E-commerce funnel: goodbye straight line, hello smart loop

Are you still thinking about the funnel? If you continue to optimize your marketing by thinking about a customer who obediently enters at the top of the funnel and, once converted, exits at the bottom, you are describing a customer who simply no longer exists. The linear, predictable, and orderly purchasing journey of the AIDA model is a historical relic, swept away by a chaotic vortex of digital touchpoints and reduced traffic from Google. Today, a customer sees a product on an Instagram reel, searches for it with a photo using Google Lens, reads a response generated by Google’s AI, watches a review on YouTube, asks a chatbot for advice, and perhaps makes a purchase via TikTok Shop during a live stream. This is marketing in 2025. So, without adding to the confusion, let’s say goodbye to the concept of the funnel and embrace the “smart loop,” a continuous cycle fueled by data and artificial intelligence, where each stage enhances the next and the previous, with the customer firmly at the center. The old TOFU, MOFU, and BOFU stages are giving way to a more dynamic cycle: Inspiration, Exploration, Purchase, and Experience.

E-commerce funnel: the classic model that worked (until yesterday)

To understand how we got to the current chaos, it’s worth taking a step back. There was a time—not so long ago—when the digital purchase journey seemed perfectly mappable. Just imagine a funnel—wide at the top, narrow at the bottom: users entered en masse, but only a fraction became customers.

The data you need is already in SEOZoom!

Analyze traffic, keywords, search intent, and rankings to understand where your funnel is getting stuck.

Find out more

Everything moved along a theoretically orderly flow: first discovery, then interest, then desire, and finally action. In short, the famous AIDA model.

• Awareness. Potential customers become aware of the brand or product through advertising, social media, SEO, or word of mouth.
• Interest. Once visitors are aware of the brand, they need to develop an interest in its products: here, quality content and detailed product information can make all the difference.
• Desire. At this stage, visitors need to be convinced that the product meets their needs or desires: customer reviews, testimonials, and product demonstrations can be effective tools for stimulating desire.
• Action. This is the final stage in which visitors become customers by completing their purchase. A simple and intuitive checkout process, together with secure payment options, can facilitate conversion.

We had begun to add an additional stage, Engagement, during which the customer engages with the brand and ideally becomes a supporter and promoter of the brand in turn.

For years, it worked. This was partly because it reflected the way people used the web: a Google search, a product listing, a click. But then behaviors, touchpoints, and expectations changed. And the funnel began to creak.

The problem? Today, people no longer move in a straight line. They discover a product while watching a reel, search for it with their camera, save it for later, talk about it with a chatbot, find it elsewhere. This is where the traditional model begins to show all its limitations.

The funnel for e-commerce today: from funnel to loop

Thinking of the sales funnel as an orderly funnel—curious people in, customers out—was convenient. It worked for years because the web was less crowded, there were few channels, and the purchasing process followed a fairly predictable logic. You attracted attention, stimulated interest, built desire, and then drove action. AIDA: simple, straightforward, reassuring.

Today, that linearity no longer exists. People discover a product on a reel, search for it with an image, read reviews from a chatbot, watch a video on YouTube, and maybe—just maybe—buy from a marketplace. No step is mandatory, no sequence is guaranteed. People enter and exit, jump back and forth. And every behavior generates data, signals, expectations. The funnel isn’t dead, but it’s no longer a funnel. It’s an open system.

People who shop online don’t follow a set path; they go through a cycle. They come back to the site, compare, read, get distracted, come back, add stuff to their cart, and then forget about it. Or they finish their purchase in five minutes. No step is linear. No user is identical to another. And if you really want to intercept them, you have to think of the funnel as an intelligent circuit, not as a ladder.

The most suitable model is the circular one: each phase feeds the others. Those who have bought can inspire new visitors, those who explore can come back later, those who have abandoned their cart can get back into the game if they find the right content. Today, the funnel is not an obstacle course, but an ecosystem of possibilities.

