Marketplace and DTC channel strategy: the complete guide

Write down what each channel is for in a single sentence, and make the sentences different. Channels that all have the same job compete with each other for the same customers and the same stock, and the store pays fees twice for demand it already had.

There are only a few jobs worth having. Discovery is reaching people who are searching for a product type rather than for you, and marketplaces are usually strong at this because that is where a lot of product search starts. Relationship is owning the customer so you can email them, sell them the second item and build repeat purchase, which only your own store does properly. Volume is moving units at a lower margin to cover fixed costs or clear stock. Credibility is being present where a buyer expects to find you, even if the channel sells little. Reach into a specific community is what a social shop or a specialist marketplace can offer.

Once each channel has a job, a lot of arguments resolve themselves. If a marketplace is there for discovery, it should carry the products most likely to be searched for generically, and it does not need your full range. If your own store is there for relationship, it should carry the bundles, the subscriptions and the newest products, because those are what build repeat purchase.

Be honest about which channel is currently paying the bills. Many brands describe their own site as the priority while most orders come from somewhere else. That is a legitimate position, but it changes what you should invest in: if a marketplace produces most of the revenue, protecting listing quality and stock there matters more than a site redesign.

Write the jobs on one page, alongside the share of revenue each channel produces today and the share you want it to produce in a year. That page is the reference for every decision that follows, and it is the thing to revisit when someone suggests adding a fifth place to sell.

Revenue by channel is a misleading comparison, because the same order leaves you with different amounts of money depending on where it happened. Build a simple contribution figure per channel first, and compare that instead.

Start with the same product and follow it through each channel. Selling price, then subtract cost of goods, the channel's commission or referral fee, any listing or subscription fee, fulfilment cost including the storage and handling charges a marketplace applies, the shipping you absorb, payment processing where you pay it, and the advertising cost per order on that channel. What remains is contribution per order.

The result is often surprising. A marketplace order at the same price can leave a noticeably thinner contribution than a direct order once commission and fulfilment fees are counted, while a direct order carries acquisition cost that the marketplace order does not. Which one wins depends entirely on your product, your weight and size, and how expensive your traffic is. That is why the arithmetic has to be done on your own numbers rather than assumed.

Add returns to the calculation. Return rates and who pays for them vary by channel, and a channel with generous returns policies can be materially less profitable than its headline margin suggests. Include the cost of processing a return and the share of returned stock you cannot resell.

Then add the second order. A direct sale that produces an email address and a repeat purchase is worth more than the first order alone. A marketplace sale that produces no contactable customer is worth what it made, once. If you know your repeat rate, apply it, and if you do not, at least note which channels can produce a second order and which cannot.

Rebuild this figure at least twice a year. Fees, fulfilment rates and advertising costs all move, and a channel that was profitable when you added it can quietly stop being so.

Split the catalogue by channel deliberately. Listing the whole range in every place looks efficient and usually is not, because different channels reward different products and an unmanaged full catalogue creates work and price pressure without adding much revenue.

Marketplaces tend to favour products that are easy to search for, easy to compare, easy to ship and stocked deeply. A well known product type at a competitive price with strong images does well. A complex, configurable or expensive item that needs explanation usually does not, because the listing format gives you little room to make the case.

Your own store is where the products that need explanation belong. Bundles, sets, subscriptions, personalised items, new releases and anything with a story that needs telling. It is also where you can control the sequence a shopper moves through, which is what makes a higher order value possible.

Use exclusivity carefully. Keeping some products only on your own site gives people a reason to buy there and protects your best margin items from fee pressure. Keeping too much back makes your marketplace presence look thin and unserious, which hurts the discovery job it was supposed to do.

Consider channel specific versions rather than strict exclusivity. A multipack, a different bundle configuration or a different set size can exist in one place without being the same listing as the single item on another. That gives shoppers a genuine reason to choose a channel and makes direct price comparison less mechanical.

Finally, prune. Products that generate returns, questions or fulfilment problems on a particular channel should come off that channel. On marketplaces especially, poor performance on a few listings can affect how the rest of your account is treated, so a small number of problem items is worth removing rather than tolerating.

Write a price policy that covers every channel, then hold it. Inconsistent pricing across places you sell is the fastest way to lose trust with shoppers who check, and on some channels it can affect how your listings are treated.

