Online sales change: from dropshipping to e-commerce

Online businesses have fallen into a trap: search engines’ AI tools intercept customers before salespeople even have a chance to speak to them. There are fewer opportunities to engage with buyers, and influencing them has become more difficult. That is precisely why the key takeaway for entrepreneurs today is this: you can no longer operate ‘blindly’. It is not enough to focus solely on sales. You need to understand what is happening within the business: how much a customer is actually worth, how margins are changing, whether people are returning, where the business is making a profit, and where it is merely generating turnover.

In this article, TON OP company highlights two processes that are currently both reshaping the market and directly reducing the profits of online businesses:

• Artificial intelligence in search results intercepts clicks before the customer even reaches your website or advert.
• Voice search generates a stream of long, conversational, often random queries that clutter up adverts and blur the accuracy of targeting. Both processes operate in parallel and hit the same target - the margins of online businesses.

The logic behind the customer journey has changed. Previously, a person would search for something, land on a search result, and hop from site to site: comparing, reading, hesitating, and choosing. And this entire process took place within an environment where the business could influence them.

Now, a significant part of this journey has shifted outside - beyond the boundaries of the e-commerce site.

People still start with a query. But increasingly, instead of a list of options, they receive a ready-made solution. For example, AI Overviews ‘intercept’ the user and Google answers the question itself. Blocks with answers generated by artificial intelligence appear above the adverts and organic results. And so, some users make a decision without visiting a single website, whilst e-commerce loses not only informational but also commercial traffic. Moreover, for informational queries preceding a purchase (such as ‘best sleep tracker 2026’), the presence of AI in search results reaches 83%. These are precisely the queries for which the buyer would previously have clicked through to the shop’s website, compared products and placed an order.

At the same time, the structure of the search queries themselves is changing, according to experts at TONOP doo ltd. They have become longer and more conversational. Previously, the wording made it possible to understand quite precisely what a person wanted. Now, the same query can be phrased in dozens of different ways-as a question, a reflection, or a description of a situation.

These formulations mix different states of mind: some people are simply curious, others are unsure, and others are almost ready to buy. The advertising system and the advertiser cannot identify specific groups of users and queries. The budget is being spent, yet the results become unpredictable for no apparent reason. And we mustn’t forget that there are over 8.4 billion voice-enabled devices in the world, and voice search only exacerbates this problem.

1. Break down CAC by channel and switch off unprofitable ones
The key change is that you can no longer rely on ‘average figures’. Average conversion rates, average customer acquisition costs, and total turnover all obscure the true picture, according to experts at TONOP doo ltd company. Average customer acquisition cost masks the failures of some channels at the expense of others. You need to calculate CAC separately for each acquisition channel: Google Ads, social media, email. It is important to understand the details.

• How much does a customer from each channel cost?
• Does the customer pay for themselves?
• Do they return?
• Where does the business actually make money?

If the first purchase becomes less predictable and more expensive, the stability of the business begins to depend on whether the customer returns. Without this, any business model becomes fragile.

Margins have long ceased to be a constant. They depend on the channel, device, region and day of the week. The same product can be profitable on Instagram and loss-making on Google Ads. TONOP experts note that a weekly recalculation of margins by channel is an absolute minimum.

Analytical data becomes crucial, as certain aspects of user behaviour may simply not be visible there (such as interactions within a search engine, for example). It is important to understand what happens after a purchase, how different customer groups behave, and where real value is created.

Modern platforms for managing dropshipping and online retail allow you to view margins in real time, forecast demand based on sales history, monitor marketing effectiveness, and automatically recalculate the procurement plan when data changes.

The TONOP programme automates the entire sales process: from supplier selection and quality control to demand forecasting and purchase planning.