At DuckDuckGo, we don’t track you, ever. That’s our Privacy Policy in a nutshell. To improve our product we develop completely anonymous ways to understand how it is being used. For example, all requests to improving.duckduckgo.com cannot be associated with any individual. We do not create unique cookies and, more generally, we architect our product and these systems so that we do not even have the ability to create a search or browsing history for any individual — it’s privacy by design. We created this page to explain how we improve our product.
First, you might notice that when you search DuckDuckGo, there may be an “&atb=” URL parameter in the web address at the top of your browser. This parameter allows us to anonymously A/B (split) test product changes we make to DuckDuckGo. For example, users in the A group would get blue links and users in the B group would get red links, and we’d be able to measure how usage of DuckDuckGo is impacted by different color links. For some experiments, we may need to ensure a consistent user experience across additional pages after you interact with a page element. In those cases, we may store non-personal data (e.g., the large and generalized user experience group you’re in, using local storage). But that’s only after your first interaction, only on your device, and never by default.
Second, we measure engagement of specific events on the page (e.g., when a misspelling message is displayed, and when it is clicked). This allows us to run experiments where we can test different misspelling messages and use CTR (click through rate) to determine the message’s efficacy. If you are looking at network requests, these are the ones going to the one-pixel image at improving.duckduckgo.com. These requests are anonymous and the information is used only by us to improve our products. We also occasionally work with browsers to similarly run completely anonymous experiments in order to improve the DuckDuckGo Private Search experience within that browser.
Third, you may notice similar atb.js or exti requests made by our browser extensions and mobile apps. These requests help us anonymously count how many devices are using an experience. For example, our browser extensions and mobile apps will send an atb.js request with each search made. These requests allow us to count approximately how many devices accessed DuckDuckGo without knowing anything about those devices or the exact searches they made.
If you’re using the DuckDuckGo app on Android version 11 or newer, you may notice a new folder in the Downloads directory named “DuckDuckGo.” Currently, we don’t store anything in the “DuckDuckGo” folder, though we might in the future. For example we might use the “DuckDuckGo” folder as the default location for files you download online using our browser. Right now, this folder just helps us estimate how many people are trying DuckDuckGo for the first time in a totally anonymous way — we simply check to see if the folder exists already, which means our browser has likely been used on that Android device before. Because hundreds of thousands of people install our Android app every day, it’s impossible for us or anyone else to use this folder to identify you or connect your behavior in the DuckDuckGo app to you personally. The “DuckDuckGo” folder is only created once per install, and you can delete it at any time.
Fourth, for certain elements in our search pages you may notice the &vis parameter being used in requests. This parameter allows us to measure usefulness by identifying which elements were actually visible on a page or received an interaction like hovering. Because of different screen sizes, visibility can be very different across devices (e.g., desktop vs. mobile vs. tablet). As users click on these elements, we use this interaction data to measure conversions effectively. The parameter values are anonymous, grouped together in batches, and are devoid of any personally identifiable information.
Fifth, while searching on DuckDuckGo, you may see &perf_id and &parent_perf_id parameters being used in certain network requests. These parameters allow us to measure the performance of essential components of a search engine, including performance of database and cache calls, fraud detection and rate limit functions, as well as return time and results processing upstream calls within queries. We use this data to help identify opportunities to improve the search experience, for example, to identify how we can deliver high-quality answers faster. The parameter values are randomly generated per search and are not associated with users, user agents, or IP addresses.
Finally, to monitor the success of certain features, we may count the number of requests for design elements hosted on metrics-content.duckduckgo.com – for example, a particular type of button illustration. This domain and process only counts the number of times these elements were loaded; in other words, it is a privacy-respecting proxy that does not capture any identifying information.
In fact, none of the above types of requests have personally identifiable information embedded in them, and we do not store any IP addresses associated with them. This is the case with all other requests to DuckDuckGo. And all these types of requests are fully encrypted in transit to DuckDuckGo so that nobody else can see them.
To be clear, this means we cannot ever tell what individual people are doing, since everyone is anonymous. That is a limit of our statistical methods, since we are private by design. With these anonymous processes, however, we can still reliably determine what changes are generally working.
We have developed these systems from scratch, instead of using third-party services, because of our privacy promise not to track you. We’re proud to be able to improve our site for you in a completely anonymous way!
If you have any concerns, please feel free to contact us.
Prior versions on GitHub.