How does Google’s process work?
Individuals or their representatives must first complete this web form. The individual will receive an automatic reply confirming that we have received the request. We assess each request on a case-by-case basis. In some cases, we may ask the individual for more information. Once we reach a decision, the individual will receive an email notifying him or her of our decision and, if we do not delist, a brief explanation.
Who may request a removal?
Individuals can request delisting of search results according to European data protection laws. We also allow people to make requests on behalf of others, so long as they can affirm that they are legally authorised to do so.
Who makes decisions about requests to delist content?
Google LLC staff make the relevant determinations. We have a team of specially trained reviewers for this purpose, based primarily in Dublin, Ireland. Our team uses dedicated escalation paths to senior staff and attorneys at Google to adjudicate on difficult and challenging cases. As of 1 November 2015, just over 30% of requests had been escalated for a second opinion.
How do you evaluate requests?
We have carefully developed criteria in alignment with the Article 29 Working Party’s guidelines. After a request is submitted to us via our webform it undergoes a manual review. There are no categories of requests that are automatically rejected by humans or by machines.
Our evaluation process consists of four steps:
- Does the request contain all the necessary information for us to be able to make a decision?
- Does the person making the request have a connection to a European country, such as residency or citizenship?
- Do the pages appear in search results for the requester's name and does the requester's name appear on the page(s) requested for delisting?
- Does the page requested for delisting include information that is inadequate, irrelevant, no longer relevant or excessive, based on the information that the requester provides? Is there a public interest in that information remaining available in search results generated by a search for the requester’s name?
If an individual files a request with us that does not contain sufficient information for us to make a decision, we may ask for supplementary information to support our evaluation.
In order to demonstrate the scope of removal requests, we have included a section of request summaries on the Transparency Report.
Are you delisting pages wholesale from your search results?
No. Pages are only delisted from results in response to queries relating to an individual’s name. So, if we granted a request to delist an article for John Smith about his trip to Paris, we would not show the result for queries relating to [john smith] but we would show it for a query like [trip to paris]. We delist URLs from all European Google Search domains (google.fr, google.de, google.es, etc.) and use geolocation signals to restrict access to the URL from the country of the person requesting the removal.
For example, let’s say we delist a URL as a result of a request from John Smith in the United Kingdom. Users in the UK would not see the URL in search results for queries containing [john smith] when searching on any Google Search domain, including google.com. Users outside of the UK could see the URL in search results when they search for [john smith] on any non-European Google Search domain.
What happens if an individual disagrees with your decision?
If we decide not to delist a URL from our search results, an individual may request that a local data protection authority review our decision.
Are you letting website owners know when content has been delisted?
It is Google’s policy to notify a website owner when pages from their site have been delisted from our search results based on a legal request. We do this in the interest of transparency. In order to respect the privacy of the individuals who have made removal requests, we only send the affected URLs, not the requester’s name.
Do website owners have any way of challenging your decisions?
Website owners who receive our notice of delisting through Google Search Console may request that we re-review a decision.
Does this ruling also apply to other Google Search services like Image Search?
Our process currently covers delisting of results from our search properties, like Google Search, Image Search, Video Search and Google News.
What are some common scenarios for delisting pages?
Some of the most common material factors involved in decisions to delist pages include:
- Clear absence of public interest: For example, aggregator sites with pages that contain personal contact or address information, instances where the requester’s name no longer appears on the page and pages that are no longer online (404 error).
- Sensitive information: Pages with content that relates solely to information about someone’s health, sexual orientation, race, ethnicity, religion, political affiliation and trade-union status.
- Content relating to minors: Content that relates to minors or to minor crimes that occurred when the requester was a minor.
- Spent convictions/exonerations/acquittals for crimes: Consistent with local law governing the rehabilitation of offenders, we tend to weigh in favour of delisting content relating to a conviction that is spent, accusations that are proved false in a court of law or content relating to a criminal charge of which the requester was acquitted. We also consider the age of this content and the nature of the crime in our analysis.
What are some common scenarios where you do not delist pages?
Some of the most common material factors involved in decisions not to delist pages are:
- Alternative solutions: There’s another avenue for the requestor to delist that page from our search results. For example, a requester may have published the content to a site that allows users to prevent the content from appearing in search results. We point requesters to information about these tools when we can.
- Technical reasons: An incomplete or broken URL is a common technical error. Requesters also sometimes ask us to delist pages for a query that doesn’t match his/her name or the name of the person the requester claims to represent.
