are certainly not going to disappear. No e-commerce manager will be able to do without these metrics in the future. However, they primarily answer a backward-looking question:

How well has our existing commerce model performed?

That is something different from the question of how well a company is positioned for future growth.

For example, a retailer can achieve an excellent ROAS and still be structurally slow. They can achieve a high conversion rate while taking months to integrate a new supplier. They can have over 100,000 SKUs but only offer a fraction of them on relevant marketplaces. And they can have outstanding product pages while their products are barely considered in AI-powered shopping experiences.

The problem, therefore, is not that classic KPIs are becoming obsolete. They simply measure one part of the value chain. They show how efficiently the existing business is running. They say significantly less about how quickly a company can tap into new products, new sales channels, and new forms of product discovery.

Perhaps this is the most important change in the logic of e-commerce KPIs:

We must start measuring adaptability alongside performance.

An unlisted product has a conversion rate of zero

Take a retailer who receives 10,000 new products from a supplier. On paper, that is enormous growth in the product range. In practice, those 10,000 products are initially just data.

Product information must be imported, verified, supplemented, categorized, and enriched. It then needs to be adapted to the requirements of each specific sales channel. A marketplace expects different attributes and taxonomies than your own shop, while another channel may require different data formats or additional content.

As long as this process is incomplete, the products cannot be sold. Your conversion rate is not at one or two percent. On that specific channel, it is at zero.

This may sound trivial, but it highlights why a KPI like time to market could become increasingly central in the future. How long does it actually take from acquiring a new product range to having a sellable product on a relevant channel?

The faster this process becomes, the sooner a product data set can be turned into revenue potential.

The question becomes even more interesting when you look beyond individual products to a company's overall ability to distribute its assortment. A retailer with 100,000 SKUs does not automatically have a digital assortment of 100,000 sellable products. Only when these products are fully and correctly available on the relevant channels can they generate revenue there.

From this, we can derive a KPI that could become significantly more relevant in the future:

product distribution rate — in other words, the share of the assortment that is actually fully available and ready for sale on the channels relevant to the company.

Product data quality is shifting from an operational task to a growth driver

This also changes the role of product data quality.

For a long time, product data quality was primarily viewed as a task for PIM, IT, or e-commerce operations. The goal was to ensure that product information was complete and correct within the system. Today, that perspective is no longer sufficient.

This is because product data influences more and more stages of the commerce value chain. Missing attributes can prevent a product from being listed on a marketplace. Incomplete information can degrade the product experience. Weak or irrelevant content can hinder organic discoverability.

AI systems also require sufficiently structured and relevant information to reliably understand products and categorize them as viable options.

This makes a product data quality score an interesting management metric. It is not just the question "How many products do we have?" that is relevant, but also:

How many of them possess the data quality required for their respective commerce channels?

This represents an important shift in perspective. Product data is no longer just a prerequisite for a functional online shop; it is becoming a core component of growth infrastructure.

From SEO to AI Visibility

This development becomes particularly clear when looking at how product search is changing.

Today, a customer doesn't necessarily search for "ergonomic office chair." They might ask an AI which office chair is suitable for someone who is 1.90 meters tall, works from home for eight hours a day, and wants to spend a maximum of 300 euros.

This fundamentally changes the role of product information. An AI system must do more than just recognize that a product is an office chair; it must understand its specific features, who it is suitable for, and how it differs from other products.

This creates a new form of visibility.

Previously, brands measured their share of voice primarily through search engine rankings, ads, and marketplace positions. In the future, it will also become relevant how often a product is identified as a viable option or actively recommended in AI-powered product searches.

Metrics such as AI Product Visibility, AI Recommendation Rate, or AI Share of Voice are currently less standardized than classic SEO KPIs. That is precisely what makes them interesting. Those who start monitoring this trend early and linking it to the quality of their own product data will gain a better understanding of how product discovery is evolving.

Perhaps the most important new KPI: Commerce Velocity

When you bring these developments together, an overarching concept emerges:

Commerce Velocity.

Commerce Velocity describes the speed at which a company can turn new commerce opportunities into sales-ready reality.

How quickly can a new supplier be integrated? How fast can 10,000 new SKUs be enriched and published? How quickly can a new marketplace be tapped into? How fast can existing product data be mapped to a new taxonomy? And how quickly can a company react to new forms of product discovery?

This is a different kind of performance measurement. It looks beyond the efficiency of the existing business to evaluate a company's ability to adapt to new market conditions.

In a dynamic commerce environment, this capability can represent a significant competitive advantage. A retailer with a smaller assortment may grow faster than a competitor with a much larger product base if they can activate new products and channels significantly faster.

Why Cost per Activated SKU is becoming more relevant

This also changes the way we look at the costs of an assortment.

Today, it is relatively easy for a company to state how many SKUs it has. However, the more interesting question is:

What does it actually cost to make a SKU ready for sale?

Product data must be imported, missing information added, content created, categories and attributes mapped, and the finished data transmitted to the respective channel. The more these steps are performed manually, the higher the costs for each additional SKU and each additional channel.

