Hai Mag and the trouble with selling more

A store can have a wonderful month and leave its owner poorer. Orders arrive, the sales chart climbs, and somewhere behind the cheerful dashboard, advertising, discounts and inventory costs quietly divide the proceeds. Hai Mag has built much of his working argument around that awkward possibility. More business is pleasant. Keeping enough of it to carry on is rather more useful.

Before Eva, Mag was a vice president at Oracle with responsibility for a business exceeding $1 billion. He had worked inside organizations capable of moving technology across countries and continents. Then he turned toward the owner trying to work out what to charge for an item on Amazon. The distance between those jobs looks enormous until you consider their common problem: making decisions when the relevant information is scattered.

The bigger job before the smaller business

Mag was born and raised in Izmir, Turkey. His career took him to Germany and Accenture, then to IBM in 2008 and Oracle in 2010. These were businesses selling systems to other businesses, where technology had to accommodate organizations with their own habits, reporting lines and competing demands. His work eventually included managing regions for Oracle.

In 2018, he moved to the San Francisco area in search of another challenge. It was a substantial change of setting after years in established technology companies. A senior corporate role supplies a certain amount of structure along with its responsibilities. Starting something requires deciding which problem deserves the structure in the first place.

“I wanted to do something new,” he recalled. The sentence is modest enough to fit almost any career change. The particulars made his consequential. He met Barry Guze, who already had years of experience running an Amazon business. Mag brought the enterprise technology background; Guze brought familiarity with the marketplace’s daily demands. Together they began examining what brands were missing.

That meeting matters more to Eva’s origin than the familiar idea of a founder having a sudden flash of inspiration. Here were two sets of experience being applied to a shared question. What could a seller reasonably be expected to manage, and what might software help the seller understand?

Three decisions, one balance sheet

Mag described their early thinking through three domains: inventory management, advertising management, and pricing and promotions. Each sounds like a department or a software category. To a seller, each is also a decision about the same limited pool of money. Buy stock, attract a buyer, choose a price. Then do it again before the money runs out.

Consider the ordinary friction between them. Advertising may bring an item to shoppers’ attention just as its available stock dwindles. A discount may speed up orders while reducing the amount available to pay for the next shipment. These are illustrations of the coordination problem, rather than accounts of a particular Eva customer. Their familiarity is the point.

Eva began with pricing intelligence and expanded into a combination of technology and operating services. Mag’s current role is Chairman and CEO. His company now works across marketplaces and direct commerce channels, with people managing accounts alongside the software. The original question has grown to accommodate more moving parts.

The diagram is simple because the relationships are easy to state. Operating them together is the work. For Mag, the attraction of artificial intelligence sits in that work: interpreting information quickly enough to inform decisions that would otherwise be made separately.

The price of a busy store

His interest in pricing also gives him a useful vantage point on the charms of a rising sales graph. A business can attract buyers by becoming cheaper. The difficult question is whether that change improves the business. Sales volume alone supplies an incomplete answer, however satisfying the notifications may sound.

When he wrote about Amazon strategy in January 2025, Mag urged sellers to connect advertising performance with margin. He also advocated configurable AI tools, giving people control over their campaign objectives and visibility into decisions. A product launch might call for an emphasis on awareness. Another product might need a firmer focus on immediate returns.

This is an interesting preference for someone running an AI business. The owner’s judgment remains inside the system. Automation needs instructions about what counts as success. Optimizing toward a poorly chosen goal can make an error more efficient, which is a distinctly modern way of creating extra work.

More orders make a good headline. The money left over keeps the business open.

The arithmetic behind Mag’s approach

His published pricing advice could also be surprisingly restrained: watch sales velocity and review prices weekly, rather than assume every team needs endless manual adjustments. The ambition is to make the decision useful and manageable. Complexity has a habit of presenting itself as sophistication, especially when someone is selling a dashboard.

A founder with a microphone

Mag’s work extends into conversations with other operators. He has been an interview guest discussing his career and a host asking specialists how their businesses work. The recurring subjects are revealing: paid advertising, organic demand, inventory, customer behavior and the practical limits of trying to do everything at once.

In one conversation, Chris Morrano joined him to discuss AI creative tools and the appeal of less polished, founder-led video. That subject carries a pleasing complication for a technology executive. Making more content becomes easier. Giving somebody a reason to care still calls for a person with something to say.

In another, Eva advertising leader Oaklyn Gale discussed how information from one platform can inform decisions on another. Mag’s questions move across the boundaries a company’s reporting structure might prefer to preserve. Customer behavior travels; the information about it ought to travel as well.

Earlier, his interview with Brent Zahradnik covered advertising alongside organic growth and the use of AI in managing commerce. Across these exchanges, Mag returns to the relationships between activities. The microphone gives him another way to examine the same problem that led to Eva.

Three channels, different jobs

By 2026, his argument had widened beyond the Amazon store. An interview with Will Christensen, recorded in January and published in October, explored when a brand should expand across Amazon, TikTok Shop and Shopify. Mag assigned the channels different roles: discovery, sales velocity and a closer understanding of customers.

That distinction changes the question facing a founder. Opening another storefront is an observable achievement. Knowing why it belongs in the business takes more thought. A shopper might discover something in one place, purchase in another and return through a third. The owner still has to pay for the whole journey.

Mag’s suggested sequence began with reaching break-even and growing from there, then looking for branded search as evidence that a dedicated store could convert existing interest. These were his judgments about expansion, rather than a promise that every brand could follow an identical timetable.

The conversation is consistent with his earlier work. At first, the connection was between price, promotion and stock. Now it also runs between platforms. Adding channels increases the number of decisions that need to agree. A founder can collect storefronts rather as someone else collects luggage, only to discover how much there is to carry.

In an August 2026 conversation with Danan Coleman, Mag again emphasized business fundamentals before increasing advertising spend. The setting and the vocabulary had broadened. The concern about paying for growth before understanding its economics remained recognizable.

What the business might pay for

There is another ambition attached to Mag’s commercial work. He has described wanting to develop AI infrastructure for children’s education, using the proceeds of a successful technology business for nonprofit purposes. He has also spoken of building Eva into a unicorn. These are stated aspirations, with a horizon beyond the next selling season.

The education goal gives the attention to profit a different context. Money retained inside a business can buy the next shipment or support the next hire. In Mag’s account of his longer ambition, it could eventually finance work outside commerce altogether. The connection depends on the commercial project succeeding first.

For now, he remains in San Diego, leading Eva and taking part in the recurring debate over how brands should grow online. His public LinkedIn handle, growthjedi, allows a little theatrical flourish into a career otherwise preoccupied with operational details. Even a discussion of inventory can make room for a lightsaber.

His professional profile lists English, German and Turkish, a small biographical detail that fits a career conducted across borders. The sequence of workplaces is international; the problem he now discusses is recognizable to anyone running a store. Stock must arrive, customers must find the product, and the price has to leave room for another day of trading.

What gives his story its shape is the decision to bring experience from large technology organizations to smaller sellers facing increasingly complicated choices. There is no shortage of ways to generate activity in ecommerce. Another campaign, another promotion, another channel will usually oblige. Mag keeps returning to what happens after the activity, when the seller counts what remains and decides whether the next step is affordable.