- Duplicate URL by same individual: A requester submits multiple requests to delist the same page for the same name.
- Strong public interest: We may decline to delist if we determine that the page contains information which is strongly in the public interest. Determining whether content is in the public interest is complex and may mean considering many diverse factors, including, but not limited to, whether the content relates to the requester’s professional life, a past crime, political office, position in public life or whether the content itself is self-authored content, government documents or journalistic in nature.
Can you provide more detailed statistics about the nature of these requests and removals?
We have provided statistics about the scale of our delisting process – updated daily – since October 2014 in this Transparency Report and have added anonymised examples of delisting decisions to provide colour. Additional data on common material factors is available for download here. A more in-depth report on data from 2014–2017 is available here. We continue to explore ways to provide more transparency into delisting decisions in an operationally efficient manner and with due regard to the sensitive and private nature of the requests.
Why do some of the charts say that the data dates back only to January 2016?
Following the court’s 2014 ruling, Google worked to get a process in place to evaluate requests as quickly as possible. Because this was a completely new process, we did not yet know what information we would ultimately want to track and collect about requests. Over time, we refined and improved our process and the amount of data that we record about each request. In January of 2016 we rolled out a new internal process based on our experience to date.
How do you classify a requester?
The requester categories are used to classify individuals who submit requests to delist – content from Search – for example, whether the requester is a private individual, minor, corporate entity or public figure. We make this determination using information provided by the requester, such as their rationale for us to delist, as well as publicly available information, such as the content at the URL.
What are the various requester categories?
- Corporate entity: Requester is filing on the behalf of a business or a corporation.
- Deceased person: Requester is filing on the behalf of a deceased person.
- Government official: Requester is a current or former government official or politician.
- Minor: Requester is under the legal age of consent.
- Non-governmental public figure: Requester is known at an international level (e.g., a famous actor or actress), or has a significant role in public life within a specific region or area (e.g. an academic who is well known in her field).
- Private individual: None of the other categories apply to the individual.
How do you classify page content?
The page content categories classify the content that appears on the individual pages submitted in RTBF requests – page content relates to the requester’s professional life, is self-authored or references criminal activity. We assign this category based on review of the content at the URL requested for delisting when we evaluate a request.
What are the various page content categories?
- Crime: Page content references the requester in connection with crimes. For example, content may relate to the requester’s conviction, witness testimony or victim status.
- Name not found: No reference to the requester's name can be found in the content page at the provided URL. However, the individual’s name may appear in the URL.
- Insufficient information: Page content was not categorised because Google required more information to process the request. For example, the requester supplied an incomplete URL or did not provide a reason for requesting that a URL be delisted.
- Miscellaneous: Page content does not fit into any of the other content categories.
- Personal information: Page content contains the requester's personal address, residence or other contact information; images and/or videos of the individual; or other types of non-sensitive personal information.
- Political: Page content contains criticism of a requester's political or government activities or information that is relevant to the individual’s public political history, platform or profile.
- Professional information: Page content contains a requester's work address, contact information or general information about their business activities.
- Professional wrongdoing: Page content references a criminal or court activity — crimes, acquittals or exonerations — specifically in relation to a professional role.
- Self authored: Requester authored all or some of the page content.
- Sensitive personal information: The page content mentions the requester's medical status, sexual orientation, creed, ethnicity or political affiliation.
How do you classify websites?
The website categories classify the website that hosts a page that we’ve been asked to delist – e.g. a news site, a social media site or a government site. This category is determined by reviewing the site during our evaluation of an individual’s request.
What are the various website categories?
- Directory: The web page is hosted on a website that serves as a directory or aggregator of information, such as postal addresses or phone numbers for businesses or individuals.
- Government: The page is hosted on an official government website; a site that references or contains government, business or legal records; a website that is the official media outlet of the government (note that public media, e.g. BBC or PBS are not included here).
- News: The page is hosted on the website of a non-government media outlet or tabloid.
- Miscellaneous: The page is hosted on a website that does not fit the other categories.
- Social Media: The page is an account profile, photo, comment or other content hosted on an online social networking site.
Why do some URLs with page content 'Name not found' appear as delisted?
When a requester’s name does not appear in content on the page in question, we may take steps to prevent this page from showing in Search results returned for queries containing the individual’s name.
Why do some months show a value of 0 for volume of URLs requested?
Google did not begin recording detailed data at the URL level until mid-June 2014.