This is why Cost per Activated SKU could become an interesting metric for scaling commerce.

Taking it a step further is the Marginal Cost of Assortment Expansion:

What does it cost a company to expand its digital assortment by another 10,000 products?

If every additional SKU causes proportionally more operational work, assortment expansion becomes expensive. If, on the other hand, import, enrichment, content, mapping, and distribution are increasingly automated, the cost curve changes.

Then scaling becomes possible without operational costs growing at the same rate.

The PIM dinosaur has a problem

This is exactly where it becomes clear why the role of a PIM system is changing.

You can think of the classic PIM dinosaur as a system that is very good at storing and managing product data. Its dashboard shows 247,381 SKUs, 98 percent data completeness, and 43 attributes per product. Everything looks tidy.

Then the e-commerce manager comes along and asks:

“We just received 12,000 new products from a supplier. Can we get them live on the new marketplace today?”

The dinosaur replies:

“No. But your data is complete.”

Of course, that’s an exaggeration. But the metaphor describes a real shift.

Classic PIM systems were primarily designed as Systems of Record were developed to create a central hub for product information, ensuring consistency and governance. This remains important.

However, the demands of modern commerce are increasingly going beyond that.

A system must do more than just know what data is available. It must be able to identify what information is missing, where to find it, what requirements a new channel has, and how existing product information needs to be transformed to meet them.

This is exactly where the transition begins from a System of Record to a System of Action – or from a classic PIM to an Agentic PIM.

An Agentic PIM wouldn't just report that a product is missing three attributes. It could identify these gaps, query the supplier or research the information, structure data using AI, and then map it to the specific requirements of the marketplace.

The process doesn't end with storing the data, but with its actual activation in commerce.

AINAVIO as an Agentic PIM

This is precisely the approach AINAVIO takes.

Supplier data can be imported automatically. Missing information can be identified and specifically requested. Product information can be researched and annotated with AI. Based on this enriched data, SEO and commerce content can be generated.

Subsequently, product data can be mapped to various marketplace and shop structures using AI and deployed via integrations.

The value lies not just in automating individual tasks, but in connecting these steps into a seamless, end-to-end process:

from the supplier's raw data set to a sellable product on the respective commerce channel.

This directly links product data management to the KPIs that drive future growth: time to market, product distribution rate, product data quality, cost per activated SKU, and ultimately, commerce velocity.

The goal is not simply to have a better-maintained PIM.

The goal is to turn existing product data into new commerce opportunities faster.

From Performance Dashboard to Commerce Readiness Dashboard

Perhaps this will also change the e-commerce dashboard of the future.

The classic KPIs remain: revenue, conversion rate, ROAS, CAC, average order value, and customer lifetime value. These will continue to show how efficiently your existing business is performing.

However, metrics that reflect the future viability of the business are becoming increasingly important:

What is your product data quality?

What percentage of your assortment is actually channel-ready?

How quickly are new products activated?

What is the cost of activating an additional SKU?

How quickly can a new marketplace be tapped into?

And how visible are products in an increasingly AI-driven product discovery landscape?

Ultimately, this represents a shift from a performance dashboard to a commerce readiness dashboard.

The question is no longer just:

"How much are we selling?"

But also:

"How quickly can we capture the next revenue opportunity?"

The future doesn't necessarily belong to the largest assortment

A company with a million products has an impressive catalog. But size alone is not a competitive advantage.

What matters is how quickly that assortment can be made available where customers are searching and buying.

A retailer with 100,000 products that can launch on a new marketplace within a few days can be more agile than a company with a million SKUs that takes six months to do the same.

This turns your assortment into a dynamic asset. It is not just about how many products a company has, but how quickly it can translate those products into new sales opportunities.

A product range only becomes a true commerce asset when it is distributable.

And distribution starts with product data.

The question is no longer just where the puck was.

Wayne Gretzky’s famous analogy about the puck perfectly captures this shift.

Most e-commerce dashboards show us very precisely where the puck was. They show revenue, conversion, ROAS, and CAC.

But the more important question for the future might be:

How quickly can we get to where the puck is going to be next?

Maybe it’s on a new marketplace. Maybe in a new country. Maybe in a new product category. Maybe in an AI-generated product recommendation.

No one can predict with certainty which commerce channel will be dominant in five years. However, companies can build their infrastructure to react quickly when the market shifts.

That is why Time to Opportunity could ultimately be one of the most interesting metrics in future e-commerce: the time that elapses between identifying a new commerce opportunity and actually activating it.

A new supplier, a new product range, a new marketplace, or a new AI commerce model is initially just a possibility.

Only when product data can be imported, enriched, structured, mapped, and deployed quickly enough does it become a real revenue opportunity.

Perhaps that is where the real KPI revolution in e-commerce lies:

The KPIs of the past measure how well we are selling today. The KPIs of the future measure how quickly we can tap into new opportunities tomorrow.

And for that, you don’t need a PIM that only knows what happened yesterday.

You need a system that helps you reach the next puck.