Almost every SME owner I talk to in Indonesia has the same setup: a Tokopedia store, a Shopee store, maybe an Instagram shop, and no website of their own. Sales are happening, so the instinct is to leave it alone. But the marketplace vs own website question is not really about which channel sells more units this month. It is about who owns the customer relationship five years from now.
Marketplaces are excellent at one thing: putting your product in front of people who are already searching to buy. That is genuinely valuable and I am not going to tell any business to abandon Tokopedia or Shopee. But marketplaces rent you customers. Your own website, done properly, lets you own them. Confusing the two, or picking only one, is where businesses leave money and resilience on the table.
What Marketplaces Actually Give You
Marketplace platforms solve discovery and trust at a scale no small business can replicate alone. A shopper searching for "kompor gas 2 tungku" on Tokopedia already trusts the platform's buyer protection, payment flow, and review system. You are borrowing that trust. For a new product or a business with no existing brand recognition, that borrowed trust is worth the fees.
The costs are real too, and they compound quietly:
- Platform fees typically run 2-5 percent per transaction before promotional spend, plus payment gateway fees on top.
- Ad spend dependency. Organic visibility on marketplaces has thinned as more sellers compete for the same search terms, pushing sellers toward paid placement just to stay visible.
- Zero data ownership. You cannot export your buyers' contact details, purchase history, or behavior patterns to build a retention program. The platform owns that relationship, not you.
- Price transparency works against you. Buyers comparison-shop within the same page, which is structurally a race toward thinner margins.
What Your Own Website Gives You Back
A website does not compete with marketplaces on cold discovery, it is not going to out-rank Shopee for generic product searches. What it does is let you keep the customers you have already acquired, whether from a marketplace sale, a referral, or a social media post, and sell to them again without paying a platform fee each time.
Owning your own site means you own:
- The customer database. Names, phone numbers, purchase history, all queryable and usable for retention campaigns.
- The margin on repeat purchases. No per-transaction platform cut once a customer buys direct.
- The brand experience. Your layout, your story, your upsells, not a template shared with thousands of competing sellers.
- Direct communication. WhatsApp broadcast, email, SMS, all reachable without asking a platform's permission or paying for its ad inventory.
This is the deeper argument in owning your customer data or someone else will: the platform that sits between you and your buyer will always optimize for its own margin, not yours.
The Sequencing That Actually Works
I do not recommend SMEs abandon marketplaces to build a website-first strategy, and I do not recommend staying marketplace-only once revenue crosses a certain point. The right sequence is marketplace-first for discovery, own-site for retention, run in parallel rather than as a replacement.
Stage 1: Marketplace-only. True for genuinely new businesses with no existing customer base. Use the marketplace to learn what sells, at what price, to whom. Do not build a website yet, you have nothing to retain.
Stage 2: Add the website once repeat-purchase signal appears. When you notice the same buyer names reordering on your marketplace dashboard, that is the signal. Those are customers worth owning directly. Build a simple site, migrate them there with an incentive (loyalty discount, faster restock notice, WhatsApp ordering), and keep marketplaces running for new-customer acquisition.
Stage 3: Split by function, not by volume. Keep marketplaces as the top-of-funnel acquisition channel. Direct all retention, loyalty, and high-margin repeat business through your own site. Track the split explicitly, if 80 percent of revenue still comes through marketplaces after 18 months, that is a signal your retention program on-site is not converting, not that the site was a bad idea.
The Fee-Plus-Data Cost of Staying Marketplace-Only
The honest math most sellers skip: platform fees alone rarely kill margins, it is fees plus the compounding cost of never building a first-party customer list. A business selling exclusively through marketplaces for five years has paid platform fees on every single transaction, including the hundredth purchase from the same loyal customer, and still has no way to reach that customer directly if the platform changes its algorithm, raises fees, or the seller account gets flagged for a policy dispute.
That last risk is not hypothetical. Marketplace account suspensions happen, sometimes for reasons outside a seller's control, and a business with no alternate channel to its own customers has no recovery path. A website with even a modest first-party list is insurance against that single point of failure.
Takeaway: Split the Job, Don't Split the Loyalty
Marketplace vs own website is not a competition to win outright, it is a division of labor. Let marketplaces do what they are good at: discovery, trust-borrowing, and volume from strangers. Let your own website do what marketplaces structurally cannot: retention, margin, and a direct line to the customers who already chose you once. If your business is still 100 percent marketplace after your first hundred repeat customers, that is the moment to build the second channel, not before, and not much later.