Messy middle, flywheel, and irregular paths

There is no longer a defined “before” and “after.” The heart of the funnel has become a tangle of actions that all happen at once: it is Google’s messy middle. The user searches, compares, reads, watches, asks, clicks, evaluates. And while doing so, they can change their mind a thousand times.

Some talk about the flywheel, where the customer is at the center and everything revolves around their experience. Others prefer to talk about micro-funnels, tailored to each behavior. But the essence remains the same: you don’t decide the path. The most you can do is design a system that can adapt, react, and remain useful at every point.

The funnel is still useful, but not as you remember it

Let’s not throw everything away. The funnel remains a useful tool for reading, analyzing, and improving performance. It helps you understand where you are losing visits, where interest is waning, and where content or offers are working. But to do this today, you need to build it in a flexible, up-to-date, and responsive way. Forget the idea of a map to follow step by step: it’s a reading structure, a way to analyze where you lose customers, where decisions get stuck, where you can act more effectively.

Each stage can be optimized: discovery can be more emotional, interest more informed, desire more engaging, action simpler.

It’s time to leave the rigid map behind and build a structure that’s better suited to the way people really move. Because it’s no longer the customer who follows the funnel: it’s you who has to follow them within the loop, where every point can also be a new beginning.

The new stages of the smart e-commerce funnel

Let’s set out to discover these four “new” fundamental and dynamic moments that every purchasing decision goes through today: inspiration, exploration, purchase, and experience.

These are the steps through which people discover, evaluate, buy, and return. There is no fixed order: these moments are triggered based on context, intention, and channel. And that’s where we need to be, useful and relevant.

Thinking about the funnel in these terms allows us to read the real behavior of users more clearly and build a strategy capable of accompanying them smoothly, without forcing them. This is how an intelligent funnel works in 2025: it follows people, adapts, and returns value at every step.

Inspiration: the moment when the idea is born, not when the search begins

At this stage, you don’t have to “be found,” but rather trigger desire, even if only accidentally. It means being present and engaging in the digital places where inspiration arises spontaneously, often when the user is not actively looking to buy. And today, the moments when the spark ignites don’t happen on Google, but on TikTok, Instagram, YouTube Shorts—these are the new challenges of e-commerce! Social commerce has made entertainment an integral part of the purchasing decision. All it takes is a well-edited video, an authentic review, or a live stream with the right product at the right time to convert a casual scroll into a sale.

Platforms such as TikTok Shop and Instagram Shopping are driving this evolution, offering an integrated, visual, direct shopping experience. Live shopping, in particular, is becoming a powerful tool for generating urgency and real-time interaction.

At the same time, technologies such as Google Lens and Circle to Search are changing the very origin of queries: search starts with an image, not a keyword. According to Google, over 10 billion visual searches are performed every month with Google Lens, and this trend is set to explode. The user sees something that catches their eye, circles it, and finds it. This means that image catalog optimization—often treated as an afterthought—is now key: high-quality images, structured data, and semantic consistency make the difference between being recommended or remaining invisible.

Vertical video content also plays a key role in this phase and is the best format for showing a product in action, overcoming the limitations of a static photo. Unboxing, tutorials, authentic reviews, and short “how-to” videos increase familiarity with the brand and build trust in a matter of seconds. You don’t need a Hollywood production: all you need is a smartphone and a clear idea. Link these videos to your catalog on Google Merchant Center and to the product listings on your website. The metrics to keep an eye on at this stage are reach, engagement rate, video views, and social referral traffic.

What you need to turn attention into desire

To really work on the inspiration phase, you need four things: a consistent visual strategy, content designed for the platform, a narrative system that creates connections, and authentic people to tell the story.

