Sourced through Scoop.it from: www.youtube.com
HURRAY! Today is September 23nd which means it’s Bisexuality Visibility Day! So I decided to turn to Google and see what people want to know about us! Wher
Sourced through Scoop.it from: www.youtube.com
Local aesthetic centre Only Aesthetic has been found to plagiarise an iconic UN campaign called “The Autocomplete Truth”. The spot was originally conceptualised by Memac Ogilvy & Mather Dubai and made headlines globally for bringing to the front line hard hitting truths about sexism and discrimination.
According to an article on Mothership, all traces of the plagiarised posts have been since taken down from the company’s Instagram. Marketing has reached out to Only Aesthetic for comment.
Images from the original campaign (below) were taken, edited and reposted by Only Aesthetic. One of the final results featured a slightly altered face, an autocomplete search of “How to get rid of acne?” and copy touting the clinic’s acne treatment solutions. The other image which was taken and edited in a similar fashion promotes the clinic’s fat reduction solutions.
Take a look at the campaign.
According to Memac Ogilvy’s website, the campaign reached 1.2 billion global impressions, with over 224 million Twitter impressions.
Sourced through Scoop.it from: www.marketing-interactive.com
Well, I have screenshots and I’ve been getting paid to manipulate Google’s search results for years versus getting paid to make sensational videos on YouTube to sell ads based on view count.
In a nutshell:
SourceFed believes Google is manipulating search results in favor of Hillary Clinton, because “Hillary Clinton cri-” did not return “Hillary Clinton criminal charges” and “Hillary Clinton in-” did not return “Hillary Clinton indictment.”
Let me share some interesting screenshots about Donald Trump that I found by searching for two of his most scandalous controversies as covered by The Atlantic.
Donald Trump was accused of rape by Ivana Trump during their divorce. She has since disavowed this, but it’s a real story in the news and many have searched for it. Interestingly, a Google search for “Donald Trump ra-” does not return anything about the rape. There are many mentions of “rally,” which makes sense, but “Donald Trump rap” has less trending searches in the last year and ranks where as “Donald Trump rape” does not.
Search trends are a significant factor in how Google Autocomplete works — this is the entire basis for SourceFed’s claim and yet according to their logic, Google must be favoring Donald Trump as well.
Google Trends screenshot for “donald trump rape” and “donald trump rap” searches
Google Autocomplete search results for “donald trump ra-”
In The Atlantic article there are numerous lawsuits mentioned, so I simply looked up “Donald Trump la-” to see what appeared.
Interestingly, there were no mentions “lawsuits” or “lawsuit,” but “Donald Trump laughing” did appear in the autocomplete search results despite having lower search volume according to Google Trends:
Google Trends screenshot for “donald trump lawsuits” and “donald trump laughing” searches
Google Trends screenshot for “donald trump lawsuit” and “donald trump laughing” searches
Google Autocomplete search results for “donald trump la-”
At this point you might be thinking to yourself, “whoa, you’re just choosing random words and searching on them!”
I’m doing exactly what SourceFed did…
I’m selectively choosing words that I think people should be searching on and I’m throwing them into Google and Google Trends to find exceptions that support my story. The difference is that I’m presenting the other side of the argument and I’d like to know what kind of agenda SourceFed has against Hillary Clinton and/or Google that they didn’t take literally two minutes to search for similar stories where Donald Trump was concerned.
The story isn’t that Google favors Hillary Clinton, it’s that Google is a complex algorithm that presents information in many ways and at this point that includes artificial intelligence!
Talking about the future of search technology is a very interesting story, but it’s not going to sell ads for SourceFed like this juicy topic has.
I’m not going to pretend that I understand everything about Google, because that’s impossible for anyone but Google’s engineers. However, I can say with certainty that Google does NOT favor Hillary Clinton.
How do I know?
Because we can simply search for her name to figure this out.
Google Autocomplete search results for “hillary clinton”
Maybe Google isn’t presenting the queries SourceFed looked up, because they assume that a human being has already typed in “Hillary Clinton” and seen queries related to “email” and “Benghazi” in the top five suggested autocomplete results before they decided to keep typing those other randomly selected terms into Google.
Is Google’s algorithm intelligent enough to know that “email” when mentioned with “Hillary Clinton” is synonymous with “indictment” and “criminal charges?”
I know this because if you select “Hillary Clinton email” after typing in “Hillary Clinton,” Google’s very first recommendation is “Hillary Clinton email charges:”
Google Autocomplete search results for “hillary clinton email”
Guess what else?
WAAAYYYYYY more people search for “Hillary Clinton email” than “Hillary Clinton criminal charges:”
Google Trends screenshot for “hillary clinton criminal charges” vs “hillary clinton email”
How many more people?
More people search for “donald trump rap” than “hillary clinton criminal charges:”
Google Trends screenshot of “donald trump rap” vs “hillary clinton criminal charges”
But why present the data in the context of other queries?
SourceFed had a a really compelling graph and sounded very intelligent, so it must be true! Who needs context when there’s anecdotal evidence about how they decided to search for two random queries out of literally millions of variations that Google has been able to process and return in more intelligent ways based on actual user behavior.
