I started experimenting with dropshipping around 2016. By 2019, I had formed a company in Delaware and was operating it as a real U.S. business.

My first year was almost comically quiet.

If I add up the entire twelve months, I probably did only around $400–$500 in sales. But I kept the store alive, kept adding products, kept improving SEO, and kept learning how the machinery behind e-commerce actually works.

That turned out to matter far more than trying to find a magical “winning product.”

Over the following years I dealt with suppliers in China, AliExpress, Oberlo disappearing, CJdropshipping, U.S. warehousing, Shopify, Google Shopping, Facebook and Instagram Shopping, TikTok Shop, trademark problems, returns across continents, payment verification, IRS paperwork, and the uncomfortable reality that a platform can change one rule and erase a large traffic source overnight.

Eventually I reached a strange conclusion:

The hardest part of dropshipping was never putting a product on a website. It was everything that happened around the product.

That realization is ultimately why I built my own dropshipping SaaS in Türkiye.

My first real advantage was patience, not advertising

My original store was built with Shopify. I wasn't selling through an established marketplace, so there was no built-in audience waiting for me.

For a long time, almost nothing happened.

Instead of spending heavily on ads, I concentrated on SEO: improving product pages, adding new products, cleaning up the site structure, and making it easier for Google to understand the catalog.

Around 2020, shortly before the pandemic dramatically accelerated online shopping, I started seeing the result.

Organic traffic began growing. Orders started arriving more consistently. Google Shopping and other product-discovery surfaces began becoming meaningful sources of traffic.

The lesson I took from that period is still the same today: organic e-commerce compounds slowly.

A product page you publish today might do nothing for months. But a properly maintained catalog of hundreds or thousands of useful, indexable pages can become an asset that keeps working without paying for every click.

That is also why I still consider search infrastructure part of the store itself rather than something you “do later.”

AliExpress taught me that your supplier is really part of your company

Like many dropshippers at the time, I initially sourced products through AliExpress and used Oberlo to connect products to Shopify.

The setup looked wonderfully simple from the outside.

Customer orders → supplier ships → I keep the difference.

Reality was messier.

I eventually realized that some of the AliExpress sellers I was buying from were effectively dropshipping themselves. I wasn't buying from the source. I was buying from an intermediary who might be buying from another supplier.

Every additional layer created another place for something to go wrong.

Prices were higher than necessary. Quality could change without warning. Stock information was unreliable. Returns were painful. A supplier could happily ship something mediocre because my store, not theirs, was the name the customer remembered.

And returning a low-value product from the United States to China often made no economic sense at all.

This is when I started taking suppliers far more seriously.

Eventually I moved much of the operation toward CJdropshipping, where I could build a more consistent sourcing and fulfillment relationship rather than treating every AliExpress seller as an unknown variable.

The broader lesson is simple:

In dropshipping, outsourcing fulfillment does not outsource responsibility.

The customer bought from you.

Your supplier's mistake becomes your support ticket, your refund, your bad review and your reputation problem.

A U.S. warehouse solved more problems than I expected

Returns were one reason I eventually looked for a physical logistics partner in the United States.

I ended up working with PrepShipping in Georgia.

Today PrepShipping describes its operation as a U.S.-based fulfillment business offering warehousing, prep and pack, return management and last-mile-related services from Tucker, Georgia.

For me, having a U.S. logistics relationship did more than give customers somewhere sensible to return products.

It gave the business a real operational footprint.

That became particularly useful when dealing with payment and platform verification.

The Shopify Payments lesson I learned the hard way

At one point I wanted proper access to Shopify's U.S. payment ecosystem and Shop Pay.

Having a Delaware company on paper was not the whole story.

In my case, being able to document a genuine logistics relationship in the United States helped demonstrate that the company had real operations there. My agreement included the company information, my information and the operational address.

But this is an important distinction:

A 3PL agreement is not a magic Shopify Payments bypass.

