I started my first ecommerce brand, Travelization, in 2017.
We sold travel-inspired home decor, but I quickly realized people were not really buying decor.
They were buying a story.
Adventure. Global citizenship. The feeling of being in a tribe who explored the world.
At the time, most Amazon sellers were obsessed with hacking the algorithm.
Keywords. Ranking hacks. Rebates. Incentivized reviews. Fast launches. Some were using black-hat tactics because Amazon had not figured out how to penalize them yet.
I went in a different direction.
I spent a lot of time in Facebook travel communities talking with the exact people I wanted as customers. I shared travel stories, advice, my own experience building a business and things I was learning along the way.
Most of the time, I was not selling anything.
Maybe one in ten posts mentioned a product. Mainly asking for feedback.
And it worked.
Not because I found some Facebook hack.
Because people care about the story and the people behind a brand much more than most sellers think.
You do not just sell a product. You sell a problem solved, a set of values and a story someone wants to be part of.
Then things went very wrong
My first year was not some clean founder success story.
About six months in, I had a massive inventory quality issue.
It got bad enough that I sold my own home to keep the business alive.
That period changed me.
It taught me that doing more work is not always the answer.
Building a system that actually works is worth much more.
If you understand your target customer better than the competition, obsess over becoming the best solution to their problem, speak their language, and genuinely earn their trust, you create something much stronger than a listing.
I fixed the business, scaled it and successfully exited Travelization less than three years after starting it.
But I kept coming back to the same question:
Why are ecommerce brands guessing what customers want?
From guessing to actually asking customers
That question eventually led me to co-found ProductPinion.
Amazon gives sellers a lot of data, but very little direct communication with customers.
So sellers would debate endlessly:
Which image is better?
Which message is clearer?
Which packaging will convert?
Which product should we launch?
The solution seemed simple.
Why are sellers guessing what converts instead of actually asking customers?
We launched the early beta of ProductPinion to my Ecom Brand Builder Mastermind and started building around that idea.
Put real customers back into the decision-making process.
It was the same lesson again.
The closer you are to your customers, the better decisions you make.
Then I saw what happens at scale
Later, working with Amazon aggregator Gravitiq, I helped operate five brands.
The ones that grew best usually had two things going for them.
-
A strong product-market fit.
-
And decisions backed by customer data, not just opinions in a meeting.
But I also started seeing something else.
Amazon loved demand coming from outside Amazon.
Creators.
Digital media.
PR.
Communities.
For Grace & Stella, I started playing with Creator Connections and developed a broader Brand Ambassador Program. In its first year, it contributed roughly over $5MIL to our annual revenue.
But the revenue was only part of the story.
Creator momentum supported media momentum.
That helped lead to opportunities like a live TODAY Show appearance on Prime Day, when the business crossed its first $1 million day.
Alongside strong paid advertising, sponsored media and great work across the team, that momentum also supported expansion into major retailers including Costco and Target.
This was when my thinking really changed.
Creators and PR were not just traffic channels.
They were building a network of third-party evidence around the brand.
Real people talking about it.
Publishers writing about it.
Customers validating it.
Retailers noticing it.
The brand was no longer the only one telling its own story.
The market was telling it too.
Then AI changed where the decision happens
Later, while consulting major CPG companies, including Fortune 500 businesses, on AI visibility, something clicked.
The biggest brands stay big partly because they keep adapting.
They do not wait until a new channel is obvious to everyone.
They test early.
They measure.
They learn.
And they keep moving.
Today, that new environment is AI.
I saw it clearly while working with a supplement brand whose product, in my view, had meaningful advantages over much more established competitors.
But having the better product did not mean AI recommended it.
So we started combining what I had learned across all the previous chapters.
The Quickfire Ecommerce Roundup - May 2026 🚀
Quickfire - Shopify Platinum Partner
2 months ago
](https://www.linkedin.com/pulse/quickfire-ecommerce-roundup-may-2026-quickfire-digital-uk-shnpe)
Before you build a website, read this. Johan van Noordwyk 6 months ago
Are Micro-products the Future of Ecommerce? Ezra Firestone 7 years ago
Amazon customer data.