1. Choose where you want to inspire and who you want to inspire

TikTok Shop and Instagram Shopping work great for emotional products, lifestyle, fashion, accessories, beauty, and food. But it’s not automatic: they require native content, consistency, and an active presence. If your brand has a more technical tone or targets a B2B audience, you might work better with short videos on YouTube, LinkedIn, or even Pinterest, focusing on tutorials, user experiences, and application scenes.

2. Create a content system, not a viral post every now and then

Inspiration does not come from a single piece of content, but from consistent exposure over time. An effective approach is to build series of content:

• Mini-pills of 15 seconds showing scenes of use
• Informal testimonials from real customers or creators
• “Routine” or “kit” videos showing product combinations
• Visual micro-narratives (“what’s in my bag,” “a day with…”)

This content can be distributed organically or sponsored, but it must be designed to trigger identification or curiosity, not to explain the product.

3. Build a flow, not isolated content

Every piece of inspirational content can—and should—lead to exploratory content. Here are three ways to do this:

• Include a product linked directly to the product page in videos, if the platform allows it (TikTok Shop, Instagram Shopping);
• Use smooth, non-promotional calls to action (“find out more in the link in bio” still makes sense, if done well);
• Link content in series, with episodic narratives or “episodes” that refer back to the same brand/product/context.

4. Engage the right people

You don’t always need an influencer with millions of followers. Often, all you need is

• vertical creators with small but active communities;
• real customers willing to create content in exchange for products;
• faces within the brand, if well managed (founders, employees, ambassadors).

The important thing is that the content doesn’t seem contrived. It works when it’s natural, spontaneous, imperfect. The opposite of advertising.

The ultimate goal: to be remembered, not recognized

True inspiration works when someone sees your product three days later and unconsciously connects it to that content they saw by chance. If that happens, you’ve hit the mark.

Exploration: where choices are made and outcomes are often decided

Here we enter the territory of the “messy middle,” the organized chaos in which users read information, evaluate, compare, validate, and seek reassurance. Your goal is not just to appear: it is to be the most useful and convincing answer at the most uncertain moment of the journey.

Google has made this moment even more complex—and interesting—with the introduction of AI Overview, which are automatically generated responses from artificial intelligence that mix information, links, and suggestions into a single summary. Appearing in these spaces is not easy, but it is not impossible: you need up-to-date, correctly formatted, semantically consistent content supported by structured data. It is a more demanding form of SEO – or rather, GEO – but also more strategic: getting into integrated shopping carousels or among the sources cited by Gemini can bring huge results.

At the same time, chatbots are becoming much more than a support window. The most advanced ones, based on LLM (Large Language Models), are capable of guiding users through comparisons, answering complex questions about products, and suggesting options based on expressed preferences. They don’t filter: they converse and guide users toward the right product, acting as a true virtual salesperson.

At this stage, video reviews and user-generated content also play a decisive role. Users don’t just trust the brand, they trust the real experiences of those who have already purchased. And comparison sites – also powered by generative AI tools – become crucial touchpoints.

The metrics to monitor at this stage are the percentage of “Add to Cart,” the average time spent on the product page, the use of wish lists, and the number of questions asked to the chatbot.

What it takes to turn curiosity into choice

When users start exploring, they’re not looking for confirmation: they’re looking for reasons to exclude you. Every uncertainty, every friction, every piece of vague information brings them closer to another option. The goal here is to simplify the choice, not push it.

1. Make your products comparable – even with each other

Don’t wait for the user to do it. If you have similar variants or different models of the same product, make the real differences clear. Tables, grids, FAQs, comparison videos: everything helps prevent them from leaving your site to seek clarity elsewhere.

2. Show the product in action, in multiple contexts

Not everyone has the same question in mind. Some want to know if it’s compatible with X, others are looking for a specific use, and others want to understand how someone else has used it. The best answers come from contextual micro-videos, visual testimonials, and concrete use cases. A few well-structured seconds are all it takes.

