IGrowth & StrategyAnalysis
Selling Through Marketplaces or on Your Own: A Choice for Makers and Dealers
A marketplace brings buyers immediately; your own store brings customers who belong to you. This assessment shows what speaks for each route, what research knows about platform operators as competitors, and how manufacturers and dealers combine both.
Key takeaways
- Marketplaces are rented land: the operator decides reach, rules, and fees. Your own online sales are owned land: slower to build, but the customer relationship belongs to you.
- Platform operators are both landlords and potential competitors. Amazon was more likely to enter successful product spaces of third-party sellers, and affected sellers grew less on the platform afterward (Zhu and Liu 2018).
- On a marketplace, the fee is charged on every purchase, including repeat purchases. The higher the repurchase rate, the more the calculation favors your own channel.
- For many businesses, the answer is a planned dual route: the marketplace for acquisition, your own channel for retention, with clear rules on prices and dealers.
The question every manufacturer and dealer faces
A manufacturer of tool accessories has sold through specialist dealers for years. Customers ask why its products are not on a large marketplace. A dealer in laboratory supplies has its own online store that is hardly found and is considering also offering its range on a platform. Both face the same trade-off: immediate reach against a customer relationship that belongs to them.
The question this article sets out to answer: when does it pay for a manufacturer or dealer to sell through marketplaces, when through its own online sales, and how can the two be combined? To answer it, we first clarify the terms, then introduce a mental model from platform research, put documented rules and findings into context, calculate both routes using customer lifetime value, weigh two views, and close with a step-by-step assessment.
On a marketplace, you buy buyers. In your own channel, you build customers.
Terms: marketplace, own sales, rented land and owned land
An online marketplace is a platform on which many providers sell their goods and the operator organizes search, payment, and often shipping. The provider pays fees for this, usually a share of the selling price. With own online sales, the business sells through its own store, an ordering portal for business customers, or a website with an ordering function. It bears the setup, visibility, and fulfillment itself.
The article Platform dependence introduced two terms for this. Rented land consists of channels where someone else sets the rules and controls the customer relationship. Owned land consists of channels where you decide whom you reach, when, and how. A marketplace is rented land, your own store owned land, even though many of its visitors come via search engines.
In platform research, the providers on a platform are called complementors: they complement the operator’s offering and make the platform more attractive to buyers. This role is the starting point of the mental model.
The mental model: the operator as landlord and competitor
To a provider, a marketplace operator is first of all a landlord. It provides reach and charges rent. But many operators also sell themselves, as retailers on their own platform. Then they are also competitors of their tenants. Feng Zhu and Qihong Liu (2018) studied this tension using the example of Amazon.
They found that Amazon was more likely to enter product spaces in which third-party sellers were successful. Amazon entered less often where growth required considerable seller effort on the platform. After entry, the affected third-party sellers pursued growth on the platform less. For buyers, by contrast, demand rose and shipping costs fell. The authors advise sellers not to rely only on a few bestsellers but on broad assortments or products requiring considerable expertise and support, and to build the ability to discover new products early.
| Operator’s role | What it offers you | What it can cost you |
|---|---|---|
| Landlord | Buyers, trust, payment, partly logistics | Fees on every sale, rules it can change |
| Data holder | Analyses of search and sales | Customer data stay largely with the operator |
| Potential competitor | More demand in the category | The operator’s own offering for successful products |
What is documented: rules and proceedings
This tension is not just an assumption of researchers. On December 20, 2022, the European Commission accepted binding commitments from Amazon and thereby closed two antitrust proceedings. Among other things, Amazon committed not to use non-public data of independent sellers for its own retail business and to treat sellers equally when selecting the featured offer on the product page. The commitments on the featured offer and on Prime apply for seven years and cover Amazon’s marketplaces in the European Economic Area except Italy, where the competition authority had imposed its own requirements. Commitments do not mean that an infringement has been established.