Alexa for Shopping questions (then Rufus).
Reviews.
Reddit conversations.
Creator content.
YouTube.
Articles.
PR.
Product positioning.
Instead of giving creators generic briefs, we started aligning their messaging with the questions customers kept asking and the product advantages we actually needed the market to understand.
We asked for comparison videos on YouTube. We asked for blogs.
Then we measured again.
Alexa for Shopping visibility started moving.
ChatGPT visibility started moving.
And suddenly something that sounded abstract, “AI visibility”, became very real.
The goal is not to manipulate ChatGPT.
It is to build a brand that ChatGPT, Alexa for Shopping, Walmart Sparky and other AI systems have good reasons to recommend.
I have seen this movie before
Building a Facebook community worked incredibly well for my brand in 2017.
I would not build the same strategy today.
Keyword stuffing on Amazon worked once too.
It does not anymore.
Platforms change.
Algorithms change.
Customer behavior changes.
And brands that adapt early can build an advantage before everyone else turns the opportunity into another crowded marketing channel.
That is how I think about AI visibility today.
It is still early.
But we are already seeing measurable results when brands treat it as a long-term Brand Authority system, not another SEO trick.
And building that authority is not simple.
It is part data, part brand strategy and part creative experimentation.
You map the questions customers are asking.
You see which competitors AI recommends.
You understand the gaps.
You improve the evidence around the brand.
You measure again.
And you keep learning.
Customer-centricity has changed
The definition of being customer-centric changes with the way customers shop.
In 2017, it often meant understanding what someone searched for and becoming the best result for that search (usually by hacking the system).
Today, increasingly, it means:
Being the brand recommended when a customer asks AI what they should buy.
I think this matters especially for consumable categories like supplements, beauty, pet and health products.
These businesses can have incredible lifetime value.
If AI helps someone make the first purchase with confidence, and their own experience proves that recommendation right, you may have earned much more than a conversion.
You may have earned a long-term customer.
That is why ARTAN AI is primarily being built for established, Amazon-first consumable brands, typically $10M+ businesses, that use Amazon for scale while also growing Shopify, retail and a long-term brand trust beyond any single marketplace.
I am also starting to hear something interesting from the retail side.
Buyers themselves are checking what ChatGPT says about brands they are evaluating.
That makes sense to me.
If creators, customers, publishers, marketplaces and AI systems consistently validate a brand, getting onto a physical shelf can become another outcome of the same underlying authority.
Why I built ARTAN AI
Looking back, ARTAN AI does not feel like a sudden pivot into AI.
It feels like the next chapter of a problem I have been trying to solve for almost a decade.
Travelization taught me to listen to the customer.
ProductPinion taught me to systematically measure what customers actually prefer.
Gravitiq taught me how customer data, creators, media and external demand compound when you operate brands at scale.
Working with major CPG companies taught me how seriously the best businesses take new technology and changing customer behavior.
And AI has changed where more of those customer decisions are starting to happen.
So the principle behind ARTAN AI is simple:
I do not want brands optimizing for AI. I want them building brands AI has good reasons to recommend.
AI visibility is the outcome.
Brand authority is the asset.
The goal is not to trick an algorithm into mentioning you.
The goal is to become the answer.
The platforms will change again. They always do.
The brands that win will still be the ones that understand their customers better, build the best solution for them and adapt fastest to where those customers go to make decisions.
In 2017, for me, that place was Facebook communities and Amazon search.
Today it is increasingly a system covering digital media, influencers, publishers, ChatGPT, Google AI Overviews, Amazon Alexa for Shopping, Walmart Sparky and other AI shopping assistants.
5-10 years from now, it will probably be something else.
That is fine.
The technology changes.
Being worthy of the recommendation does not.
If you are building an established Amazon-first consumer brand and want to understand where AI recommends you, where competitors beat you and what you can actually do about it, reach out.
I would be happy to compare notes and see if ARTAN AI can help.