3. Align content and channels

If the user searches on Google, they find in-depth content. If they come from YouTube, they need to find a consistent experience. If they switch to a comparison site, they need to find the same price, the same tone of voice, and the same credibility. Any inconsistency undermines trust.

4. Make information support seem like part of the process, not an afterthought

A chatbot that just offers links or responds with canned phrases wastes time. One that asks “what will you be using it for?” or “do you prefer more autonomy or lightness?” can really guide the choice.

Don’t ask for trust yet. Give confirmation

At this stage, the brand doesn’t have to convince: it has to reassure. This is achieved through detailed reviews, answers to common questions, guarantee badges, and public comparative tests. Any element that cannot be verified leaves room for doubt. And at this stage, doubt means escape.

Purchase: if everything has worked, this should be simple

The user has made their decision. You no longer need special effects, but rather meticulous work to eliminate friction. Now every millisecond of delay, every extra field to fill in, every uncertainty is a potential lost customer. There is only one goal: to make payment an instant, frictionless action.

This is where you measure the distance between a funnel that drives traffic and one that generates revenue.

The moment of action is not just checkout: it is everything that precedes and accompanies it.

A slow, confusing, or distracting process can undermine even the best of inspirational strategies. Conversely, a simple and consistent experience can close the sale without the need to push.

The key metrics for this phase are the classic ones: conversion rate, average order value (AOV), and cart abandonment rate.

What it takes to turn intention into action

When the user is ready to buy, you need to remove every possible barrier, not add new ones.

1. Checkout: one page, no surprises.

Conversions are in the details. A form that is too long, a mandatory registration requirement, a missing payment method, or shipping costs that are not disclosed until too late are all common reasons for abandonment. The ideal flow is:

• single-page checkout
• guest mode always available
• fields reduced to the essentials
• expected and up-to-date payment methods (card, wallet, BNPL),
• and, if possible, a clear estimate of time and costs before the user even looks for them.

Implement a One-Page Checkout, offer Guest Checkout (purchase without registration), and integrate diverse payment methods, including BNPL (Buy Now Pay Later) services such as Klarna or Scalapay, which according to various studies can increase conversion rates by up to 20-30%.

2. Immediately visible trust elements

Security badges, summary reviews visible during checkout, return guarantee, or active real-time support. No need to add text: just make them visible at the right time. It’s a silent confirmation that reduces perceived risk.

3. Purchase support campaigns, not just remarketing

Purchases can also be stimulated before the decisive click. Google’s Performance Max campaigns are an example of this: they use AI to automatically intercept users with a high propensity to purchase, distributing creative content across the entire Google inventory (Search, YouTube, Gmail, Display, Discover). The interesting thing is that assets can be generated automatically in a matter of seconds, thanks to generative artificial intelligence. But the real advantage is being able to create customized assets automatically, based on real data and consistent audience segments. No complex graphics are needed: all you need is a clear message, the right product, and a CTA consistent with the intent. You can test a huge amount of creative and let the algorithm find the winning combination.

4. An abandoned cart is an interrupted conversation, not a defeat

This is where intelligent automation comes into play. It’s not enough to send an “You forgot something” email. You need personalized content based on:

• what they left in their cart (and how long ago)
• what they viewed immediately before
• what incentive might reactivate their interest (targeted discount, alternative product, testimonial) and allow you to recover the cart.

A follow-up message that takes these elements into account reopens the funnel without disrupting the experience. For example, if a product is out of stock, AI can suggest a similar alternative. If the cart exceeded a certain threshold, it can offer free shipping. It should feel like a logical continuation, not an attempt at recovery.

Experience: the most overlooked (and most profitable) part

Did you sell? Great, but the funnel and your work are not over. Those who bought can become promoters, repeat customers, or—if neglected—a silent loss. The post-sale phase is often the one that receives the least attention, but it is here that business sustainability is built. It is the principle of customer retention: a returning customer costs up to 5 times less than a new one and spends more on average.