Andreoli-Versbach and Gans (2024) counter that alternative perspectives on a marketplace’s business model and incentives often get short shrift in such proceedings. For a provider, the dispute among lawyers and economists matters less than the sober conclusion: conditions on a marketplace are determined by rules, proceedings, and decisions of the operator over which the provider has little influence.
Since July 2020, business users have had some rights under the EU regulation on fairness on intermediation platforms, such as the requirement that changes to terms be announced at least 15 days in advance. How strongly a single rule change can hit is shown by the study on Apple’s privacy prompt described there: highly dependent online retailers lost between 8 and 40 percent of revenue relative to less dependent ones, depending on the method.
The calculation: fee per sale versus cost of acquisition
Economically, the routes differ mainly in when costs arise. On a marketplace, the provider pays a fee on every sale, even when the same customer buys for the fourth time. In your own channel, you pay mainly once to win the customer and low ongoing costs afterward. According to industry overviews, the selling fees of large marketplaces usually range between about 7 and 15 percent of the selling price, depending on the category; the operator’s current fee schedule is binding.
The following worked example uses the customer value ladder from the guide Calculating customer lifetime value: at a 75 percent repurchase rate, a customer stays around four years on average. The customer buys 2,000 euros a year, and the contribution margin before sales costs is 35 percent, that is, 700 euros a year.
Worked example with freely chosen figures: €700 contribution margin per year before sales costs, four-year lifetime. Marketplace: 700 − €240 fee = €460 per year. Own store: 4 × €640 − €600. Dual route: €460 in the first year, then 3 × €640.
In the example, the two pure routes are close together. The difference arises only through retention: if you bring the customer into your own channel after the first purchase, you save the fee on all subsequent purchases. With a low repurchase rate, the calculation reverses. If most of your buyers purchase only once, the marketplace is often cheaper than the costly buildup of your own visibility. The guide Contribution margin per order describes how to set up the calculation for your offerings.
Two views: reach or control
The first view emphasizes reach. Buyers search on large marketplaces, trust their payment and returns handling, and compare there. A small provider reaches them without a long buildup. The operator’s logistics can also offer delivery times a small business can hardly achieve alone. If you are not on the marketplace, you leave the buyers there to others, possibly even to dealers who offer your own products in an unplanned way.
The second view emphasizes control. On the marketplace, the operator determines fees, ranking, and rules, customer data stay largely with it, and successful products can attract it as a competitor. Your own channel is slower, but whatever it builds in customers, data, and reputation remains with the business.
I consider both views justified and see the decision in three questions. How high is the repurchase rate? How much explanation does the product need? And how much revenue would a single marketplace carry? As a rule of thumb, I consider it risky when more than a third of orders come through a single channel you do not control.
The trade-off at a glance
| Characteristic of your business | Rather favors the marketplace | Rather favors your own channel |
|---|---|---|
| Repurchase rate | Low, mostly one-time purchases | High, regular reordering |
| Product | Standard goods, easy to compare | Needs explanation, with advice and service |
| Customers | Many small, unknown buyers | Known business customers with agreed terms |
| Assortment | Broad, many items with small revenue | Narrow, own brand with a clear profile |
| Existing sales | No specialist dealers who would be annoyed | Dealer network that needs protecting |
| Resources for visibility | Low; buildup would take too long | Available; time to build your own content |
For manufacturers with dealers, there is a further question: if the manufacturer sells on the marketplace itself, it may compete with its own dealers. That is a channel conflict as described in the comparison Sales channels compared. The rule there: each task needs exactly one channel in charge per customer group.
Without a plan
- The entire assortment is listed
- Prices deviate from specialist dealers in an unplanned way
- Buyers remain anonymous order numbers
- The marketplace grows into the largest channel
With a plan
- Selected items for new buyer groups
- Pricing rules that dealers know
- Inserts, service, and registration lead into your own channel, where permitted
- A cap on the share of any one channel
A five-step approach
Clarify customer value and repurchase rate
For your most important product groups, calculate how often customers reorder and what they are worth over their lifetime.
