Marketplace Platform in New Zealand

The software is not the hard part

Listings, search, profiles, messaging, payments, reviews — all of it is well-trodden. Five developers will quote you competently for it and the range will not be enormous.

What kills marketplaces is that on launch day one side of the market is empty, and an empty marketplace is worse than no marketplace. Buyers arrive, find four listings, and never return. Sellers list, hear nothing, and stop.

Before you commission anything, be able to answer: which side are we solving first, and how? The usual honest answers are:

• Recruit supply manually. Ring people. Sign them up one at a time before a line of code ships. Unglamorous and it works.
• Go narrow. One city, one category. Twenty sellers looks thin nationwide and looks like a market in Tauranga.
• Be the supply. Fulfil it yourself at first, and open it up once there is demand worth having.

If none of the developers quoting asks you this, they are quoting software rather than helping you build a business.

Decide whether money moves through you

This is the single biggest fork in cost and in obligation.

Introductions only. You connect the parties and they settle between themselves. Much cheaper to build, no payment risk, no chargebacks — and your revenue depends on people paying for something they can get around after the first deal.

Money through the platform. Split payments, payouts, refunds, disputes, and a seller dashboard showing what they are owed. Far more expensive, and it is what makes the marketplace worth being on.

Stripe Connect is how most New Zealand marketplaces do the second. It handles seller onboarding, identity verification and payouts, which is a very large amount of work you do not have to build. What it does not decide for you:

• Who is the merchant of record, and therefore who carries the chargeback risk.
• How long you hold funds before releasing them to the seller.
• What happens to your fee on a refund.
• Whether a dispute pauses a payout, and who adjudicates.

Those are commercial and legal decisions. Make them before the build, because they shape it.

Leakage: design against it, do not police it

Both sides will eventually notice they can do the next deal directly. Hiding phone numbers and scanning messages for email addresses annoys everyone and does not work.

What actually keeps people on a platform:

• Payment protection worth having — the buyer's money is safe, the seller knows they will be paid.
• A review history that has value only where it lives.
• Dispute resolution that someone actually runs.
• Scheduling, invoicing, or insurance that is genuinely easier through you.
• Being where the next job comes from. Repeat demand beats every restriction.

Marketplaces that charge for the introduction alone leak steadily. Ones that make the transaction better keep people without trying.

Trust and safety is a running cost

Every marketplace deals with fake listings, misrepresentation, disputes, refunds, and occasionally someone behaving badly in a message thread. These are normal operations, not edge cases, and they arrive in week two.

Budget for the tooling in the build — a moderation queue, the ability to suspend an account, a view of a dispute with both sides' history — and for somebody's time afterwards. A marketplace with no admin tooling forces your team to work in the database, which is exactly as bad as it sounds.

New Zealand adds its own layer: where you stand between buyer and seller affects your obligations under the Fair Trading Act and the Consumer Guarantees Act. An hour with a lawyer before launch is cheap next to finding out afterwards.

Build the smallest thing that proves demand

A great many marketplaces could have been tested for $20,000 instead of $200,000. Before the platform:

• A landing page and a form, with you matching people by hand.
• A spreadsheet of supply and an email introduction.
• An existing group or directory you already have access to.

If matching twenty deals by hand is miserable and people keep coming back, build the platform. If you cannot find twenty deals, the software would not have found them either.

Ask each developer what they would build first, and how long it would take to learn something. The answer tells you whether they have watched a marketplace launch before.

Comparing five quotes

Read for:

• Who asked which side of the market you are solving first?
• Who raised leakage before you did?
• Whose quote includes moderation and dispute tooling rather than only the buyer and seller screens?
• Who has integrated split payments and payouts before?
• Who proposed a smaller first version?

Describe the two sides, what a transaction looks like, where your first fifty sellers come from, and what you would charge. Developers who have built one will spend most of the conversation on the last two.