The problem is that this phase is often ignored. Everyone is obsessed with traffic and conversion, but very few invest energy in building value over time. And that’s a shame: a funnel that closes on the cash register is sterile, one that reopens relationships generates margin.

The key metrics here are Customer Lifetime Value (CLV), repurchase rate, and Net Promoter Score (NPS). The point is not to “keep selling.” It’s to build trust, habit, and belonging. If you succeed, you don’t have to start from scratch every time.

What it takes to turn a customer into a promoter

A satisfied customer is an asset. An enthusiastic one is a multiplier. But you can only get there if you play your cards right at this stage, working on four levels: intelligent automation, relationship management, and encouraging participation and sharing.

1. Intelligent automation that talks like a person

Your customer data is gold if you use it wisely. If a customer has just bought a DSLR camera, don’t send them a discount on another one. Send them content on the best lens for beginners. If they don’t return to the site after 45 days, try an incentive that makes sense to them, not to everyone. Want more? AI can analyze data to predict the future! Choose a tool that integrates predictive AI capabilities to segment your audience dynamically and automatically: implement churn prediction models (to identify customers at risk of leaving and take action), cross-selling models (to suggest complementary products), and next-purchase prediction models (to anticipate the next purchase and send an offer at the right time).

2. Customer care that arrives before the problem

Don’t wait for an angry email. If you see a negative review, if a negative mention appears, if you receive multiple tickets about a product: take action immediately. You can use sentiment analysis, automatic alerts, and proactive flows to contact those who need you before they ask.

3. Loyalty programs that don’t look like supermarket cards

Classic loyalty programs are giving way to tailored experiences. You don’t need 10 points for $1. But

• early access to exclusive products
• restricted content
• preview invitations to events or promotions
• targeted surprises for frequent buyers or those who haven’t bought in a while.

Those who feel part of something stay, recommend, participate.

4. UGC not as a favor, but as part of the game

Don’t expect reviews if you don’t ask for them. They won’t come spontaneously if you don’t make it easy to give them. And above all: show the ones you get. Real photos, videos of people unboxing products, and honest comments do much more than any campaign. Those who create content for you are not just helping you. They are fueling the inspiration phase of other users. The circle closes, the funnel turns into a loop. And at that point, it really works.

What you (really) need to make every stage of the funnel work

Models and strategies are useful, but then you have to get your hands dirty.

Every stage of the funnel can be enhanced with tools, content, and operational choices that make a difference: here are some concrete levers that help turn attention into action, and the first conversion into a stable relationship.

• Read behavior with heatmaps

Analytics tell you a lot, but not everything. To really understand what works (and what doesn’t), you need to observe how people move around your site, where they click, what they ignore, and where they get stuck. Heatmaps are a valuable tool for finding out if a CTA is invisible, if a description isn’t being read, or if an image isn’t generating any interaction. They are also useful for identifying less visible problems, such as unclear navigation, slow loading, or promotional banners that no one notices. Looking at this data with a critical eye allows you to take precise action: revise a layout, simplify a text, or optimize a product page.

• Feed the funnel with useful content (not just promotional content)

Starting a blog and filling it with articles without direction is useless. But a strategic blog can be a powerful ally. People are looking for answers, examples, comparisons. And well-crafted content can be the first point of contact, or the decisive step that leads to conversion. You don’t need “company news”: practical guides, product comparisons, real case studies, and honest answers to frequently asked questions are better. If the content is useful, traffic will come. And if readers find value, they will be more likely to trust (and buy).

• Build ads that speak to the right people

It’s not enough to “do advertising”; you have to do it well. The first step is to really know who we want to reach, the second is to segment and personalize. Demographic targets, dynamic retargeting, similar audiences: today’s platforms offer sophisticated tools to intercept interests, behaviors, and intentions. But they only work if the message is clear, the format is consistent, and the landing page is up to par. Ads need to be tested, measured, and adapted. And metrics aren’t limited to clicks: time spent on the page, actions taken, and the next steps matter too.

