Marketing
Get clear answers from the web with a source for every claim, so you can make decisions with confidence.
The shift
An AI answer can sound right and still be out of date. Checking it by hand takes an hour you don’t have.
A chatbot will still answer without opening a page. Every answer here is backed by pages you can open.
The outcome
Every claim links to its source.
If the search does not support the answer, the reply says so. It does not smooth the gap into a paragraph.
It reads what anyone can open online, and keeps your private files out of it.
Save answers with their links for later.
Where it earns its keep
The question is whether a public statement is still true. The reply shows the pages, including one that contradicts it if that page exists.
The question is specific. The reply is short, with the links under it, so the room can open the page that matters.
The question asks for the disagreement, not a blend. The reply keeps the two pages apart and says where they differ.
The operating model
From first conversation to a system that runs itself.
We map how the work gets done today, including the calls only a person should make, and agree on it before anything is built.
Built on Cloudflare and connected to the tools you already use, so nothing has to move.
Tested on your real work from last week before any customer sees it.
We launch with close monitoring, keep it running, and you can change the rules at any time. It’s yours to keep.
The decision
A chatbot, a researcher on staff, and a sourced answer are different levels of trust.
| Off the shelf | Hire in-house | Abracadabra | |
|---|---|---|---|
| What can be checked | Often nothing. The paragraph is the product. | Their notes, if they keep them. | The page under the claim. |
| When the source is missing | A fluent answer may still arrive. | They say they could not find it, or they fill the gap. | The reply says it could not find a page. |
| What it will not read | It depends on the product, and is easy to forget. | Whatever they can log into. | Public pages only. Private files belong in document search. |
| When you don’t need us | If the answer is already in your files, use them. For a casual question, a normal chat is fine. | A researcher is the right hire when the question needs interviews, judgment, or access that cannot be put on the public web. | The fit is a public fact, a real decision, and an unsourced paragraph that would be dangerous to trust. |
Named work
The blogs on pirx.ca are the public record of this kind of work: pages written to answer a question, with the search history kept on that portfolio note. This page does not repeat the figures. Open the note for the dated Search Console pull.
Running in production today.
Priced to the project. Tell us what you need and we’ll quote it.
Questions, answered
faq-title
The answer has to point at pages. If it cannot find a page, it says so.
No. It reads public pages. Private files belong in answers from your documents, not here.
No. Every claim meant to be trusted carries a link you can open. A remembered citation is not good enough.
It gives you the facts and sources. The decision is yours.
As current as the pages opened for that run. A saved answer is frozen on that day. Ask again if the week matters.
No. It shortens the reading. A person who knows the field still decides what the pages mean.
Bring a question you need answered before your next decision, and we’ll show you a sourced answer.