Trade Me Scraper for New Zealand Property Data

Yes. The ScrapeIt Trade Me API returns listing id, the price string and sale method, suburb, district and region, bedrooms, rateable value and open home times as JSON from an endpoint we host, refreshed on the schedule you set. Motors, jobs and marketplace rows come through the same endpoint with their own columns. The same data also comes as CSV or XLSX files or straight into your database.

Yes, and the per week part matters. New Zealand quotes residential rent weekly, not monthly. Trade Me's own rental search takes a minimum and maximum price in NZD per week and returns a rent per week value. A team that assumes a monthly figure and multiplies by twelve will be wrong on every row, because a year holds 52 weeks rather than 48. We deliver the number as published and label the column per week. If your model needs a monthly figure we add a derived column, rent times 52 divided by 12, and keep the original beside it. We do not convert silently.

A listing page shows the structured basics: listing id, title, address, suburb and district, price display, bedrooms, bathrooms, floor area, land area, property type, agency details, photos and open homes. Two things buyers ask for regularly take more work. Rateable value or capital value is entered by the seller when the listing is created, and chattels - the fixtures, fittings and appliances included in the sale - are normally written into the description rather than shown as their own field, so they need parsing out of free text. The third is history. What a listing asked last month is not on the page. Price reductions, a change of sale method, and withdrawal followed by a relist exist only in a record taken repeatedly.

Treat them as temporary. Listings run to a closing date and leave the active search once they expire, sell or are withdrawn. That is the strongest argument for continuous collection over a single large pull: whatever you did not record before a listing left is gone from your side of the fence. We put a first seen date, a last seen date and a collection timestamp on every row, so a disappearance becomes a data point rather than a gap. When a listing id reappears at a different price, you can read it as a relist instead of counting a new property. Auction outcomes and the spread between an initial asking figure and a final result are built the same way, from the sequence of observations rather than from one snapshot.

By default we deliver listing and commercial data: the property, the price, the sale method, the location, the agency. Agency name and branch are business attribution and we keep them. Individual agent names, direct phone numbers and email addresses can be excluded from your delivery on request, and we will exclude them by default if you say so at scoping. Private sellers do appear in property and in the general marketplace, so this is not a theoretical point. New Zealand's Privacy Act 2020 governs the collection and use of personal information about identifiable individuals. We collect only publicly available data - everything a visitor can see on Trade Me - and we collect it legally, with nothing behind a login.