PricingAlmanacBot

Identifying it

Every request sends this user-agent, unaltered:

PricingAlmanacBot/1.0 (+https://pricingalmanac.com/bot)

It never disguises itself as a browser, never rotates user-agents, and never uses a residential or proxy pool. Requests come from a single small server. If you see traffic claiming to be PricingAlmanacBot that does not behave as described below, it is not us, and hello@pricingalmanac.com would like to know.

What it does

It reads published pricing pages — typically one or two URLs per site — and records the plan names, prices and stated feature availability, along with the URL and the time it read them. Those figures appear on this site's pricing pages, each one citing your page as its source with a nofollow link and the date it was checked.

It does not:

• Crawl your site broadly. It requests only the specific pricing URLs configured for it
• Submit forms, create accounts, sign in, or attempt to reach anything behind authentication
• Execute JavaScript or load your images, fonts, stylesheets or analytics
• Follow links out of the pages it is configured to read
• Republish your page content. It records figures and cites you; it does not mirror your copy

How often, and how gently

Frequency

Roughly once a day per configured URL, spread across a 4-hour window with a stable per-site offset, so the whole set never arrives at the same minute.

Rate

At least 3 seconds between requests to the same host, with a maximum of 2 concurrent connections. In practice that is a handful of requests per day, not per minute.

Conditional requests

It sends If-None-Match and If-Modified-Since. If your page has not changed, a 304 ends the exchange and nothing is downloaded — which is most nights, for most sites.

Backing off

A 429 or a 5xx puts your host in a cooldown and retries are exponentially spaced. Requests time out after 15 seconds rather than holding a connection open.

robots.txt

Fetched before anything else, cached for 24 hours, and obeyed — including Crawl-delay. A disallow is treated as permanent: the URL is dropped rather than retried later.

What it stores

The response body of any page that produced a figure is kept for 90 days. That copy is not published and is not served to anyone; it exists so that if a figure on this site is disputed, the question "what did the page say on the day it was read" has an answer. After that window it is deleted. Only the extracted figures, the URL and the timestamp are kept permanently.

Blocking it

You are entitled to, and it will be respected immediately — no appeal, no contact required. Add this to your robots.txt:

User-agent: PricingAlmanacBot
Disallow: /

The next run reads that before requesting anything else and stops. Your product's pages on this site are then removed at the next rebuild, because a figure that can no longer be re-verified stops being published — this site does not keep serving a price it is no longer allowed to check.

If you would rather not wait for a crawl, or want the data removed sooner, email hello@pricingalmanac.com and it will be switched off by hand.

If something looks wrong instead

If your pricing is misrepresented here rather than merely present, reporting it is faster than blocking, and the correction is usually a one-line fix to how your page is read. The methodology documents how PricingAlmanac extracts, verifies and ages every figure.