The simplest workable policy is one price for the same item everywhere, with differences expressed as different offers rather than different prices. Free shipping thresholds, bundles, subscription savings and loyalty benefits can vary by channel without the item itself being cheaper in one place, and they give people a reason to buy direct without starting a race to the bottom.

Be aware that some channels compare your price with other places the same product is sold, including your own site, and use that comparison in how prominently they show your offer. Undercutting your own marketplace listing on your own site can therefore cost you visibility on the channel, which is an expensive way to save a small amount of commission.

Decide who can change a price, on which channels, and within what range. In a small team this is one person with a documented range. Without that, prices get changed in one place for a good reason and never changed back, and nobody notices until a customer points it out.

Plan promotions across channels rather than per channel. A sale on your own site while a marketplace listing sits at full price is confusing to a shopper who sees both, and a sale on a marketplace you forgot to end can run for months. Keep a single promotions calendar covering every place you sell.

Finally, watch other sellers of your products if you sell wholesale or through distributors. Losing control of pricing usually starts with a single reseller clearing stock, and the earlier you spot it the easier it is to handle.

Treat the listing as a product page you do not fully own, and control every part you are allowed to. On channels where shoppers cannot leave to read your site, the listing has to do the whole job of selling.

Start with the title, because on most marketplaces it carries a large share of the search relevance and it is what shoppers scan. Write it for a person and for the search at the same time: what the product is, the brand, and the attributes people filter and search by such as size, quantity, colour, material or compatibility. Avoid stuffing it with terms nobody would say out loud.

Then the images. Most channels have strict rules about the main image, typically requiring the product on a plain background with no added text or graphics. The images after the first are where the selling happens: scale, use, what is in the box, the detail that answers the main objection. This is where lifestyle and customer photography earns its place.

Fill every structured attribute the channel offers. These power the filters shoppers use to narrow a category, and a missing attribute quietly removes you from a set of results you could have appeared in. It is unglamorous work and it usually moves visibility more than rewriting the description does.

Use whatever enhanced content the channel makes available to brands, such as richer description modules, comparison blocks or video. These often require registering your brand with the channel, which is worth doing for the content alone as well as for the protection it gives against other sellers using your listings.

Keep the content in one place internally so channels stay in step. When you improve a description on your own site because it answers a real customer question, that improvement should reach the marketplace listing too. Most stores update one and forget the others, and over a year the listings drift apart.

Decide in advance which channel gets the stock when there is not enough for everyone, because that decision made under pressure is always made badly. The rule should follow contribution per order and the strategic job of the channel, not whoever shouts first.

Start by knowing your true available quantity across all locations, including stock sitting in a marketplace's fulfilment network that you cannot easily move. Split inventory across fulfilment networks carries a real risk: units in the wrong place are effectively unavailable, and pulling them back takes time and costs money.

Set buffers per channel. Selling the last unit on three channels at once creates cancellations, and cancellations damage your standing on marketplaces in ways that are slow to repair. A small held back buffer on the channel with the harshest penalties for cancelling is usually worth its cost.

Decide the priority order. A common approach is to protect the channel where running out is most damaging, then the channel with the highest contribution, then the rest. Whatever you choose, write it down so the person doing allocation on a busy day is following a rule rather than guessing.

Plan for going out of stock rather than pretending it will not happen. On your own site that means a back in stock signup and an honest restock date. On a marketplace it usually means the listing loses momentum, so decide whether to keep a small quantity flowing to hold position or accept the gap and plan the return.

Finally, connect stock levels to advertising. Advertising a product you cannot ship wastes budget on both sides. Set rules so that paid promotion pauses when stock drops below a level, on every channel, automatically where the tools allow it.

Split advertising by what each type of spend buys. Money spent inside a marketplace buys position in front of people already choosing between products. Money spent outside buys awareness and traffic you can send wherever you want. They are not substitutes.

Inside a channel, the spend is close to the purchase and usually converts well, because the audience arrived intending to buy something in that category. The trade off is that you are paying for a customer the channel keeps, on top of the commission you already pay. That is acceptable when the contribution still works and when discovery is the job you gave the channel.

Outside a channel, you decide where the click lands. Sending outside traffic to a marketplace listing is sometimes the right choice, particularly if the listing converts better than your own page or if the channel offers an incentive for external traffic that improves your effective fee. Check whether such a programme exists on the channels you use, because it can change the arithmetic.

More often, outside traffic should go to your own store, because that is the only place the customer becomes yours. The exception is when your own page cannot compete on delivery speed or on price, in which case sending a buyer to a listing that closes is better than losing the sale.