Data + context. It matters.
Let’s see if mainstream media bothers to do their homework or simply picks up this completely bogus story spreading it further.
Now for the real kicker…
Because SourceFed told you to look up these queries, they’ve just manipulated Google’s search results.
Think about that for a minute. Google Autocomplete is powered by user behavior, personalization, trends, and lots of other factors. By telling hundreds of thousands of people (and growing) to search for these queries, SourceFed has just sent Google data supporting a massive spike of interest in these terms.
It’ll be very interesting to see what happens with these queries from here.
As someone who has been paid to to manage online reputations and displace negative Google search results for years, I have to wonder if there was a different motivation behind this video, because it was either very poorly done or very strategically executed. Whatever the reason, I hope if you’ve read this far you now have a better understanding of how Google Autocomplete works and that this has absolutely nothing to do with favoring anyone.
This is taking off and there are some good points being made.
Fact: Google suppresses “rape” terms across the board, so that was a poor example by me. However, “lawsuits” is valid and the “racist” inquiry in the edit below still stands.
Just take my word for it, here’s a screenshot of Google Trends data for the two queries SourceFed looked up.
Let’s keep watching what happens. The real story is whether these continue to be filtered out after this huge surge in interest for both terms. If they do it could be because of prior suggested searches or that they’ve already started they will filter some queries based on the content those queries contain.
Edit: I’m starting to get overly critical responses about whether I bothered to look at Bing and Yahoo. Yes, I did, and the most damn evidence is present there! “Donald Trump ra-” doesn’t mention the rape, but it’s all over queries that include “racist!”
The story that matters is HOW Google Autocomplete works not that it favors anyone.
My belief is that they’re factoring in prior suggested searches you ignored plus filtering out queries that will lead to certain types of content. The latter is a suggestion from Kristine Schachinger during our 3am debate and exchange of related screenshots.
Sourced through Scoop.it from: www.youtube.com
Google’s Auto suggest function can be incredibly damaging to a person or brand, If Google suggests words such as “scam” and “reviews” and “complaints” in association with a brand or persons name, such as the following example:
According to Google, as much a 64% of search users will choose an auto suggest from the pop-up list, rather than completing their original search phrase. In the above example, neither “reviews” nor “complaints” would not be considered offensive in a take-down request, but these suggestions by Google demonstrate how businesses today stand or fall by Google.
Google has been sued on several occasions because of defamatory auto suggest results appearing through this function. Google has lost in some countries, but not in the United States.
If there are illegal or offensive autosuggestion terms appearing when your name, or brand is Googled, then Google does provide a form to submit a removal request from the Auto Suggest library.
Although, as per Google’s normal track record with Internet defamation, we do not hold out a lot of hope that they will approve your request. But, all you have to lose his time, the form is located here:
If Google will not remove the auto suggest phrase from the drop-down display, you can mitigate the damage being done to reputation by taking control of the resulting search engine result pages (“SERPs”).
Contact us to get started.
Sourced through Scoop.it from: defamation911.org
Google Suggest is the name of Google’s auto-complete function. If a user enters a letter or a word in Google’s search field, they are automatically shown associated terms in a dropdown menu. These suggestions are generated based on the most frequently searched terms. If a user enters the word “car”, for example, associated terms like “carmax,” “cartoon network”, “cars” and “car credit” are suggested.
This function helps users save time and may also provide them with additional information about the topic they are researching. Google Suggest is not only a useful function from a usability perspective, but it’s also essential for businesses and figures of public interest.
There are several potential risks and opportunities associated with Google Suggest, both from a marketer and a user standpoint.
How Did Google Suggest Emerge?
Pretty much everyone who searches online comes into contact with Google’s autocomplete algorithm. The feature has been around for 5 years and is mostly taken for granted nowadays. Most people, among them – unbelievably – search engine optimizers, neglect the following questions:
- How can I use Google suggest to my advantage?
- How could it potentially be harmful to my client?
Before we answer these questions, first lets explore how the Google Suggest function works.
Why Google Launched Auto Suggest
Basically Google’s auto suggest function aims to improve the usability of its search engine. Suggestions help the searcher save time entering long phrases like “Samsung Galaxy S4”. Thanks to Google Suggest, that term can be chosen after “Sams” is entered.
Another helpful aspect of the auto complete function is the checking of search terms for spelling mistakes. If the user enters “hairdesser” in the search mask, the search engine giant offers the term “hairdresser”.
Influence Factors of Google Suggest
Figure 1: Search rate, search manner, search frequency and search location influence the recommendations of Google Suggest.
Search Rate, Search Manner, & Search Frequency of Google Users
Here Google tends to be transparent. Official statements by the search engine giant stated that previously entered search phrases do influence the auto completion. For example, if “holiday mallorca” is often entered in the search mask chances increase that “holiday mallorca” will be suggested to another user when he or she enters just “holiday”.
There are a couple of other elements which are also considered. Although these “objective” factors haven’t been made public by Google, other aspects like the search frequency can easily be identified. When the interest for media events like Prince William’s and Kate’s wedding gradually fades, these search suggestions – e.g. when “wedding” is entered – are displayed less and less often. Constant terms like “plan wedding” on the other hand rise in popularity again after the furor of a big event has died down.