• Strengthen trust with social proof

If one product has a hundred positive reviews and a similar one has none, guess which one generates more trust? Other users’ opinions influence every stage of the purchasing journey. Showing feedback, quoting testimonials, highlighting social mentions or sales figures is not just aesthetics: it is building trust in real time. Consistency is key here too: reviews must be collected, moderated, and valued. And it’s not enough to just “have” them: they need to be visible in the right places on the site, at the right times.

• Give them a reason to come back (and stay)

Building loyalty isn’t just about “collecting points.” A good loyalty program is based on simple, clear incentives that are perceived as beneficial. Exclusive discounts, reserved content, early access, or personalized rewards work when they’re part of a smooth, engaging system. Even a touch of gamification—such as badges, progress, and levels—can turn a one-time purchase into a lasting relationship. But be careful: the program must be well known, easy to use, and integrated into the experience. If customers don’t know it exists or don’t understand how to use it, it’s a waste of time.

The key role of first-party data

This complex and intelligent loop can only work if it is fueled by the right fuel: first-party data. In the post-third-party cookie era, the data that customers consciously choose to share with you is the most valuable asset you have.

Today, success is not about who captures the most audience, but who knows how to collect, organize, and use their data in the right way. Every click, purchase, email open, and product view is valuable information that can fuel all stages of the loop: inspiration, exploration, purchase, and experience. The competitive advantage is no longer “having access to data”—too many brands collect information that then sits idle. It’s knowing how to turn it into action, connect the dots, and turn a messy mass of events into active, actionable knowledge. And that requires strategy.

How to build a data-driven strategy

In short, you shouldn’t aim to have “lots of data,” but rather know which tools to use, for what purposes, and in what sequence. This is where three key technologies come into play. When integrated correctly, they can transform the way you build and manage your eCommerce funnel: GA4, CRM, and CDP.

Google Analytics 4 (GA4) is the tool for monitoring onsite behavior. It tells you what users do on your site: where they come from, which pages they visit, how long they stay, and whether they take meaningful actions (add to cart, fill out a form, make a purchase). But GA4 is not enough: it gives you events and sessions, not reports.

That’s why you need a CRM, the tool that allows you to associate those behaviors with real people, with a complete history of interactions. In CRM, you can manage contacts, segment them, personalize communications, see who has been a customer for the longest time, who has high potential but is inactive, who has abandoned their cart three times without converting. It is the relationship environment, where every contact has a face, a story, a temperature.

Finally, the CDP (Customer Data Platform) is what connects everything. It is the technology that collects data from all sources (GA4, CRM, email, apps, social media, chatbots, customer care, etc.) and unifies it into a single, up-to-date customer profile. It does not work in silos, but creates a cross-sectional view. Thanks to the CDP, you can know that a user has visited the product page three times, opened the last email but didn’t click, left a positive review last year, and today has an open ticket with support.

This type of insight is the only possible basis for a truly personalized, responsive, scalable funnel.

From collection to value: what to do with the data once you have it

Once collected and integrated, data must generate action.

If you know that a user often returns to a category but does not buy, you can understand that they have an unmet need. If a customer buys every 40 days like clockwork and has not done so for 60 days, you can intervene before they go elsewhere. If a segment clicks on emails but does not convert, perhaps the tone, timing, or proposal needs to be reviewed.

This is where Artificial Intelligence comes into play: not to replace marketing, but to make it faster, more targeted, and more sustainable. It analyzes patterns, predicts behavior, and suggests responses. You remain at the center, but you have a co-pilot working while you sleep.

The point is not to know that a customer bought something two months ago: the point is to decide what to say to them today, based on that behavior. And in what tone. And on which channel. The ultimate goal is ambitious, but now necessary: to offer personalization on a large scale, to make every customer feel seen, heard, and recognized.

