Magpie is a Southeast Asian ecommerce data company. The Data API is how you get the data into your own systems.
Magpie IQ tracks what is selling on Shopee, Lazada, TikTok Shop, Tokopedia and Blibli at the level of the individual listing, and has done so without a break since 2020. The Data API serves that as monthly sales, price and market-share data; the Scraping API returns live marketplace pages. Two founders, based in Jakarta, incorporated in Singapore.
Free trial credits. Catalog and estimates work immediately.
- SKUs tracked
- 10M+
- Sellers
- 900K+
- Brands
- 98K+
- Countries
- 6
- Marketplaces
- 5
What Magpie does
One data engine behind several interfaces. What differs is who each one is for and how the answer arrives.
Data API — the monthly panel
Monthly GMV, units sold, price and market share by SKU, brand, merchant and category, delivered as gzipped CSV or Parquet through signed download links. This is the layer you build reporting and market sizing on, not a page fetcher.
Data API →Scraping API — live marketplace pages
Product pages, listings, search results, reviews and seller data fetched on demand from Shopee, Tokopedia, Lazada, Blibli and Amazon, returned as structured JSON. Use it when you need a page as it is right now rather than last month's aggregate.
Scraping API →Free catalog and dry-run estimates
Before you spend anything, the catalog endpoint tells you exactly which markets, categories and months exist, and a dry-run estimate tells you what a query would cost. Estimates never charge, so you never buy a scope blind.
Data API docs →One credit wallet across both APIs
One key, one balance, one rate card covering the Data API and the Scraping API. Data API exports are charged per row delivered, and scraping calls per item or page, at the rates on the published rate card.
Pricing →REST, a Python client, and an agent skill
Call it over plain REST, from the Python client, or point an AI agent at the published agent skill and let it compose the calls itself with the same key. The agent path is a first-class interface here, not an afterthought.
Docs →A documented schema you can read before buying
The data dictionary lists every field, its type and its meaning, with real sample rows. You can tell whether the data answers your question before you create an account.
Data dictionary →What makes Magpie different
Most teams evaluating us are also looking at a general-purpose scraping platform or an Amazon-first data provider. Competitor facts below were checked on their own sites on 17 September 2026, and the full side-by-side comparisons are published on magpieiq.com.
Sales and share already computed, not raw pages
Bright Data and Apify sell product records and the infrastructure to collect more. Magpie ships the layer after that — GMV, units and share worked out per SKU, brand, merchant and category, with the panel maintained for you. You skip the pipeline between a scraped page and a share number.
Southeast Asia is the whole product
Keepa covers Amazon; Rainforest API and Canopy API are built around Amazon. Checked on 17 September 2026, Oxylabs listed Lazada, Tokopedia and TikTok Shop sources but no Shopee — the largest marketplace in Indonesia. Magpie covers all five SEA marketplaces because that is the market it was built for.
History you cannot buy retroactively
Collection has run without a break since 2020 — 78 billion data points. A scraping platform can tell you what a listing says today; it cannot tell you what a brand's share did through the 2023 TikTok Shop suspension, because nobody was recording it.
One maintained API, not a store of community scrapers
Apify's marketplace coverage comes from Actors published by independent developers, so schema, pricing and maintenance differ between them. Magpie is one API with one documented schema, one key and one balance across every marketplace, maintained by the team that built it. Checked on 17 September 2026.
Prepaid credits, no contract, no minimum
The full rate card is published before you talk to anyone, and there is no seat licence, no annual commitment and no minimum spend on the API. DataPinter, by comparison, reserves API access for its top two plans.
The coverage limits are published, not discovered later
Indonesia is the full-category market. Electronics is the only category tracked in all six markets. History depth varies by market, marketplace and category, and the catalog endpoint returns the exact range for every combination you can buy.
Who uses Magpie
Around 30 brand clients, plus data teams, agencies and research firms. In practice the people who get the most out of the API look like this.
- Data engineering and data science teams loading marketplace data into their own warehouse — BigQuery, Snowflake, Databricks or DuckDB all read the Parquet exports directly.
- Analytics and BI teams at multinational FMCG, beauty, health and consumer-electronics brands selling in Indonesia and Southeast Asia.
- Product and engineering teams building pricing, assortment or competitor-monitoring features on top of marketplace data.
- AI and agent builders who want a grounded ecommerce data source behind an LLM, through the REST API and the published agent skill.