Select the assortment for the marketplace
Start with items meant to reach new buyers that affect specialist dealers little, instead of listing the entire range.
Strengthen your own channel
Make sure your website, ordering route for business customers, and contact options work. The checklist A website that brings inquiries describes what a website needs.
Create routes into your own channel
Service, registration for warranty or spare parts, and a newsletter with consent give buyers reasons to come directly, as far as marketplace rules and data protection allow. The guide B2B newsletters describes how a newsletter prepares conversations.
Review shares quarterly
Record what share of orders comes through which channel, and set a cap for the marketplace.
Conclusion: rent acquisition, own retention
When does the marketplace pay off, when your own online sales, and how can the two be combined? The marketplace pays off for standard goods, many small buyers, and a low repurchase rate, because it brings reach without a long buildup. Your own channel pays off for reordering, advice, and known business customers, because no fee is charged on every subsequent purchase and the customer relationship stays with the business. Research shows that operators can be landlords and competitors at the same time, and the European Commission’s proceedings show that this dual role needs rules. For many businesses, a planned dual route is therefore the best answer: use the marketplace for acquisition and build retention in your own channel.
- This week: Determine the repurchase rate for your three most important product groups.
- Next week: For one group, calculate the customer lifetime value for the marketplace, your own channel, and the dual route.
- In a month: Decide which items are sold through which route, which pricing rules apply, and the maximum share any channel may carry.
A marketplace is a good place to meet customers. It is a bad place to keep them.
Frequently asked questions
Should I sell on Amazon and other marketplaces or run my own store?
That depends mainly on the repurchase rate, the product, and your existing sales. Marketplaces suit standard goods and many one-time buyers; your own channel suits reordering, advice, and known business customers. Many businesses do best with a planned dual route.
How high are the fees on online marketplaces?
Large marketplaces usually charge a share of the selling price, which according to industry overviews often ranges between about 7 and 15 percent depending on the category, plus, depending on the model, base, shipping, or storage fees. The operator’s current fee schedule is binding.
Can a marketplace operator compete with me?
Yes. Many operators also sell themselves. A study of Amazon found that the company was more likely to enter successful product spaces of third-party sellers. In 2022, the European Commission accepted binding commitments from Amazon on the handling of seller data and the equal treatment of sellers.
May I bring marketplace customers into my own store?
That is governed by the terms of the respective marketplace and by data protection law. Direct contact using platform data is often restricted. Permissible routes can include service, registration for warranty or spare parts, and a newsletter with consent. This is not legal advice.
Does a marketplace harm my dealers?
It can, if the manufacturer approaches the same customers there at different prices. This can be prevented with a clear selection of items, known pricing rules, and an assignment of customer groups to channels. Contractual price requirements for dealers are limited by competition law.
Sources
- Zhu, F.; Liu, Q. (2018): Competing with complementors: An empirical look at Amazon.com. Strategic Management Journal, 39(10), 2618-2642.
- Andreoli-Versbach, P.; Gans, J. (2024): Interplay between Amazon store and logistics. European Competition Journal, 20(2), 460-497.
- European Commission: acceptance of Amazon’s commitments in cases AT.40462 and AT.40703, December 20, 2022
- Regulation (EU) 2019/1150 on promoting fairness and transparency for business users of online intermediation services (EUR-Lex)
- Aridor, G.; Che, Y.-K.; Hollenbeck, B.; Kaiser, M.; McCarthy, D. (2026): Evaluating the Impact of Privacy Regulation on E-Commerce Firms: Evidence from Apple's App Tracking Transparency. Management Science, 72(8), pp. 7009-7027.
Note. This article is a general assessment and not legal advice. Fee information may change. Charts and tables without a source are worked examples with freely chosen figures or the author’s own classification.



