This article emerges from the conversations at the Retail & eCommerce Minders Meetup – Buenos Aires, where Joaquín Bossié, Gonzalo Oliva Vélez and Andrés Kloster shared how growth in retail and eCommerce is evolving: from building products that scale to using data, AI and customer journeys to turn traffic into loyal customers.
How to Connect Product, Data and Experience to Scale
Growth in retail and eCommerce has never had so much potential. At the same time, it has rarely been this difficult to sustain.
Many companies are investing more in marketing, launching new features and sending more campaigns. Effort increases, yet results do not always follow. The problem is rarely a lack of work, but something deeper: how organizations learn—or fail to learn—from user behavior.
Understanding that difference is what separates teams that simply operate from those that truly scale.
Sustainable digital growth rarely comes from doing more things. It comes from building a system that continuously learns from user behavior.
Why Digital Growth Became Harder
For a long time, digital growth seemed relatively straightforward to explain. If a company wanted to sell more online, the path was fairly clear: invest more in acquisition, optimize conversion on the website and launch new product features. Marketing brought users in, product improved the experience and CRM re-engaged those who did not return.
That model worked for years. However, in many companies today it is starting to reach its limits.
Over time, a frustrating pattern begins to appear: companies acquire users, lose users and then acquire them again at a higher cost. Effort increases, budgets grow, but growth does not scale at the same pace. Meanwhile, margins shrink and the internal perception is that everyone is working harder, but results are not improving proportionally.
The issue is rarely a lack of effort. In most organizations, there is plenty of activity.
The real problem is often more subtle: the illusion of control.
Joaquín Bossié – Co-founder & CRO at Minders
The Illusion of Being Data-Driven
Many companies believe that having dashboards, marketing reports, sales metrics and active campaigns means they are truly data-driven organizations.
In practice, something else often appears.
Many organizations have data. What they do not always have is learning.
There are metrics, but not necessarily better decisions.
There is activity, but not necessarily impact.
When Teams Work but the System Does Not Learn
A common scene in digital retail companies illustrates this clearly.
It is Monday morning and the weekly performance meeting begins. Marketing explains that acquisition spending increased, but ROAS dropped. The product team reports that new features were launched, although activation did not improve significantly. CRM shows that more campaigns were sent, yet churn barely moved. Meanwhile, the data team tries to explain what is happening, often encountering instrumentation issues, incomplete events or misaligned data sources.
At some point someone usually says the next step should be implementing AI-driven personalization.
Everyone agrees.
But very few people are actually clear on what that means in the specific business context.
Each team is doing its job. Marketing runs campaigns, product ships features, CRM sends communications and data tries to explain what is happening.
Yet the system as a whole is not necessarily learning.
Product measures usage. Marketing measures campaigns. The business measures revenue. But when those layers are not connected, improving outcomes becomes extremely difficult.
As a result, organizations often optimize isolated parts of the system without fully understanding which user behaviors are actually driving the business.
The consequence is that many companies operate with a huge amount of activity but relatively little accumulated learning.
That is one of the central challenges of modern digital growth.
And solving it requires changing the way organizations think about growth.
The Paradigm Shift: From Activity to Learning
Companies that manage to scale sustainably usually operate with a different logic.
In essence, growth begins to look less like a marketing plan and more like a learning system. They grow because they learn faster what actually works.
This learning is built around a relatively simple but extremely powerful loop: formulate hypotheses, run experiments, learn from the results and scale what works. When that cycle repeats with speed and discipline, each iteration improves the next. The system begins to accumulate knowledge about real user behavior.
In that context, the goal is no longer simply to generate more activity. The goal is to improve the quality of decisions.
That distinction may seem subtle, but it profoundly changes how organizations work.
Instead of asking what features they should launch or which campaigns they should send, teams begin asking what behaviors they need to understand better and what experiments they should run to learn something meaningful.
Companies that build this type of system develop a competitive advantage that is extremely difficult to copy: learning velocity.
Gonzalo Oliva Velez – Product & Engineering Advisor on E-commerce | Ex Chief Product Officer at Vestiaire Collective | Lazada (Alibaba group) | Mercado Libre | Tiendamia | IBM
The Three Real Levers of Digital Growth
When growth is analyzed through this lens, most of the impact tends to concentrate around three key levers.
Activation
The first lever is activation. It refers to the moment when a user experiences value for the first time within a product or service. It is the instant when someone truly understands why the experience is worth it.
In many cases, this moment is filled with invisible friction: unnecessary registration steps, unclear value propositions, complex onboarding flows or overly long checkouts.
When activation does not work well, the entire system suffers. Marketing must spend more money to compensate for that friction, and retention becomes harder because users never fully experience the product’s value.
Retention
The second lever is retention. This is not simply about getting users to come back, but about building a habit.
Many companies confuse retention with recurring promotions. True retention appears when users return because they find value in the experience, not because they received a temporary discount.
