A small studio that ships scrapers.
Harpoon builds and runs Apify Actors: single-purpose scrapers with a documented input, a clean dataset out, and a price per row you can check before you run anything.
What we build
Every Actor does one job. The posts on a TikTok profile. A single Amazon product page. A Target store stock. A job board listings. Narrow scope is what makes them fast, and it is what keeps them working after the source site changes its markup.
Today that covers 12 platforms: social feeds, marketplaces, real estate, job boards and search. The catalog is deliberately small. 21 Actors that stay maintained beat a thousand that quietly stop returning rows.
When a site you need is not covered, a custom Actor with a fixed scope is usually the right answer. Send the URL and the fields, and the feasibility and the price come back before any work starts.
How we work
- Fast by design
- Benchmarked and profiled against the alternatives, then tuned. Speed is part of the product, not a footnote on a chart.
- Priced per result
- You pay for the rows that land in your dataset. No seats, no minimum, no retainer.
- Continuously improved
- Frequent releases keep every Actor leaner and more reliable as the sites it reads change their markup.
- One job, done right
- A documented input, a documented schema, and rows that drop into a sheet, a warehouse or a pipeline without cleanup.
Who it is for.
The same Actors serve very different jobs, which is why the output schema matters as much as the scrape.
| Data and analytics teams | Recurring datasets that feed a warehouse, without standing up a scraper team to keep them alive. |
|---|---|
| Growth and lead generation | Profiles, posts, products and reviews turned into lists a sales or marketing team can actually work. |
| Market and competitor research | Price and assortment tracking across marketplaces, refreshed on a schedule instead of by hand. |
| AI agents and automation | Actors callable over the Apify API or through MCP, returning typed JSON a model can use without parsing HTML. |
What we will not do.
Boundaries are part of the product. These ones are not negotiable for any volume.
| No login-walled data | Every Actor reads public pages only. No cookies, no accounts, no private groups. |
|---|---|
| No resale of personal data | We publish scrapers, not people. |
| No seat licensing | Billing is per result, on your own Apify account, at the rate published on each page. |
| No open-ended scope | Custom work is quoted before it starts, with the price per row stated up front. |
Working with us.
Everything runs on your own Apify account, so billing, scheduling and storage stay in one place and you keep the data. For custom work, one person answers, and the answer includes a price.
| [email protected] | |
| Apify | apify.com/harpoon |
| GitHub | Harpoon-Org |
What do you need to scrape?
A URL, the fields you need and roughly how often. That is enough for a straight answer on feasibility and cost.