Scrape Anything by Context.dev — Investment Research Report

Category: Developer infrastructure / live web data API

Stage: Early-stage, Y Combinator Summer 2026

Founder: Yahia Bakour, Founder & CEO; small team

HQ: San Francisco, California (YC profile)

Funding: YC-backed; amount, instrument, valuation, and complete investor list not publicly verified

Business Model: Usage-based B2B API subscriptions, credit overages, and enterprise plans

Product Hunt Launch Date: 2026/10/12 (sheet display value)

Report Date: 2026/10/15

Initial Investment Scorecard

Dimension	Score / 10	Evidence confidence	Assessment
Product and developer experience	8.0	Medium	Broad APIs and mature docs; independent reliability benchmarks are unavailable.
Market potential	8.0	Medium	Agent products need live web data, although category spend is still emerging.
Founder and team	7.5	Medium	Relevant engineering and startup experience; team depth needs confirmation.
Traction	6.5	Low–medium	Customer stories and growth claims are encouraging, but financial metrics are private.
Business model	7.0	Medium	Recurring usage model can expand; infrastructure margins are unreported.
Defensibility	5.5	Low–medium	Reliability and integrations may compound, while core features are replicable.
Provisional overall	7.0	Medium–low	DD: diligence is justified before an investment decision.

Executive Summary

Scrape Anything is Context.dev’s October 2026 Product Hunt launch for web extraction. One API returns Markdown, HTML, screenshots, or structured fields, and supports crawling, search, answers, monitoring, and brand data. It replaces in-house browser, proxy, retry, parsing, and refresh infrastructure. Its value depends on dependable production results, not launch votes.

The opportunity is credible: agents, retrieval systems, support bots, and research tools need fresh web data. Context.dev combines scraping, structured extraction, and brand data; published examples include Mintlify, SiteGPT, and Sourcely. Company and YC pages cite hundreds of organizations. An investor separately claims 500-plus paying customers and 50% monthly growth, without public financial records to substantiate it.

Founder-market fit is promising. YC identifies Yahia Bakour as CEO, citing engineering leadership at Sunrun and Amazon and a prior StockAlarm.io startup that reached 225,000 users before acquisition. Context.dev announced YC backing, but has not disclosed financing terms, revenue, margins, retention, or concentration. Product Hunt showed #9 daily rank and 104 points, useful visibility but not durable-adoption evidence.

Recommendation: DD. Request cohort, revenue, reliability, and cost data. A capable founder, customer integrations, and accessible usage pricing warrant a diligence meeting. Investment conviction depends on verifying recurring paid usage, attractive margins when proxy or browser escalation is needed, and a measurable advantage over Firecrawl, Bright Data, Cloudflare, and open-source tools.

Product Overview

Customers submit a URL, domain, sitemap, or file for Markdown, HTML, screenshots, images, or schema-shaped JSON. Other endpoints support search, answers, batches, monitoring, and brand enrichment. Context.dev bundles rendering, anti-bot handling, proxies, and retries; SDKs cover TypeScript, Python, Ruby, Go, and PHP.

The product can refresh chatbot knowledge, extract sourced facts for agents, or enrich onboarding with company context. Endpoint breadth supports account expansion, but reliability, billing, and support must stay consistent as the API grows.

Founder and Team

YC lists Bakour’s Sunrun engineering leadership and Amazon technical role. StockAlarm.io reportedly reached 225,000-plus users and 100,000-plus daily notifications before acquisition; Bakour says he built and sold two startups. This is relevant experience for developer infrastructure.

Team size is inconsistent across public snapshots: the YC launch narrative says four, while its directory lists six. Confirm current headcount, ownership of reliability and customer support, and whether critical systems depend on the founder alone.

Market Opportunity

A bottom-up underwriting assumption: 25,000 software and AI teams could have recurring workloads for a managed web-data API. At $24,000 annual spend per mature account across scraping, crawling, monitoring, search, and extraction, the illustrative serviceable wedge is $600 million. This is an analyst estimate, not a published market statistic.

At 3% penetration, the model implies 750 accounts and $18 million ARR; at 10%, 2,500 accounts and $60 million. The thesis depends on teams preferring an API vendor to model search, internal crawlers, or larger data platforms—and using it repeatedly in production.

Traction and Growth Quality

Product Hunt showed #9 daily rank and 104 points. The company page has a 5.0 rating across 19 reviews, a Context.dev-wide count rather than a clean measure of this launch’s conversion. YC’s profile currently says 600-plus companies, while the founder’s YC launch narrative says 400-plus. The official customer page describes SiteGPT migrating from Firecrawl, Mintlify using Context.dev for branded documentation, and Sourcely crawling academic journals and PDFs. These examples show workflow integration, but company-published stories do not disclose contract values or retention.

