What the audit is for
It is a free tool with a real bill behind it. Every run costs us money at Google, at our ranking data supplier and at the model that writes the findings, and it stays free because the volume stays sane. This page is what “sane” means.
What is fine, including the part people expect us to forbid
- Auditing a business you own — The obvious one.
- Auditing a competitor — This is allowed and we are not going to pretend otherwise. Everything the audit reads is public — a Google profile, a map position, a page’s own source, a public performance score. It is what twenty minutes of searching would find, and tools have done exactly this on arbitrary domains for years.
- Auditing a business you are considering buying from, or buying — Also fine.
- Sharing a report — The link is yours. Send it to whoever you like.
What is not fine
- Automating it — Scripts, scrapers, headless browsers and anything else that runs the audit without a person in front of it.
- Working around the limits — Rotating addresses, connections or browsers to get more runs than the limits allow. The limits are layered precisely so that evading one does not help.
- Running a list — Working through a set of businesses one after another. One at a time about businesses you have a reason to care about is the use this was built for; a list is a data collection exercise using our budget.
- Reselling it — Presenting our findings as your own product, or building a dataset out of them to sell.
- Prising the tool open — Attempting to extract the scoring weights, thresholds or rule definitions behind a finding, whether by probing the interface or by reverse-engineering the responses.
- Attacking any of it — Probing for vulnerabilities, overwhelming the service, or interfering with anybody else’s use of it.
- Misrepresenting a report — Altering one, or presenting a finding about a business as current when the report says when it ran.
The limits, and what happens when you hit one
The audit is rate-limited per connection, per browsing session, and per business scanned — so repeatedly auditing one business is capped however many people do it, and auditing many businesses is capped however many connections you have. A handful an hour is the shape of it.
We do not publish the exact numbers, for the reason you would expect. If you are using this the way it is meant to be used you will never see one.
Hitting a limit is not a ban. You are told, the report you already have keeps working, and the counter resets.
We do not block on suspicion
A session that looks automated gets a slower, cheaper audit — cached and free checks instead of paid ones — rather than an error page. A false positive should cost somebody a slightly worse report, never the ability to become a customer.
Blocking outright is for the list above, and it is done by address, by connection or by business. We will say what happened if you ask.
When our own crawler visits your site
This runs the other way too. Our crawler identifies itself as AllTradesDigitalBot and this page is the address in its user agent string.
- It obeys robots.txt — Including a rule naming it specifically. A path you disallow is not read, and the audit records that it was refused rather than going round it.
- It reads one page and stops — Your robots.txt, then the home page, within a few seconds. It does not crawl a site, it does not follow links through it, and it does not come back on a schedule.
- It takes nothing private — There is no login to attempt and no form to submit. It reads what a first-time visitor on a phone would see.
- It can be told to stop — Write to the address below and we will stop reading a domain entirely.
Reporting something
Abuse of this service, a report you think is wrong, or our crawler behaving badly — all three go to abuse@alltradesdigital.com.
This page is part of the terms of use. What we do with the data an audit produces is the privacy policy.