Google’s New Offence Guidelines

Google are aware of problems with inappropriate and misleading content on their search side, and they’ve recently been taking steps to combat the rise of offensive and inaccurate sites padding out browsers’ search results. New guidelines have been issued to their valuable human Quality Raters (whose feedback informs Google’s much vaunted algorithms) helping them to identify sites issuing offensive content, extremist views and the Fake News which has dominated debate recently.

Google is aware it’s walking a fine line: it’s ideal is to streamline the browser’s experience rather than apply censorship. As Paul Haahr, one of Google’s lead search engineers has explained “Maybe on the inaccurate side, they like satire sites or maybe on the hate side, they hate people. Google should not prevent people from finding content that they want.”.

Google’s approach is to introduce its raters (a 10,000 strong worldwide army of search result assessors who conduct searches, grade the results for usefulness and feed back to Google) to a new tag for search results: “upsetting-offensive”. This allows them to mark results as unhelpful if they promote hate against specific groups, which in time will lead to these results slipping down Google’s rankings, and making people far less likely to encounter such content. It also includes instructions to help people make nuanced decisions, for example sorting between sites parroting racist views, and reputable, scholarly sources that discuss the history of racism without promoting it.

While its human Quality Raters start to establish a baseline for this content, Google’s engineers are working on technology that can interpret the intent of someone’s searches, so if they are trying to find such extreme material (for any reason) it will remain available to them: Google are not in the business of trying to police the internet, merely making sure its users enjoy a frictionless experience that doesn’t unexpectedly confront them with explicit material. The curious can find Google’s 200 page Quality Rater Handbook here.

Given this and other issues Google has recently found itself in hot water over it’s important to remember that people are getting more discerning about what they see online, from their search results to their advertising. The Stop Funding Hate Campaign has also drawn attention to the wider social issues involved and has motivated boycotts and letter writing campaigns. Brands need to be increasingly aware of the damage that can be done by an advert in the wrong place. In 2017, it appears, there is such a thing as bad publicity.

Programmatic is a great solution that makes sure your ads are kept where the right audience will see them, without consuming all your resources with manual bidding for space. It’s important to make sure your programmatic service has safeguards in place to make sure you can block your ads from sensitive sights. Geode, Illuminate’s own Native platform has, needless to say, a robust combination of automated filters and proactive human operators that can block ad content by URL and keyword that keeps your brand from associations you don’t think are healthy.