We built the tool we kept wishing existed.

There is a whole category of work that most data tools do not acknowledge. It is not data engineering. It is not analytics. It is the Monday morning task of checking competitor prices across twelve websites, or the afternoon spent pulling product names into a spreadsheet because there is no export button and the API costs more than the budget allows. It is skilled, necessary work that happens to look like copying and pasting.

The tools built for this job fall into two camps. One camp requires you to write code - Python, browser automation, regular expressions. The other camp requires you to upload your data to a server somewhere, process it in the cloud, trust that nobody is reading it. Neither camp acknowledges that the person doing the work might just want to get on with it.

We built SiteScoop for the second group of people. The ones who are very good at their jobs and would prefer to spend less of their Monday on this particular part of it.

We built SiteScoop in Rust, compiled to WebAssembly, running entirely inside your browser tab. There is no server. There is no account required to get started. When you extract data from a page, that data goes directly into your clipboard or your downloaded file. It does not pass through our infrastructure, because we do not have infrastructure for it to pass through.

This is not how most tools are built, because it is genuinely harder to build this way. It also means we cannot see what you are extracting, cannot log your queries, cannot accidentally leak your data in a breach. Forty-one percent of Chrome extensions collect user data. We thought that was a strange approach for a tool whose whole job is to work with data you care about.

The technical choice is also a values choice. Your data is yours. We want to be unambiguous about that.

We did not build SiteScoop for developers. Developers have excellent tools already. We built it for the people who work with websites all day and have never written a line of code and have no plans to start. Procurement teams tracking supplier and competitor prices. Market researchers building datasets from public listings. Ecommerce sellers monitoring how their category moves. Journalists checking whether the numbers in a press release match what is actually on the retailer's website.

What these people share is a workflow that involves a lot of switching between a browser and a spreadsheet, and a quiet awareness that most of their afternoon could have gone better. SiteScoop slots into that workflow. It does not ask anyone to learn new software, change how they think about data, or get sign-off from IT. It is a Chrome extension. It works on the pages they already have open.

We are particularly proud of how it handles the pages where data is not neatly structured - the ones where the automatic detection does not find a clean pattern and the user needs to point at an element and say "that one, and everything like it." That fallback matters more than the easy case.

We are not building a platform. We are not building an AI that predicts what data you want. We are building a very good tool for getting structured data off websites, and we intend to keep making it better at that specific thing for a long time.

The things we care about improving: the accuracy of automatic pattern detection, the breadth of sites it works well on, the smoothness of the multi-page extraction flow for sites with pagination or infinite scroll. The export formats. The speed. The cases where it currently does not work and probably should.

We will keep the free tier free. We will keep the data local. We will not add an account requirement. These are not features we are planning to renegotiate.