The First 90 Days in E-Commerce: What to Do and When?

June 16, 2026 · KârHesapla · Ferhat Demir · E-commerce Specialist

When the notification of the first order arrives, most sellers do the same thing: they pack it excitedly, hand it to the carrier and wait for the next order. What really matters — whether the product’s pricing logic has been set up correctly — only becomes clear a month later, when the first return arrives or a competitor cuts prices. This article is not about product selection or setting up a company — those steps belong to the preparation before you start selling. The topic here is the first 90 days after sales begin: what to look at in which week, and which decisions backfire if made too early.

Weeks 1-2: Settle the operational rhythm; don’t touch the numbers yet

The only goal of the first two weeks is this: the process of taking orders, packing, handing over to the carrier and answering customer questions should run without getting stuck. In this period there is not yet enough data to change prices, question the commission rate or panic over “did I pick the wrong product?” Fifteen orders say nothing statistically.

Even so, there is one thing that must not be overlooked: a record of what you actually earn on each order. Three columns in Excel are enough — selling price, product cost, shipping fee. Sellers who do not keep this record in the first two weeks are the ones who, in the third month, end up saying “I thought I was profitable, but I wasn’t”.

Weeks 3-4: The first real price test

Once fifteen to twenty orders have accumulated, something can finally be said: is the conversion rate low, or is the price high compared with competitors? The most common mistake here is lowering the price without ever calculating the break-even point. Every discount given without knowing below which price a loss begins — after commission, VAT and shipping are deducted — is a gamble.

Practical step: before changing the price, work out your break-even price clearly. Run the test above this floor, in small steps (for example 5-8%); don’t go for a 20% discount in one go. Even if a large discount increases sales, you will notice where the margin went three weeks later.

Example: if your break-even price is 180 TL and your current selling price is 220 TL, you have a margin buffer of about 40 TL. Instead of putting this entire buffer into a discount in one go, first see how a 10 TL reduction affects the conversion rate. If the number of units sold does not rise noticeably, the problem is not the price but the product visuals or description — and that is not solved by lowering the price further.

Weeks 5-6: The first return, the first real cost lesson

The first return usually arrives at the start of the second month and surprises new sellers, because most people think the cost of a return is only “taking the product back”. In reality, it also includes the return shipping fee, repackaging and, in some categories, the risk of the product becoming unsellable. A seller who does not account for these three items cannot understand, in a category with a return rate of 8-10 percent, why the profit table does not add up at the end of the month.

At this point you need to add the return share to your net profit calculation; we covered how to calculate it in detail in the calculating the cost of returns article. If you have seen a single high-return product in the first 90 days, don’t panic yet — the real decision is made at the end of the third month, when the rate has stabilized.

Weeks 7-8: The first campaign invitation

Marketplaces usually start inviting sellers to campaigns in the second month. Thanks to the sales history accumulated so far, when an offer arrives you can now answer the question “will this discount push me into a loss?” with real data. The trap inexperienced sellers fall into is accounting for the volume increase a campaign will bring but not for the loss of profit.

Running the campaign profitability calculation before deciding takes a few minutes; it shows how much the number of units sold needs to rise after the campaign. Most invitations accepted without this calculation end with the seller unknowingly giving products away at a loss.

Sidebar: when the first low-rated review arrives

Another event that is almost inevitable in the first 90 days is the first low-rated review. Most new sellers perceive it as a crisis about product quality and, in a panic, cut the price or remove the product entirely. Yet a single negative review says almost nothing statistically; in a sample of ten or fifteen reviews, the weight of a single data point is very low.

What you should do is read the content of the review and check whether there is a recurring pattern. A complaint like “the shipment arrived late” points to an operational problem, while a complaint like “the product doesn’t match the description” may require a correction in the listing text. Panicked reactions (hastily cutting the price, removing the product) usually cost more than the problem itself.

Weeks 9-10: Which product is really working?

You usually start with more than one product, and by the ninth week a clear difference emerges between them. Some sellers decide with the wrong criterion at this point: they think the best-selling product is “the best product”. Yet a high-volume but low-margin product can bring in less profit than a low-volume product with a healthy margin.

A simple comparison makes this clear:

Although product A stands out in units sold, its real contribution to profit lags behind B’s. A seller who allocates stock and ad budget according to “the best-seller” has actually channeled resources into the less profitable product. The right question should not be “which product sells the most” but “which product contributes the most to total profit”.

Sidebar: the first restocking decision

If bought in the right starting quantity, most products begin to sell out towards the end of the second month. Here there are two opposite mistakes new sellers make. The first is panicking a few weeks before stock runs out and placing an urgent order in the same quantity (or more) — which means locking capital into the same distribution again before it is clear which product really brings in profit. The second is the opposite: delaying the restocking order out of worry that “maybe it won’t sell” and running out of stock, which damages search ranking on the marketplace and customer trust.

The profit contribution table you obtained in weeks nine and ten is exactly what helps here: make the restocking decision not by “which product is running out” but by “which product contributes the most to total profit”. Instead of restocking a low-contribution product in the same quantity, shifting that resource to the high-contribution product directly improves profitability in the third month.

Weeks 11-12: Continue, stop or scale?

At the end of the third month you have three months of real data, and the decision falls into a clear table:

The second row of this table is the most frequently overlooked situation. The fact that sales are coming in does not mean the business has been set up correctly — it only means that the demand exists. If it is a margin problem, postponing it in the hope that sales will “sort it out eventually” turns a mistake that could be fixed in the third month into a chronic one by the sixth month.

When does this calendar not apply?

The sequence above is for sellers who start with a single product or a narrow category. For a seller who publishes a broad catalog (more than 50 SKUs) all at once from the start, the weekly tracking logic must kick in much earlier — because the number of errors also grows proportionally. Likewise, in stockless (dropshipping-like) models, returns and shipping dynamics work differently, so the return assumptions in the weeks 5-6 section do not apply one-to-one.

Checklist: what you should have at the end of the first 90 days

• A real net profit record for each product (selling price − cost − commission − VAT − shipping)
• The average return rate by category
• At least one campaign offer evaluated with a profitability calculation
• A clear ranking of which product contributes the most to total profit
• A written threshold for the criterion the continue/stop/scale decision will be based on (e.g. “if the margin falls below 10% two months in a row, remove the product”)