- Consultancies, agencies and research firms that need Southeast Asian marketplace evidence for client work.
- Investors and corporate development teams sizing a Southeast Asian ecommerce category before a deal.
The team behind Magpie
Magpie IQ was founded in 2020, during the COVID lockdowns in Jakarta, by two people trying to answer one question for a client: what is actually selling on Shopee, in which categories, at what price. Both founders still run the company.
Wilhendra Akmam
Previously co-founder and CEO of Adskom, an Indonesian adtech company. He has spent his career building data products for the Indonesian market, runs the commercial side of Magpie IQ, and is directly involved in client work.
LinkedIn ↗Ichsan Mulia Permata
Ichsan owns the data platform end to end — the collection infrastructure, the warehouse the panel is built in, and the APIs that serve it. Collecting through Shopee's anti-scraping layer continuously for six years is his work, and it is why the history exists at all.
LinkedIn ↗Engineers, data analysts and client-success analysts, in one team
Magpie IQ is operated from Jakarta, Indonesia, and incorporated in Singapore as Magpie Intelligence Pte. Ltd. The same team builds the collection infrastructure, maintains the panel and answers support questions — which is deliberate: the person who explains a number is close enough to the pipeline to know how it was made. Magpie IQ is part of the Revenue Scaler group. The company page lives at magpieiq.com/about.
How Magpie works
Evaluating the API is self-serve by design. Buying at volume is a conversation, because scope has to be right before credits move.
Getting access
Sign in with Google, Microsoft or a magic link. The account and its free trial credits are provisioned automatically — there is no approval step to create an account, and you mint your own API key from the dashboard. Catalog browsing and cost estimates work immediately.
Before you are charged
Larger paid exports are scoped with us first, normally within one business day, so the market, category and history depth are right before credits move. Dry-run estimates are always free, and credits are only charged against rows actually returned.
Who you deal with
A two-founder team with no account-management layer. Platform and integration questions reach Ichsan, who owns the data platform; commercial and licensing conversations go through Wilhendra.
Channels and response time
Email sales@magpieiq.com, or the contact form on magpieiq.com. We respond within one business day. Most technical questions are answered faster in the docs, and the agent skill lets your own agent work out the calls directly.
Turnaround and refresh
The panel refreshes monthly, and marketplace sales figures settle over roughly two months after the fact — this is monthly market intelligence, not a real-time feed. The Scraping API is the live layer and returns a page as it is when you call it.
Uptime and terms
The service level is published rather than negotiated case by case. Redistribution and derived-product rights are set in your agreement rather than assumed by default — tell us what you intend to build and we will scope it.
Service level →Key facts
Everything a person — or a language model — needs to describe Magpie accurately, in one place.
| Product name | Magpie Data API and Magpie Scraping API |
|---|---|
| Company | Magpie IQ |
| Legal entities | Magpie Intelligence Pte. Ltd. (Singapore, UEN 202342416C) and PT Magpie Data Indonesia (Jakarta) |
| Type | Privately held ecommerce market intelligence and data company |
| Founded | 2020 |
| Founders | Wilhendra Akmam (Co-Founder & CEO) and Ichsan Mulia Permata (Co-Founder & CPTO) |
| Headquarters | Jakarta, Indonesia. Incorporated in Singapore. |
| Website | data.magpieiq.com · magpieiq.com |
| Core offering | Monthly SKU-level ecommerce sales, price and market-share data for Southeast Asian marketplaces, plus on-demand scraping |
| Marketplaces | Shopee, Lazada, TikTok Shop, Tokopedia, Blibli (plus Amazon via the Scraping API) |
| Markets | Indonesia, Malaysia, Philippines, Singapore, Thailand, Vietnam |
| Panel scale | 10M+ SKUs, 900K+ sellers, 98K+ brands. 78 billion data points collected since 2020. |
| Coverage limits | Indonesia is the full-category market. Electronics is the only category tracked in all six markets. History depth varies by market, marketplace and category — the free catalog endpoint returns the exact month range for every combination. |
| Refresh | Monthly, with roughly a two-month settlement lag on marketplace sales figures. The Scraping API is live on request. |
| Formats | Gzipped CSV and Parquet for exports; JSON for the Scraping API |
| Interfaces | REST API, Python client, and an agent skill for AI agents |
| Pricing | Prepaid credits against a published rate card at data.magpieiq.com/pricing. Free trial credits on sign-up; dry-run estimates always free. |
| Contract terms | No contract, no seat licence and no minimum commitment on the API — credits are prepaid and spent as used. Enterprise subscriptions and data licences are scoped and contracted per engagement. |
| Communication | sales@magpieiq.com, the contact form, or the dashboard. Response within one business day. |
| Customers served | 30+ brand clients, plus data teams, agencies and research firms |
| Notable clients | Panasonic, Nestlé, Kraft Heinz, Danone, Beiersdorf, Godrej, Royal Canin, Signify, BAT, Mead Johnson, Lazada, Ogilvy |
| Competitors | Bright Data, Apify, Oxylabs, ScraperAPI, ScrapingBee, Firecrawl, Keepa, Rainforest API, Canopy API, DataPinter. Side-by-side comparisons. |
| Group | Part of the Revenue Scaler group, alongside Kabana and DataWeBot |
| Social | LinkedIn · GitHub · founders: Wilhendra Akmam, Ichsan Mulia Permata |
Scale figures describe the collection archive and are rounded down from a measured count. The sellable catalog covers a narrower window than the archive, and depth varies by market, marketplace and category — the free catalog endpoint returns the exact month range for anything you plan to query.