Shopify's current U.S. requirements state that the business must provide an official physical U.S. business address where the business operations primarily occur. Shopify can also request EIN documentation and identity verification.

Shop Pay itself is closely connected to that payment infrastructure; Shopify documents it as an accelerated checkout available through Shopify Payments, with some third-party-gateway support under separate eligibility rules.

So the useful takeaway from my experience is not “rent any address and Shopify will approve you.”

It is this:

If you run a legitimate U.S. company remotely, build legitimate U.S. operations that you can document.

A real 3PL, warehouse or fulfillment relationship can be valuable operationally even before a platform asks you to prove anything.

Keep your EIN documentation somewhere safe

One surprisingly annoying document in U.S. business life is EIN verification.

Shopify itself lists documents such as the IRS SS-4 or 147C Letter as documents it might request if it cannot verify an EIN automatically.

Traditionally, one way to obtain EIN verification was to call the IRS Business and Specialty Tax Line at 800-829-4933. The IRS still lists that number for businesses that need help recovering or confirming an EIN.

The phone route I learned was essentially to choose English, navigate to the EIN section, select the option for an existing EIN, and tell the agent that you need an EIN Verification Letter / 147C. The agent verifies that you're authorized to receive the company's information. Menu options can change, so I would not treat a specific sequence of button presses as permanent.

If fax delivery is offered, ask about receiving it through a private, secure fax. A secure digital fax can also be much more practical than waiting for international mail, but confirm the current procedure with the agent.

There is now an even better option for some businesses.

As of 2026, the IRS says eligible users of its Business Tax Account can download an EIN verification notice called CP575, which can be used as a substitute for the older CP575 notice series and Letter 147C.

That's exactly the kind of tiny administrative detail that can save you hours when a payment provider suddenly asks for verification.

I used USFormation rather than trying to figure out every U.S. filing myself

When I established the company, I used USFormation.

For someone outside the United States, I found it far cleaner to have a formation service handle the process rather than trying to understand every state and federal requirement from scratch.

USFormation currently describes its services as including company formation, registered-agent service, federal tax registration support, banking documentation and ongoing compliance assistance. It also explicitly notes that EINs are ultimately issued by the IRS, not by USFormation itself.

For someone starting from outside the U.S., that's the type of assistance I would seriously consider again.

USFormation — the referral link I use

And regardless of which formation company you use, learn what filings your specific entity requires. Forming the LLC is the beginning of compliance, not the end of it.

Google Voice gave the company a practical U.S. phone layer

Another surprisingly useful piece of infrastructure was Google Voice.

Instead of maintaining a complicated traditional U.S. phone setup, I could make and receive calls from a computer or phone and handle SMS from the same environment.

Google's current Voice offerings include U.S. calling and U.S. texting, with unlimited domestic calling/texting available on applicable plans subject to Google's acceptable-use rules. Business editions integrate with Workspace as well.

For a remote U.S. operation, having a consistent business number makes dealing with suppliers, customers and services much cleaner.

The IP problem is much bigger than checking a logo

One of the ugliest lessons I learned came from intellectual-property disputes.

When sourcing products from enormous catalogs, especially from China, it is easy to assume that because thousands of sellers carry the same product, it must be safe to sell.

That assumption is dangerous.

I encountered situations where apparently ordinary products later attracted enforcement claims. Some disputes felt extremely opportunistic: a product would circulate through large supplier networks, hundreds or thousands of stores would list it, and only later would legal demands begin arriving.

Whether an individual claim is valid is a legal question, not something I would decide from an angry email.

But as a merchant, the practical lesson is straightforward:

Do clearance work before a product becomes successful, not after.

I regularly used the USPTO trademark search to investigate brand names and registered marks. The USPTO still provides its public Trademark Search system for this purpose.

There is another important distinction I wish more dropshippers understood:

Trademark search alone is not enough.

A trademark generally concerns source identifiers such as names and logos. A distinctive physical product design may instead involve a design patent, and artwork or graphics may raise copyright issues.

