Diffbot is an extraction API for developers: send a URL, get JSON back in Diffbot's schema. Paid plans start at $299 a month, and Crawl, the part that finds the pages for you, starts at $899. If you are a pricing, sales or research team that wants specific columns delivered to a sheet every morning, WebRobot does that from one plain-English sentence for $79 a month, with no code to host.
Diffbot figures taken from its published plans and credit table
Robot console
Standing by Running · s Complete · rows
1 · Pick a target
3 · Fields to extract
Agent log
Crawl graph
Extracted data · rows
Want this data fresh every morning, without lifting a finger?
Comparison
| Dimension | Diffbot | WebRobot |
|---|---|---|
| What it is | Four developer APIs: Extract, Natural Language, Knowledge Graph and Web Search, plus Crawl | A hosted scraping robot: describe the data once and it runs on a schedule |
| Who sets it up | A developer who calls the API, stores the JSON and schedules the job | Whoever needs the data, in a sentence; no code |
| Free option | Free plan: 10,000 credits a month, 5 requests a minute, no card | No free tier; the interactive demo is the free taste |
| Entry paid plan | Startup $299/mo: 250,000 credits, 5 requests a second, overage $0.001 a credit | Launch $79/mo ($63 yearly): 5 robots, 10,000 records/mo, daily scheduling, API |
| Mid plan | Plus $899/mo: 1,000,000 credits, 25 active crawls, 3 seats, overage $0.0009 | Scale $249/mo ($199 yearly): 25 robots, 100,000 records/mo, hourly, agent actions, 5 seats |
| Top of the range | Enterprise, custom credits, 100+ active crawls, dedicated support | Autopilot $699/mo ($559 yearly): 100 robots, 1,000,000 records/mo, real-time monitors, 15 seats |
| Finding the pages | Crawl only on Plus and up; below that you supply every URL | The robot follows listings, pagination and detail pages on every plan |
| Output schema | Fixed per page type (Article, Product, Discussion and others); custom fields need rules or your own code | The columns you name, in the order you want them |
| Company and people data | Knowledge Graph of organizations, people and products at 25 credits an entity | Not included; the robot collects what is published on the pages you point it at |
| Logins and forms | Public pages; sessions and form steps are yours to script | Agent actions on Scale and up: signs in, searches and paginates unattended |
| Delivery | JSON over the API; Google Sheets and Excel connectors | CSV, Excel, Google Sheets, Slack, Zapier, webhooks, REST API; change alerts on Scale |
| Contract | Monthly, cancel any time | Monthly or yearly, cancel any time |
Their pricing
| Plan | Price | Credits a month | Overage |
|---|---|---|---|
| Free | $0 | 10,000 | None; 429 error at the cap |
| Startup | $299 | 250,000 | $0.001 a credit |
| Plus | $899 | 1,000,000 | $0.0009 a credit |
| Enterprise | Custom | Custom | Negotiated |
| Action | Credits |
|---|---|
| Extract one page | 1 |
| Extract one page through Diffbot's datacenter proxy | 2 |
| Crawl a site for links and pages (Plus and up) | 0 |
| Natural Language, one document up to 10,000 characters | 1 |
| Web Search, one search | 1 |
| Knowledge Graph, export or enhance one entity | 25 |
| Knowledge Graph, facet record or enhance with refresh | 100 |
Two things shape the real cost. First, page extraction is cheap per credit: Startup works out to $1.20 per 1,000 pages, or $2.39 through the proxy. Second, the plan gates matter more than the credits. The Free plan's 5 requests a minute means 10,000 pages take about 33 hours of calls, and Crawl, which is how Diffbot finds product or listing URLs on its own, is not on Free or Startup at all. A team without a URL list is looking at Plus at $899.
Worked bill 01
About 9,000 extractions a month. That fits Diffbot Free if you have a developer to run the calls, store the JSON and schedule the job. With nobody to build that, WebRobot Launch runs it daily for $79 and lands the rows in your sheet.