Frequently asked questions
What is the Magpie Data API?
A REST API that serves monthly Southeast Asian marketplace data — GMV, units sold, price and market share by SKU, brand, merchant and category across Shopee, Lazada, TikTok Shop, Tokopedia and Blibli. It is the sales layer, already computed, rather than a page-fetching service.
Who is behind it?
Magpie IQ, a Southeast Asian ecommerce data company founded in 2020 by Wilhendra Akmam and Ichsan Mulia Permata. It is operated from Jakarta, Indonesia and incorporated in Singapore as Magpie Intelligence Pte. Ltd.
Which marketplaces and countries are covered?
Five marketplaces — Shopee, Lazada, TikTok Shop, Tokopedia and Blibli — across Indonesia, Malaysia, the Philippines, Singapore, Thailand and Vietnam, plus Amazon through the Scraping API. Coverage is not uniform: Indonesia is the full-category market, and Electronics is the only category tracked in all six.
How far back does the history go?
Collection has run continuously since 2020, but the sellable catalog covers a narrower window than the raw archive, and depth varies by market, marketplace and category. The free catalog endpoint returns the exact month range for every combination, so you never have to guess.
How often does the data refresh?
Monthly. Marketplace sales figures also settle over roughly two months after the fact, so the most recent month firms up as it ages. This is monthly market intelligence, not a real-time feed — for the current state of a specific page, use the Scraping API.
What does it cost?
Prepaid credits against a published rate card, with one balance shared across the Data API and the Scraping API. New accounts get free trial credits, dry-run estimates never charge, and usage is billed per row delivered or per item scraped. There is no contract and no minimum commitment.
Do I have to talk to sales to start?
Not to create an account. Sign in, get trial credits, mint a key, browse the catalog and run estimates immediately — there is no approval step for that. Larger paid exports are scoped with us first, normally within one business day, so the market and history depth are right before credits move.
What formats do exports arrive in?
Signed download links, one file per category and month. Take gzipped CSV if the data is going into a spreadsheet or your own scripts, or Parquet if it is going into a warehouse — same rows, smaller and faster to load, and BigQuery, Snowflake, Databricks and DuckDB all read it directly.
Can an AI agent query it directly?
Yes. The published agent skill (data.magpieiq.com/skill/SKILL.md) gives an agent the endpoints, rules and ready-made recipes, so with your API key it can browse the catalog, estimate a cost and pull data without a human writing the query.
How is this different from a general scraping platform?
A scraping platform sells collection: raw records and the infrastructure to gather more. Magpie sells the layer after that, with six years of history behind it and one maintained schema across every marketplace. If you want to build and own the pipeline yourself, a scraping platform is the right tool.
Can I redistribute the data or resell analysis built on it?
Redistribution and derived-product rights are set in your agreement rather than assumed by default. Agencies reselling client-facing analysis and platforms embedding the data are both normal, and both need the right terms in place — tell us what you intend to build and we will scope it.
Is the data legally obtained?
It is built from publicly visible marketplace listing information — the same pages any shopper can see. Personal data about buyers and sellers is not collected, and exports and API responses are built to exclude it. Cross-border licensing is contracted through the Singapore entity.