Understanding retention means identifying which behaviors predict loyalty, which experiences drive recurrence and how to accompany users throughout their relationship with the brand.
Monetization
The third lever is monetization.
This is where decisions related to purchase propensity, product recommendations, relevant cross-sell, loyalty programs or new models like retail media appear.
However, these strategies only work well when they are grounded in a deep understanding of real user behavior. Otherwise, they simply become noise—or spam.
Activation, retention and monetization form an interdependent system. Improving one dimension often impacts the others.
Companies that scale successfully do not treat these areas as isolated initiatives, but as parts of the same growth engine.
The Central Role of User Behavior
For that engine to function correctly, organizations must understand what users are actually doing inside the product.
Business metrics such as revenue or conversion are important, but they often arrive too late to explain what is happening.
By the time a metric appears in the P&L, the behavior that generated it has already happened.
What truly reveals the system are behavioral signals: the steps users take within the product, where they drop off, which sequences of actions precede a purchase or abandonment.
Product analytics tools exist precisely to answer these types of questions.
Funnels, cohorts, journey analysis and experimentation allow teams to observe the product not as a list of features, but as a dynamic system of behaviors.
When organizations instrument behavior correctly, clarity begins to emerge. It becomes possible to identify specific frictions, understand which experiences generate value and design improvements that genuinely impact the business.
From Insight to Action
Understanding behavior is only the first step.
The next challenge is acting on that behavior at the right moment.
This is where experience orchestration comes into play. It is not simply about sending mass campaigns, but about intervening in the user journey when a relevant signal appears.
A reminder when someone abandons a cart. A personalized recommendation based on recent behavior. A contextual experience that makes the next action easier.
When this orchestration is connected to real behavior, the experience stops feeling generic. It begins to feel relevant, timely and personalized.
The Missing Piece: The Growth Operating Model
Even when data and tools exist, many organizations still struggle to improve results. The reason is often organizational.
Sustainable growth rarely comes from a single team. It comes from a system of work.
In many companies, product, marketing and business operate with different logics and different metrics. Each team optimizes its own domain, but no one is optimizing the system as a whole.
That is why more and more companies are adopting what is often called a growth operating model: a way of working where product, data and marketing function as parts of the same learning system.
Decisions are prioritized based on expected impact, experiments are designed to answer concrete questions and results become cumulative knowledge.
Instead of operating through rigid roadmaps or isolated campaigns, organizations begin working through continuous learning cycles.
When this model works well, growth stops depending on isolated initiatives and becomes a systematic process of improvement.
Andrés Kloster – Founder & Chief Revenue Officer at Eleven
The New Discovery Layer: From SEO to GEO
User behavior is not only changing inside products. It is also changing in the way people discover products.
For years, much of retail growth depended on mastering traditional SEO: appearing at the top of Google search results.
However, the emergence of AI-based assistants is beginning to reshape that model.
More and more users ask questions directly to systems like ChatGPT, Perplexity or new AI-powered search experiences. Instead of exploring a list of results, they receive synthesized answers that cite a limited set of sources.
This introduces a new dynamic.
It is no longer just about ranking a page in Google. It is about making content citable by AI models.
In this context, a new concept is emerging: GEO (Generative Engine Optimization).
Optimizing for these systems means creating structured content that models can understand, summarize and reference. In retail and eCommerce this often takes concrete forms: product comparisons, buying guides, “top product” rankings and informational content that answers real user questions.
The objective is no longer only to attract traffic. In some cases, traffic may even decrease.
But when content becomes a source cited by AI systems, the quality of demand that arrives tends to be significantly higher.
AI as a Growth Accelerator in Retail
Artificial intelligence is also changing how marketing and content operations scale.
In eCommerce, for example, generating content for product pages, creating advertising creatives or producing UGC-style content can now scale much faster using AI.
This enables something that was previously difficult: faster experimentation.
Marketing teams can test multiple content variants, creatives and acquisition formats at much lower costs. Product teams can generate richer content for large catalogs. Growth teams can iterate faster on which messages actually convert.
In that context, AI does not replace strategy.
But it becomes a powerful operational accelerator for executing and learning at higher speed.
Learning Faster Than the Market
Ultimately, sustainable growth in retail and eCommerce does not depend solely on technology, marketing or product individually.
It depends on an organization’s ability to continuously learn from user behavior.
Companies that move fastest tend to share one characteristic: they have built a system that converts behavior into learning—and learning into decisions.
That system combines three fundamental layers:
-
The ability to understand user behavior
-
The ability to act on that behavior through relevant experiences
-
An operating model that connects both to business decisions
When those layers work together, growth stops depending on isolated initiatives and becomes a continuous improvement process.
And in a market where competition intensifies and user expectations keep rising, the hardest competitive advantage to copy is not a campaign or a feature.
It is the ability to learn faster than the rest of the market.