USPTO's Patent Public Search lets you search issued patents and published applications, and the USPTO specifically explains that design patents can protect the ornamental appearance of manufactured products.

So before committing serious traffic to a product, I would at minimum investigate the brand, the product name, distinctive logos and the product's unusual visual design. For something that is becoming commercially significant, professional legal clearance is cheap compared with discovering the problem after thousands of orders.

Don't build your entire company on somebody else's traffic switch

For years I benefited from product discovery through Google Shopping, Bing, Facebook and Instagram.

Then platform rules changed.

At one point Facebook-related shopping traffic that had been sending thousands of visits to the business dropped almost overnight. The company was American, the accounts were verified and the operation itself had not suddenly disappeared—but managing the business from outside the United States became a practical platform problem.

I spent a lot of time talking to support, testing different approaches and trying to understand what had happened.

Later I experimented heavily with TikTok Shop.

That experience was almost surreal.

Products began selling organically with no serious advertising spend. I remember waking up to the phone constantly firing order notifications. For a while it felt like I had stumbled into the most powerful organic commerce channel I had ever seen.

Then the problems started.

TikTok raised fulfillment concerns. I supplied tracking information and delivery evidence, including carrier proof showing packages reaching customers, but the dispute still did not end the way I expected.

Later I tried again with a completely different structure: a formal relationship with a U.S. furniture supplier, fast domestic fulfillment and local shipping.

Sales started again.

Then platform eligibility and location issues became a problem again.

Whatever anyone thinks about those individual decisions, the business lesson was much more important:

A marketplace can be an extraordinary sales channel, but it should never be your only asset.

Your domain, customer relationships, catalog data, supplier relationships, operational systems and search presence are assets you can control.

Marketplace distribution is rented land.

Search Console, Bing and Merchant Center became part of my basic setup

Whenever I launch an e-commerce site now, search-engine onboarding happens immediately.

I register the property in Google Search Console, submit the sitemap and make sure important pages are crawlable.

But there is a misconception worth correcting: submitting a URL does not mean Google will immediately index it, and it certainly does not guarantee that you will outrank a competitor.

Google explicitly says an indexing request does not guarantee inclusion and recommends submitting a sitemap when you have many new or updated pages.

The value is that you are making discovery easier and giving yourself visibility into crawling and indexing problems.

I do the same with Bing Webmaster Tools. Bing currently recommends IndexNow for automated notification when pages are added, changed or deleted. It even notes that repeatedly submitting the same URL does not accelerate indexing.

And for an e-commerce business, I consider Google Merchant Center equally fundamental.

Eligible products can appear through Google's free listings across surfaces such as Search, Shopping, Images, Lens and YouTube. That visibility is not guaranteed, but it gives a well-maintained product catalog another organic distribution layer.

Reviews matter, but not because they magically improve SEO

I also recommend establishing your business on Trustpilot early and collecting genuine reviews from real customers.

Not because opening a Trustpilot profile gives you some mysterious Google “trust score.”

It doesn't work that way.

The useful part is independent reputation data.

Google's own documentation lists Trustpilot among the supported independent review partners it can use as one of the sources for store ratings.

That is a much more defensible reason to collect reviews.

Good reviews also have a value no algorithm can fully summarize: when a new customer searches your company name before handing you their credit card, they find evidence that other humans actually received what they paid for.

Eventually I stopped asking how to survive these systems and started building one

By 2025 I had spent years dealing with the same categories of problems from different directions:

supplier reliability, stock changes, marketplace APIs, product imports, returns, tracking, invoices, customer support, platform verification and fulfillment.

At some point the obvious question became:

Why am I repeatedly solving these problems manually instead of turning what I learned into software?

That became Gymtayt Dropshipping.

I started building the platform in Türkiye toward the end of 2025 around a simple idea: let partners concentrate on actually selling while software and the supplier operation absorb as much repetitive work as possible.

