About the site
Privacy & cookies
This website is built to be as privacy-friendly as possible. No tracking cookies are set and no profiling takes place — the only cookies that exist are two functional ones remembering whether you chose light or dark mode. On the public domain, two cookie-free, anonymised measurement scripts are loaded that count visits without cookies and without personal profiles. You don't need to accept any cookie banner — because there is nothing to accept. We also count anonymous page views to know which pages are used. Exactly what is counted, and what is deliberately not counted, is described further down.
What we don't do
- We use no tracking, analytics or advertising cookies. The two light/dark mode cookies are described further down and identify nobody.
- We use no cookie-based or profiling analytics tools — neither Google Analytics, Matomo, Meta Pixel nor similar. The two cookie-free measurement scripts are described below.
- We store no IP address, no user agent, no session and no device fingerprint. The only exception: if a question in the ask function breaks the rules, a salted cryptographic hash of the IP address is stored for 30 days, solely as an abuse block. The hash cannot be reversed into an IP address and is never used for saved pages, read status or any other personal feature.
- We share no data with third parties.
- We have no ads and no ad networks.
- All open facts, search, comparisons and maps require no sign-in. An optional free member account exists for saving pages and watching one municipality — it stores your email address, see the member account section below.
Cookie-free measurement scripts
sverigefakta.com loads two external measurement scripts from privacy-friendly analytics providers. Both are cookie-free and used for aggregated visit statistics — they set no cookies, build no user profiles, are not used for ads, and nothing is sold or shared onward. The scripts are not loaded in previews and not on internal pages under /intern.
We also measure the page's loading performance (Web Vitals), fully aggregated per day and page type, without IP address, user agent, ID or query parameters. The measurement respects the browser's Do Not Track and Global Privacy Control.
What belongs to your member account
Saved pages, read status and followed municipalities belong to your member account and are stored in our database, linked to your account id. They require sign-in, are visible only to you and are deleted when you delete the account. No IP address, no cookie-based visitor identity and no fingerprinting is used for this. Without an account nothing of this kind is stored, and all open facts, search, comparisons and maps still work.
What is stored locally in your browser
Everything below is stored only on your own device. None of it contains a user ID, is linked to your IP address or is sent to the server, and none of it follows you between browsers or devices. You can clear it all at any time via your browser's settings.
- Research Workspace — saved charts and excerpts live in the browser's IndexedDB on your device and never leave it.
- Light or dark mode — your choice is stored in
localStorageand in two functional cookiessf-themeandsf-theme-explicitso the page can render in the right mode immediately. They contain only "light" or "dark" and identify nobody. - Price base — the "Original prices ↔ 2026 prices" toggle is stored under
invnetto.inflationwith the value"original"or"today". - Preferences and dismissed notices — e.g. recently searched municipalities, the map guides and the editorial note you already closed. Only yes/no markers and your own choices.
- Tab marker for the counting —
sf.seeninsessionStoragewith the value "1", so the counting knows whether a page view was the tab's first. It disappears when you close the tab.
External links
When you click links to PDF reports (e.g. Ruist's ESO report or the National Institute of Economic Research's special studies), these open with the respective sender. What they log is governed by their own privacy policies — we have no insight into or control over that.
How we count page views
We want to know which pages are actually read, so effort is put where it helps. Every page view is therefore counted in an aggregated table. It also underlies the "Most read pages, last 7 days" list in the footer. The only things stored per page view are these six fields:
- The path (e.g.
/skattetrycketnever a query string or fragment. - Date (day, not time of day).
- Language —
svoren, derived from the path. - Traffic class — one of five fixed strings:
google,bing,social,direct,other. Den fullständiga hänvisningsadressen läses i minnet, reduceras till en av dessa och slängs. Den lagras aldrig. - Country code — two letters from the hosting provider (e.g.
se), orxxif unknown. No region, city or position. - Landing flag — whether the page was the first in the tab. A plain yes/no flag.
What is not stored: no IP address, no user agent, no session identifier, no user ID, no device fingerprint, no geolocation beyond the country code, and no full referrer address. The IP address is used only in memory for a short-lived rate limit and is never written to disk. The rows are pure sums per page and day — they cannot be linked into a visitor trail and therefore cannot be used to follow an individual person.
The counting respects your browser's Do Not Track and Global Privacy Control: if either is enabled, nothing at all is sent.
How we count shares
Pages have share buttons (X, Facebook, Reddit, copy link and save as image). When you use one of them, the event is counted in an aggregated table, the same way as page views. The only things stored are four fields:
- The path to the page — never a query string or fragment.
- Block ID — which section or footer the button sat in.
- Action — e.g.
link-copiedorimage-saved. - Succeeded/failed plus date (day, not time of day).
No IP address, session identifier or referrer address is stored. The rows are pure sums per page and day and cannot be linked to an individual visitor. So we see that a page was shared — never who shared it. If you click through to X, Facebook or Reddit, their own privacy policies then apply; we have no insight into what they log.
So the same share isn't counted several times when the page re-renders or when you navigate back and forth, a technical marker is stored in your browser's sessionStorage (path, block ID and action). It never leaves your device, contains no identifier, and is deleted when you close the tab.
Hosting & server logs
The website runs on a standard web server via Cloudflare. Like all web hosting, the hosting provider can technically log the IP address, time and which page was requested — this is necessary to deliver the content and to protect against abuse. We do not use these logs to profile visitors, and they are never linked to an identifiable individual by us.
Source code & transparency
All calculation happens in your browser, based on publicly available figures from Ruist (ESO 2018:3), the National Institute of Economic Research (Special Study 117) and Statistics Sweden. See Method and Sources for full documentation.
Purposes, consent and retention
The purposes are kept apart and are never mixed. The account stores an email address, an internal account id, verification status, language choice and an optional first name. The municipality watch is a service you start yourself and stores only the municipality code and frequency; it can be ended separately without deleting the account. An editorial newsletter requires its own explicit yes and never follows automatically from the account or the watch — newsletter mailings are not built yet. Payment does not exist, and no payment or billing data is stored.
Retention: accounts that never verified their email are deleted after 30 days, ended watch rows after 180 days, technical history of watch changes after 365 days and the chat abuse guard after 90 days. Your recorded choices are kept for the lifetime of the account and deleted with it. These periods are documented defaults that operations can adjust.
You see and change all of this yourself under My Sverigefakta: account details, active watch, newsletter choice, export of your data and permanent deletion of the account.
GDPR & your rights
Without a member account, Sverigefakta.com collects no personal data: the page-view counting stores only aggregated sums without any identifier, neither direct nor indirect. With a member account, the data described above is processed. You can then request, correct and delete it — all the way from My Sverigefakta, without contacting us.
What the hosting provider (Cloudflare) can technically log for a request — IP address, time, path — is processed only for delivery and abuse protection, and is not stored by us. If you have questions about this, contact us via /kontakt. For rights exercised against Cloudflare, see their own policy.
Changes
The date of the latest change is shown in the page header ("Last updated") and mirrored in the page's metadata. A full trail is available in the changelog.