Not with an email with their name in it, but with experiences, products, content, messages, and offers that truly resonate with what that person is looking for, at that precise moment.

A well-designed system allows you to:

• predict a user’s next purchase based on their history
• catch signs of disaffection before they turn into churn
• create dynamic messages in real time
• adapt ad creatives based on data
• personalize the site for each segment.

This is where marketing automation stops being a static flow and becomes a living ecosystem that adapts based on what each user does or doesn’t do. And if you really want to make a difference, Artificial Intelligence gives you that extra edge: identifying hidden patterns, suggesting more precise segmentations, activating behavioral triggers that you wouldn’t have predicted manually.

Stagnant funnel? It’s time to change your approach

If your e-commerce funnel is spinning but not taking off, the problem may not be in the structure… but in the tactics. The context has changed, users are more impatient, more distracted, more accustomed to deciding elsewhere. So you need a real overhaul, made up of smart choices, not minor tweaks, starting with these solutions.

Write content that sells!

With AI Writer, you can create customized texts for every stage of the customer journey: inspiration, comparison, decision.

Find out more

• Cross-selling, but done good

Showing “related products” is no longer enough. True cross-selling works when it manages to anticipate desire, not chase it. Amazon makes billions from it. You can start with something simple: don’t just recommend similar products, but useful, real, contextual suggestions. And if you want to take it to the next level, there are tools that create automatic recommendations based on behavior and turn them into carousels, micro-stories, and visual formats. Less “product grid,” more experiences.

• Customer care within the funnel

People who get stuck in the buying process often have just one question. And if they don’t find a quick answer, they disappear. Customer support can’t stay on the sidelines: it has to get in the game, talk to those who need clarification, perhaps using personalized emails or advanced chatbots to reactivate interest and bring back those who gave up halfway. It doesn’t take much: a recognizable signature, a direct link, a human communication. And the customer is back.

• Monitor searches for your brand

Do you know where many people end up after seeing a product? On Google, looking for discount codes, reviews, alternatives. So monitor those queries: branded keywords are part of the funnel, even if they happen “outside the home.” Optimized content, targeted landing pages, strategic FAQs: you need to be there, not a third-party site. And yes, the copy must speak like a good salesperson would: informed, reassuring, convincing.

• Copy that guides (not just sells)

Writing “well” is not enough. You have to write for intent, understand what the user is really looking for, and take them where they need to go. Semantic search is for this: deciphering expectations, not just keywords. Strategic copy like this not only converts better, but ranks better. And when it works, it does both.

• Look at the funnel, really

One last tip: look at what’s happening in the funnel. Where are people leaving? Where are they getting stuck? Google Analytics (and GA4 in particular) tells you everything, if you know how to read it. Data isn’t just for reporting; it’s for making operational decisions. Change a CTA, fix a passage, resolve a bottleneck. That’s where improvement comes from.

Mistakes to avoid in a modern funnel

Positive things are counterbalanced by obstacles and mistakes, that much is clear. Obviously, every funnel has a weak point—and it’s almost always where you don’t expect it. Even a well-designed strategy can fail if it remains stuck in old patterns, loses touch with the customer, or wastes energy on content that no one really needs. Here are the most common mistakes we see every day—and the ones you should avoid so you don’t waste time, traffic, and potential.

1. Still thinking in a linear way

The idea of an orderly path, where the user discovers a product, learns about it, then evaluates it and finally buys it, no longer holds water. Customers enter and exit the funnel when they want, and often make logical leaps that you can’t predict: they discover a product in a live stream on TikTok, compare it on an external site, read reviews on Reddit, then search for you on Google to see if you have a loyalty program. Those who set their strategy based on fixed sequences miss the opportunity to be useful at the right moment, which can come from any direction.

2. Segmenting poorly (or out of habit)

Using only demographic data or pre-packaged segments is like trying to fish in a lake with a leaky net. Segmenting by age, gender, or device may still make sense in certain contexts, but today you need much more: behavior, frequency, value, intent, and journey. Each segment should be a lever, not a category. Above all, it should evolve over time, just like the people who make it up.