Investor Taro Fukuyama’s public LinkedIn post claims 500-plus paying customers and 50% monthly growth. This is positive but unverified: no revenue series, account definition, or cohort data accompanies it. The company evolved from Brand.dev, so diligence should separate legacy brand-data customers from new scraping and agent workloads. Request monthly active organizations, paid accounts, ARR, gross/net retention, churn, usage concentration, and revenue by endpoint for at least 12 months.

Competitive Position

Firecrawl is the closest developer-focused rival across scraping, crawling, mapping, search, monitoring, and agent features. Context.dev differentiates through combined web extraction and company/brand context, and says rendering and proxy handling are included in base scrape pricing. A Product Hunt reviewer praised support and concurrency compared with Firecrawl, while noting that brand-color extraction needed refinement. This is useful anecdotal feedback, not a controlled benchmark.

Bright Data has deeper proxy and enterprise scraping infrastructure, while Apify and open-source tools such as Crawlee or Crawl4AI let teams compose their own systems. Cloudflare and model-provider browsing can make basic retrieval cheaper or more convenient. Context.dev’s potential edge is developer experience, reliable access, structured outputs, and operational learning. Scraping itself is replicable; sustained quality and integration depth would need to form the moat.

If the largest platform bundles reliable web extraction and live search into its existing developer or model subscription, what measurable quality, coverage, compliance, or workflow advantage would keep customers paying Context.dev? The company should show that advantage with independent benchmarks and customer migration data.

Business Model and Economics

Official plans list Developer at $19/month, Pro at $99, Growth at $299, and Scale at $499, plus enterprise pricing. A scrape starts at one credit; extraction and specialty endpoints cost more. Most failures are not billed, while rendering and proxies are included in the base scrape rate.

The model is easy to trial and expands with usage, but raises a margin question: protected pages vary in browser, proxy, retry, and support cost. Scale’s one million credits for $499 imply $0.0005 revenue per included credit before endpoint mix—not gross profit per page. Request costs and margins by endpoint and customer cohort. Monitoring and integrations may raise retention; simple scraping may be easy to switch.

Path to a Unicorn and Valuation

A $1 billion valuation is not a base-case forecast. At 20–25× ARR, it would require about $40–50 million ARR. At $24,000 annual spend, that means roughly 1,700–2,100 mature accounts; at $5,000 spend, 8,000–10,000 accounts. Such multiples require growth, high retention, strong margins, and category leadership; they may compress if basic web access becomes bundled.

A plausible path is a scraping wedge, expansion into search, answers, monitoring, and brand data, then enterprise workloads with security and SLAs. Public evidence supports early product signals, not a unicorn forecast. With financing price and valuation undisclosed, the round cannot yet be underwritten.

Key Risks

Commoditization: Model providers, cloud vendors, or open-source libraries may make basic scraping a low-margin feature.
Reliability: No independent benchmark or domain-level methodology supports public success-rate claims.
Cost exposure: Browsers, proxies, retries, and support may compress margins on hard targets.
Traction quality: Public customer counts conflict and do not prove paid, active, retained usage.
Source dependence: Websites can block crawlers, change layouts, or restrict automated access.
Legal exposure: Terms, copyright, privacy, database rights, and robots rules vary across sites and jurisdictions.
Security: URLs and extracted content may contain sensitive data; enterprise controls need validation.
Product sprawl: Expanding across scrape, search, answers, brand, and monitoring can dilute execution.
Team capacity: A small team may struggle with 24/7 reliability and enterprise commitments.
Conversion and concentration: Free credits may not convert, or revenue may depend on a few large users.

Final Scorecard and Decision

Dimension	Score / 10	Evidence confidence
Product and developer experience	8.0	Medium
Market potential	8.0	Medium
Founder and team	7.5	Medium
Traction	6.5	Low–medium
Business model	7.0	Medium
Defensibility	5.5	Low–medium
Overall	7.0	Medium–low

Final Decision: DD. The company merits diligence, not an investment commitment based on launch attention. Upgrade toward Invest if financial records verify recurring paid usage, strong retention and gross margins, customer diversity, and reproducible reliability advantages. Move to Watch or Pass if “paying” counts materially include free accounts, usage does not recur, unit economics deteriorate at scale, or customers view the API as replaceable.

Due-Diligence Questions

What are monthly ARR, net retention, gross margin, churn, and paid-account cohorts since January 2026?
How do the 400-plus, 600-plus, and 500-plus paying-customer claims reconcile by date and definition?
What share of revenue comes from the top 10 customers, legacy Brand.dev, and each API endpoint?
What are success, latency, and retry rates by domain type and proxy/browser requirement?
What are unit costs and gross margins for scrape, extraction, search, and Answers?
What caching, opt-out, robots, and legal-review policies govern source requests?
What data-retention defaults, security controls, compliance scope, and incident history exist?
Who are the four versus six team members, and which systems have single-person ownership?
What are the exact YC and other financing instruments, dilution, cash, and runway?
What proves customers expand across endpoints, and what are the main churn reasons?