Measure each type on its own terms. Inside channel advertising is best judged on contribution after commission and fulfilment. Outside advertising is best judged on the blended effect across every channel, because a shopper who sees your ad may buy where they are most comfortable rather than where you sent them.

Expect brand searches to be contested. Other sellers can advertise against your brand name inside marketplaces, and shoppers searching your brand may land on a competitor's product. Decide whether defending that position is worth the spend for you, and treat it as a defensive cost rather than a growth channel.

Sort your channels by whether you end up with a contactable customer, because that single difference decides how much a first order is worth and how hard you should work to acquire it.

Your own store gives you the customer: an email address, consent to contact them, purchase history, and the ability to run a post purchase sequence, a replenishment reminder and a win back. That is what makes a direct first order worth more than its margin, and it is why stores can afford to spend more to acquire a direct buyer.

Most marketplaces do not. You get the order, the fulfilment details and often a masked contact route limited to service messages about that order. You cannot add the buyer to your list, and marketing to them through the channel's messaging is usually restricted. Treat that revenue as a series of first orders unless the channel offers a genuine subscription or follow up mechanism.

This changes what you should do on each. On channels where you keep the customer, invest in the second order: the post purchase flow, the replenishment reminder, the reason to come back. On channels where you do not, invest in what improves the first order: listing quality, price position, delivery speed and review volume.

There are legitimate ways to build a relationship from a marketplace order, and they are worth using within the channel's rules. Warranty or registration processes, product support material, and brand content within the channel's own tools all create a connection. What is generally not allowed is content that pushes buyers off the channel or asks for positive reviews, and breaking those rules risks the account.

Finally, when you compare channels, be explicit about this. A comparison that treats a marketplace order and a direct order as the same thing understates the value of your own store and overstates the case for chasing marketplace volume.

Every channel measures how you operate and adjusts how visible you are accordingly. Treat those standards as part of your marketing budget, because falling below them costs you position that no amount of advertising fully replaces.

The measures are broadly similar wherever you sell: how often you ship on time, how often you cancel, how quickly you respond to messages, how often orders arrive late, and how many defects or complaints you generate. Small failures accumulate, and the penalty is usually reduced visibility before it is anything more serious.

That has a direct commercial effect. On channels where a single product can be offered by more than one seller, service performance affects who gets the prominent position on the page. Losing that position can cut sales of a product sharply without anything changing about the product itself.

Build the operational side to the strictest channel's standard rather than to your own. If one channel demands same day dispatch and another does not, running two standards creates mistakes. One process that satisfies the toughest requirement is simpler than two processes that mostly work.

Watch the messages. Response time to customer questions is measured on most channels, and unanswered messages are one of the easiest ways to slip below a standard. Assign the inbox to a person, not to whoever notices.

Finally, keep an eye on account health as a routine rather than a crisis. Check the performance dashboards on each channel on a schedule, deal with the warnings while they are warnings, and keep records of resolved issues. A suspended channel account can remove a large share of revenue overnight, and reinstatement is slow, which is a good reason not to let any single channel become the only thing keeping the business alive.

Judge channel decisions on the total business, not on each channel's own report, because channels move each other. A shopper can see an advert in one place, search in another and buy in a third, and every platform will claim the credit.

Build a simple blended view. Total revenue, total marketing spend across every channel, and the resulting ratio, tracked over time. Then contribution by channel underneath it. This will not settle every attribution argument, but it answers the question that matters: is the whole business getting more efficient or less.

Use holdouts when the decision is big enough. Turning a channel's advertising off in one region or for a set period, then watching what happens to total sales rather than to that channel's reported sales, tells you more about incrementality than any attribution model. It is uncomfortable and it is usually worth doing before a large budget shift.

Set review points for adding channels. A new channel takes real work: listings, images, attributes, fulfilment setup, service standards, someone to own it. Add one at a time, give it a defined trial period with a target, and judge it on contribution after all its costs including the internal time it consumes.

Have a rule for leaving. A channel that does not reach its target after a fair trial, that consistently returns thin contribution, or that requires more service work than it is worth should be closed or reduced rather than carried indefinitely. Wind it down properly: run stock out, keep listings accurate until they close, and handle the returns tail.

Finally, revisit the one page you wrote at the start. Have the jobs changed. Is the revenue mix moving the way you wanted. Doing that once a quarter turns channel strategy into something you manage rather than something that happens to you.