3. Don’t follow the customer after the purchase

A customer is not “done” after they pay. In fact, that’s where the real value-generating part begins: repurchasing, reviewing, sharing, word of mouth, loyalty. Yet many brands completely ignore this phase. No follow-up, no dedicated content, no after-sales attention. Just silence. A modern funnel cannot afford this oversight. Without a positive post-purchase experience, the loop is broken. And starting from scratch every time is the most expensive way to grow.

4. Building generic content instead of precise answers

Content marketing does not mean publishing random articles, newsletters, or social media posts. It means being the useful answer to a real question, at the right time, in the right format, with the right tone. Generic, filler content that is disconnected from the customer journey does not drive qualified traffic, educate, or convert. You need content that anticipates objections, clarifies doubts, and shows the product in the context in which it will be used. Content “made to please Google” is no longer enough: it must appeal to readers and keep them moving forward on their journey.

E-commerce funnel: FAQs and key questions to answer

The shift from a linear funnel to a smart loop is not just a terminology update, it is a paradigm shift. It requires us to stop thinking about isolated channels and start designing a seamless, interconnected, and personalized customer experience.

Technology (AI, Google, TikTok) is not an end in itself, but a means to make this loop more efficient and satisfying. The next time you analyze your marketing strategy, don’t ask yourself, “How do I get customers into my funnel?” Ask yourself, “Is my loop smart, fast, and rewarding enough to convince customers to stay and bring others with them?”

The answer to this question will determine tomorrow’s e-commerce winners. And speaking of questions and doubts, here is a list of FAQs that may help you clarify the last gray areas.

1. What is an e-commerce funnel?

It is the journey—not always linear—that a person takes from their first contact with your brand to their purchase and beyond. It is not a simple sequence of rigid steps, but a useful model for understanding where you are losing customers, where you can intervene, and how to improve the overall experience. Today, there are four stages to consider: inspiration, exploration, purchase, and experience.

2. Does the AIDA funnel still work for an e-commerce site?

As a theoretical model, it still has its merits. But in practice, no: it is no longer enough. The AIDA model – Awareness, Interest, Desire, Action – was created for a much simpler context, where the customer followed a fairly orderly path. Today, users move around chaotically, go back, skip steps, use multiple channels, and are influenced by external sources. A more fluid vision is needed, such as that of the flywheel or the smart loop.

3. How do you optimize an online sales funnel?

You work on each stage with specific objectives, avoiding universal solutions. In the inspiration phase, you need to be present where interest arises: social media, videos, content. During exploration, the site must offer clear information, reassurances, and reviews. The purchase must be as simple and quick as possible, while the after-sales experience is what fuels loyalty and word of mouth.

4. What metrics should be used to evaluate the e-commerce funnel?

There is no single metric that tells the whole story. Each stage of the funnel has its own indicators. In the exploration phase, look at how long people stay on product pages or how many interactions occur with the chatbot. In the purchase phase, the number of items added to the cart and the conversion rate are important. After the sale, repurchases and customer lifetime value say a lot about your ability to retain value.

5. What is the difference between a sales funnel and a marketing funnel?

The marketing funnel is primarily concerned with the acquisition and nurturing of the contact, i.e., taking them from the first click to interest. The sales funnel, on the other hand, comes into play when the customer is closer to a decision and needs to be guided through the purchase. In e-commerce, these two areas overlap and often work together to provide continuity throughout the customer journey.

6. How can AI be used to improve the e-commerce funnel?

Artificial intelligence can be the invisible engine that makes every stage more precise and effective. From content personalization to ad optimization, predictive behavior analysis to tailored messaging, AI helps you be faster, more relevant, and more consistent. When well integrated, it acts as a value multiplier throughout the funnel.