The project has since grown into something much closer to an e-commerce operating system than the small connector it started as.

The current system handles marketplace/store connections, product imports, variant-level stock and price synchronization, orders, logistics and returns. The codebase also contains asynchronous workers and reconciliation flows for things such as fulfillment, invoicing and marketplace synchronization rather than assuming every external API call will work perfectly on the first try.

That last part matters.

Real SaaS integrations fail.

Webhooks arrive late. APIs time out. A marketplace accepts a product but rejects one variant. Stock changes while another job is running. A payment succeeds while an invoice provider is temporarily unavailable.

The answer is not another button.

The answer is queues, idempotency, retries, reconciliation and observable state.

That is the engineering knowledge I did not have when I uploaded my first AliExpress product years ago.

Today Gymtayt's public platform exposes catalog, stock, price, order and logistics workflows, while its integration layer covers Shopier, Trendyol, Shopify and ikas-related flows.

The Gymtayt Dropshipping app is also publicly listed in the ikas App Marketplace, where its listing describes product import plus automatic stock and price synchronization.

That is an important milestone for me because this didn't start as a theoretical SaaS idea.

It came from years of being the person on the other side of the screen when something went wrong.

If I were starting again today

I would make one decision before almost anything else: which country am I actually building the operation around?

If I were targeting the United States again, I would set up the legal entity properly, keep EIN documentation accessible, establish a real business phone number, choose reliable sourcing, create a legitimate return/fulfillment solution, perform intellectual-property checks, build the store, and configure Search Console, Bing and Merchant Center immediately. I would use marketplaces as acquisition channels rather than allowing any one of them to become the company itself.

If I wanted to start from Türkiye instead, I would probably take the opposite approach.

Rather than reconstructing the infrastructure I spent years building, I would establish the appropriate business structure—often a şahıs şirketi for someone starting small—open the sales channel I actually intend to use, and connect it to an existing operational system.

For people following that route, Gymtayt Dropshipping is the system I built specifically for that problem.

Gymtayt Dropshipping

Our Trendyol partners, for example, can connect their seller integration, import products and let stock and price synchronization continue behind the scenes. Shopier and ikas offer other paths depending on how the merchant wants to sell. The goal is not to promise that somebody will magically become profitable in seven days.

The goal is to remove the ridiculous amount of repetitive operational work between finding something to sell and actually delivering the order correctly.

I also want to give a special thank-you to Zeynep Ercan, who supported the establishment of the Gymtayt Dropshipping operation and has put enormous effort into its logistics side. Software can automate a surprising amount of commerce, but eventually a real person still has to make sure the right physical product reaches the right customer.

That work matters.

The tools I would bookmark before starting

These are the services and resources that either formed part of my own journey or that I would want close at hand if I were building the operation again:

You can also turn this article into your own roadmap with an AI agent

There are enough moving pieces here that I would not try to remember all of them manually.

Give this article to an AI agent together with your country, target market, budget and preferred sales channel, then ask it something like:

Analyze this dropshipping experience and build me a step-by-step launch plan for my situation. Separate legal/company setup, suppliers, IP checks, store infrastructure, payments, logistics, search engines, marketplace integrations and ongoing operations. Tell me which steps depend on my country, verify any requirements that may have changed, and do not move to the next phase until the dependencies of the previous phase are clear.

That turns somebody else's ten years of mistakes into a checklist you can adapt to your own situation.

And that may be the biggest thing I learned from all of this.

Dropshipping is often marketed as a shortcut around building an e-commerce operation.

It isn't.

You still need sourcing, logistics, customer service, compliance, payments, technology, product research and distribution. What dropshipping really does is remove one enormous requirement from the beginning: buying inventory before you know whether anybody wants it.

That is still a powerful advantage.

But the more of the remaining operation you can make reliable and automated, the less your business feels like gambling on products—and the more it starts behaving like an actual system.

That is the system I eventually wanted for myself.

So I built it.