Per-Page Pricing: Examples & Companies

What is it

Per-Page Pricing is a billing unit where customers are charged per page crawled, parsed, or rendered — the meter for web scraping and document parsing. The page is where web-data infrastructure and document AI converge on the same noun: Firecrawl meters pages scraped into clean markdown, LlamaIndex’s LlamaParse meters PDF pages parsed into structured output, Mistral AI’s OCR bills $2 per 1,000 pages read, Unstructured charges a flat $0.03 per page of document ETL, and Exa and You.com both sell full-page content retrieval at $1 per 1,000 pages beside their per-call search APIs.

The unit’s appeal is the same as the document’s: a buyer with a URL list or a PDF folder can count their job before running it. Its weakness is that “a page” abstracts over enormous compute variance — a static blog post and a scanned 600-dpi financial table are the same unit and very different work — which is why most of this cohort prices modes on top of the page count rather than one blended rate.

The SEO and docs tools carry the unit in subscription form: Frase and Scalenut bundle audit-page quotas into monthly tiers, and Mintlify treats the documentation page itself as the rendered unit inside its docs platform — pages as an allowance or a hosted deliverable rather than a metered API line.

How it works

The metered cluster prices pages × mode rate, usually denominated in credits:

Worked example — the mode is the price. The four bills above are the same 100,000 pages: Firecrawl scraping lands on the Standard tier’s 100k-credit allotment ($83), Mistral OCR runs $200, and Unstructured’s flat $0.03/page ETL comes to $3,000 whether the pages are clean text or VLM-parsed images. LlamaParse spreads widest — $125 on Fast (1 credit/page) to $5,625 on Agentic Plus (45 credits/page), a 45x swing on an identical page count set entirely by how much reasoning the parse applies.

Worked example — endpoint drift. Firecrawl’s “1 credit = 1 page” holds for Scrape, Crawl, Map, and Monitor — but Search bills 2 credits per 10 results and Interact bills 2 credits per browser-minute. A workload that mixes endpoints can’t be estimated from page count alone; the usage-metric guide calls this unit drift, and it’s the first thing to audit on any credit-denominated rate card.

Companies using this

9 in-corpus companies meter pages: the API cluster (Firecrawl, LlamaIndex, Mistral AI, Unstructured, Exa, You.com) billing pages scraped, parsed, OCR’d, ETL’d, or fetched, plus the content pair (Frase, Scalenut) bundling audit-page quotas into subscription tiers and Mintlify metering documentation pages inside its docs platform.

Patterns observed

The commodity floor is real and public: plain page retrieval settled at $1 per 1,000 pages at both Exa and You.com — undifferentiated page-touching is priced like a utility, and margin lives only in the intelligence layered above it.

The category shows a clear repricing gravity toward simpler meters over time. Unstructured walked from compute-hour billing (~$12.93/1,000 pages) to strategy-tiered per-page (Fast $1, Hi-Res $10 per 1,000) to a single flat $0.03/page, and LlamaParse’s v2 collapsed a 1–90-credit maze into four flat tiers — each step trading a “fairer” ladder for a legible one.

Credits are the standard wrapper. Firecrawl and LlamaParse both denominate pages in credits with tier-sized monthly pools, which buys repricing flexibility without touching the dollar price of a credit. Mintlify applies the wrapper to a docs platform: documentation pages are the hosted unit, but AI features draw on 5,000 included credits at $0.01/credit overage — the page and the credit live side by side, with a hard cap so AI spend can’t surprise the buyer.

Counterexamples & variants

Frase and Scalenut show the unit defanged: pages appear as audit quotas (50/mo on Frase Starter, 1,000 at the top) inside flat subscriptions, with no per-page overage rate published — the page count gates the tier you need rather than metering a bill, which makes them packaging, not pricing.

Unstructured is the sharpest counterexample to the mode ladder: one flat rate for every strategy deliberately removes the “if I turn on the good parser, what happens to my bill?” objection — the opposite bet from a per-mode credit spread. And Mintlify is the semantic variant: its “page” is a published documentation page, not a crawled or parsed one, so the meter is closer to a hosting deliverable than a per-request API line — a reminder that “pages-rendered” spans both the input a scraper consumes and the output a docs platform serves.

What this means for buyers vs vendors

For buyers

Price the mode, not the page: get a representative sample of your documents and run them through the cheapest tier that passes quality, because the spread between modes (1–45 credits on LlamaParse, versus one flat rate on Unstructured) dwarfs the spread between vendors at any single mode. On crawl workloads, ask what bills on failure — redirects, soft-404s, and bot walls are a real fraction of any large URL list — and watch reprocessing, since re-running a corpus after a chunking or embedding change pays full per-page rates again. For bundled audit quotas (Frase, Scalenut) or docs-page platforms (Mintlify), translate the tier into an effective per-page price at your actual monthly volume before comparing against the metered APIs.

For vendors

Pick a lane and commit to it: publish the mode ladder with each rung’s quality difference demonstrable, or flatten it entirely as Unstructured did — the category’s repricing history shows buyers reward legible page pricing and punish mode mazes. Keep the commodity rung at the market floor as the funnel, take margin on intelligence above it, and if you wrap pages in credits, hold the page-to-credit exchange rate stable per endpoint — every footnote on “1 credit = 1 page” is forecast error you’re exporting to the buyer, and the prepaid-credits guide covers why that trust erosion compounds. Buyers estimating token-and-page costs alongside the parse bill can sanity-check model spend with the Mistral pricing calculator.

Explore this theme in the knowledge graph →