Diffbot

Diffbot turns public web pages and text into structured data through APIs, from page extraction to a knowledge graph and search. It is best suited to teams processing public web data at scale; its breadth is compelling, though monthly credits and request rates shape what each plan can support.

Overview

Diffbot combines tools for extracting page data, crawling sites, analyzing text, querying public entities and searching an index. That range is useful when a team needs several ways to work with unstructured web content, but it may be more than buyers seeking only a focused enrichment workflow require.

Web Search offers an API and a self-hostable index that can run on a machine or in the cloud, including offline use. Closed media sites such as Facebook, Reddit, LinkedIn and Instagram are generally off limits, so it is not a route to comprehensive search across those services.

Key features

Extraction and crawling

Extract classifies a page and returns structured JSON using a standard page type and ontology. That offers a consistent format for page-level data. Crawl starts from seed URLs, spiders sites and extracts pages into structured datasets, making it the more relevant option for teams collecting information across sites rather than processing isolated pages.

Language, knowledge and search

The Natural Language service identifies entities, relationships, coreferences, salience and sentiment in plain text. The Knowledge Graph supports structured queries about companies, people, news and other public web entities. Together, these give teams options for turning text into relationships and querying entity data, not just extracting page fields.

Web Search adds a self-hostable index and API. Python and TypeScript libraries and agent skills are listed as interfaces in the documentation. Subscribers receive a unique API key; Diffbot collects login, query history and API call data to provide its services.

Pricing

The Free plan costs nothing, includes all four APIs and 10,000 credits per month, and requires no credit card. It is capped at 5 requests per minute. Once the credit quota or rate limit is exceeded, requests receive a 429 Quota Exceeded response. Quotas reset at the start of each billing period, making Free suitable for evaluation or modest usage rather than uninterrupted high-volume processing.

Startup costs $299 a month for 250,000 credits per month, all four APIs and 5 requests per second; overage is $0.001 per credit. It removes the free tier’s much lower request rate and raises the credit allowance, but the 5-request-per-second ceiling may constrain faster workloads.

Plus costs $899 a month for 1,000,000 credits per month, 25 requests per second and overage at $0.0009 per credit. It adds Crawl access for 25 active crawls and three user seats. The higher allowance and request rate suit larger workloads, while the published seat count matters for teams sharing the account.

Enterprise has custom pricing, a custom credit allotment, 100 or more active crawls, custom user seats, dedicated support and 25 or more requests per second. It is the fit for organizations that need more scale or tailored access; buyers must obtain a price to compare its cost.

Technical and implementation help is available at [email protected]. Account, billing and sales inquiries go to [email protected] or the listed phone number.

Platforms

Diffbot is available through API, web and self-hosted platforms. Its self-hostable search option is a distinct advantage for teams that need to run an index on their own machine or cloud, including offline use.

Who it's for

Diffbot fits teams that need to extract, crawl, analyze or search public web data, especially when several of those tasks belong in one toolkit. Its stated enrichment types include person and organization profiles, firmographic data, and employment and location attributes. The published Startup allowance of 250,000 credits per month makes it a substantial recurring commitment for teams that have moved beyond evaluation. Buyers seeking CRM sync or waterfall enrichment should look elsewhere: neither is offered.

Pros and cons

• Broad set of data tools: page extraction, site crawling, language analysis, entity queries and web search cover different stages of working with public web information.
• Self-hostable search: the index can run on a machine or cloud and support offline use, giving teams deployment options beyond a hosted API.
• Meaningful free evaluation: 10,000 monthly credits and all four APIs are available without a credit card, though the 5 requests per minute cap and quota response limit sustained use.
• Paid plans have defined throughput and quotas: buyers can compare allowances, overage rates, request limits and, on Plus, active crawls and seats before choosing a tier.
• Not suited to every enrichment workflow: CRM sync and waterfall enrichment are absent, and closed social platforms are generally outside Web Search’s scope.

Alternatives

For an open-source option that runs on your own infrastructure, OpenGTM has a free self-hosted plan; third-party provider costs may apply. Choose it when self-hosting and avoiding a subscription matter more than Diffbot’s combination of extraction, crawling, language analysis and knowledge graph services.

Clay has a free plan with 500 actions and 100 data credits per month, plus unlimited seats and tables and up to 200 rows per table. It is worth comparing when those enrichment-specific allowances and a free trial better match the work.

B2BEnrich offers a free plan with 50 monthly credits for person and company enrichment, aimed at testing and small projects. It is a narrower-budget option for those needs rather than Diffbot’s broader web-data toolkit.

Coresignal offers a 7-day free tier with 2,000 credits, 5 requests per second and email support. Consider it when a short trial and that smaller credit allowance are a better fit.

Elvesora Enrichment has a $30 /mo month-to-month plan for 1,000 enriched company records, with REST and remote MCP, bulk processing, XLSX export and quality support. It may suit buyers who want a lower-cost company-record workflow rather than a broad public-web platform.

People Data Labs offers API-based enrichment with free access and credit-tiered IP Enrichment APIs. Compare it when API enrichment is the priority rather than crawling, text analysis or self-hosted search.

PredictLeads has API, web and self-hosted platforms, with usage-based API-call tiers. It is an alternative to consider when those access options and usage-based calls suit the workload.

TargetWise has a free plan with 20 work emails, five business-phone results, and dashboard, API and MCP access. It is a more focused choice when those contact results are the main requirement.

See the data enrichment software directory for more options.

Verdict

Choose Diffbot if your team needs multiple ways to turn public web content into structured, searchable data, and can match its credit and rate limits to the workload. Its breadth and self-hostable search are the strongest reasons to choose it; look elsewhere if you need CRM sync, waterfall enrichment, or a low-cost focused enrichment service.