Worked bill 02
You know the category page, not the 20,000 product URLs under it. On Diffbot that needs Crawl, so Plus at $899 a month. WebRobot follows the category, pagination and detail pages itself: 20,000 records a month on Scale at $249.
Worked bill 03
At 25 credits an entity that is 200,000 credits, inside Startup at $299. WebRobot has no company database, so for firmographics pulled from a graph, Diffbot is the right purchase and we would not pretend otherwise.
Fair play
Advantage 01
A queryable graph of organizations, people, products and articles, with facets and enrichment. If your product needs firmographics or entity matching, that is a dataset, not a scraper, and WebRobot does not sell one.
Advantage 02
For a data team feeding a model or a search index with text from thousands of unrelated sites, a consistent Article schema at about a tenth of a cent a page is hard to beat.
Advantage 03
10,000 credits a month with no card is generous. A developer with a known URL list and a few thousand pages a month may never need to pay. Our cheapest plan is $79.
Advantage 04
Entities, sentiment and facts from text sit on the same credit pool as extraction. If you need the page and its analysis in one pipeline, that is convenient.
Switching triggers
Trigger 01
Free covers a prototype. The moment you need more than 5 calls a minute or crawling, the bill is $299 or $899 a month, which is a lot for a team that wants a few thousand competitor prices a day.
Trigger 02
The Product API returns Diffbot's idea of a product. Your analyst wants SKU, pack size, shipping threshold and stock badge in named columns. The gap between the two is code someone has to write and keep working.
Trigger 03
An API returns JSON. Scheduling, retries, storage, deduplication and delivery to Slack or a sheet are yours. Teams whose engineer moved on find the pipeline quietly stopped weeks ago.
Trigger 04
Supplier portals, wholesale price lists and member directories need a session and a few clicks before the data appears. WebRobot's agent actions sign in and fill the search form on Scale and up. The guide to scraping a site that requires a login covers how that works.
Shortlist
| Tool | Best for | Entry price | Our write-up |
|---|---|---|---|
| Firecrawl | Developers turning pages into Markdown or JSON for LLM apps | Hobby $19/mo, $16 billed yearly | Firecrawl pricing |
| Zyte | Scrapy teams that want ban handling billed only on success | Pay as you go from $0.13 per 1,000 requests | Zyte pricing |
| Apify | Prebuilt scrapers for specific sources, usage-based bill | Starter $19/mo plus usage | Apify pricing |
| Bright Data | Large volumes and ready-made datasets on enterprise contracts | $1.50 per 1,000 records pay as you go | Bright Data alternative |
| Import.io | Managed extraction with a query allowance | Standard $199/mo billed annually | Import.io alternative |
| WebRobot | Business teams that want named columns on a schedule, including behind logins, without code | Launch $79/mo, $63 billed yearly | WebRobot pricing |
Migration
Step 1
Open whatever reads Diffbot's JSON. The handful of keys it keeps, plus anything you compute afterwards, is your column list.
Step 2
"Every product in this category with name, SKU, price, sale price, pack size and stock status, all pages." Start from the category URL, not a URL list.
Step 3
Point the robot's webhook or REST API at the same endpoint, or skip the code and deliver to a Google Sheet. Compare twenty rows field by field.
Step 4
If Knowledge Graph lookups feed your CRM, keep Diffbot for that at a lower tier and move the page scraping over. Your credit use drops to the entity calls.
Teams leaving Diffbot tend to run one of a few jobs, each with its own page: competitor price monitoring, ecommerce data scraping for catalogs and reviews, news scraping for media monitoring, and website change monitoring. If you want the robot behind your own code, the web scraping API page shows the request and response format.
FAQ
Describe the data in plain English. The robot finds the pages, runs on our servers every day and delivers 10,000 records a month to your sheet for $79.