What Else Plays an Important Role for the Google Suggest Function?
Although it should be self-evident, for the sake of completion, the language used and the place where the search query is entered also play a role. In earlier iterations of the algorithm only the country was used, but now Google has been successfully differentiating between individual cities of a country. This has to be taken into consideration for SEO, since different search results are displayed depending on the place, and these are even dealt with preferably. The following example makes this aspect clear:
Figure 2: “Shopping basket” has a higher overall search volume than “shopping center london” Although the term “shopping basket” has a higher search volume than “shopping center london” the local search term is preferred if just “shopping” is entered in the Google search mask. Figure 3 makes clear that Google even differentiates between single districts in Berlin.
Figure 3: Local search queries have a high significance in the selection of Google’s recommendations.
However, these suggestions only seem to emerge if the user has entered his or her location in the settings and already did multiple search queries. Otherwise, the suggested terms are broader with the actual location being the first suggestion, no matter if it was determined in the settings or not.
Figure 4: Suggestions without any local search queries beforehand
It is important to note that when changing the settings one can only choose locations that do exist in the respective country. In order to get search results from other countries one needs to change the settings on the country-specific Google cover page. In order to verify our thesis, changing the location should lead to other suggestions in local searches.
Figure 5: Changing the location in Google’s search settings
And in fact, if we search for “shopping center” in google.co.uk, no districts are suggested, even if “London” was chosen as the location. The suggestions are identical to those that appear if no location was entered at all. Consequently, at google.com, the same terms are suggested no matter if with or without location (e.g. New York). Depending on the different Google cover pages (co.uk / com), however, we get different suggestions.
Figure 6: No districts are suggested in google.co.uk …
Figure 7: … and neither in google.com
One can therefore assume that it is the multitude of local search queries responsible for the search suggestions. After all, in the previous examples no city-specific districts were suggested but rather a thematically larger bandwidth of terms, even if those consisted mainly of different cities.
When it comes to mobile search, another phenomenon can be seen: Even though we searched with google.co.uk and chose London as our location, the first suggestion is “shopping center berlin” (our actual location).
Figure 8: Mobile devices know our actual location
Since smartphones always know where we really are, getting local search results from other locations seems to be impossible – contrary to stationary devices. Instead, local search suggestions always display a person’s present location – no matter what the settings are. It is very likely that this will be the case on stationary devices in the near future, too, and local search results will noticeably enhance Google’s suggestions.
Why then, for some search queries are no suggestions are displayed at all? This is due to the fact that Google follows strict guidelines when it comes to pornography, copyright violations, violence, and similar negatively charged terms. So the search engine provider automatically filters out terms that can be classified into the above-mentioned problem areas. On the other hand, it can take some time until breaking new is validated by Google and implemented as suggestions.
Auto Suggest: Chances and Risks for Reputation Management
Nowadays, how an enterprise is perceived online and in public strongly depends not only on a professional position, but also on internet users. Thanks to Twitter and Co., information and reviews spread like a virtual wildfire. Today companies are dependent on the opinion of a huge and partly anonymous crowd. This has effects on the auto completion of Google Suggest, for example as a result of usage of certain search terms over the course of time.
The problem example is this: Sometimes, Google associates a brand with fraud, scam, complaints or other negative terms. Since search behavior and search frequency affect Google Suggest, this phenomenon can be explained by short-run trends. For example, if a well know blogger writes a scathing (not necessarily true) blog. When this happens, potential clients can be turned off of the brand.
On the other hand, these same mechanisms can have positive consequences for a company. If terms like “experiences” and “opinions” are suggested after a brand name, this can lead a potential customer to read positive reviews and signal that others are interested in the brand.
Chances and Risks in Affiliate Marketing
An interesting train of thought how Google Suggest can have positive influences on the business model of an affiliate shop owner is the following:
Let’s assume a Google user searches for an online shop to get information on the latest events, etc. If this online shop offers vouchers and other users have searched for these vouchers before, the user will be suggested a combination of the chosen shop and the word “voucher”. This gives the user a whole new idea since he could even save money during the purchase! Affiliates who rank high for a voucher term in connection with a brand can be happy about this suggestion by Google. On the other hand, another shop owner may lose important traffic because of the suggestion.
Risk: Distracting the Searcher
The next example goes one step further. If a Google user searches for “credit card” , he has already made the decision to get one. However, while searching he stumbles upon the term “credit card free”, his intention might change. Therefore, the searcher ends up searching for free credit cards instead of a particular company he was initially interested in.
Unjustified Brand Advantages?
The following figure illustrates another risk, respectively, or an advantage for some companies:
Figure 9: When searching for “car”, the first suggestion is “carmax”.
Here the question emerges: To what extent is this suggestion justified? On the one hand, these brands’ websites are a good suggestion because they are often searched for. On the other hand, these suggestions influence the user and could cause distortion of competition. Consequently, one could question whether this was Google’s initial intention.
Furthermore, even if just the letter “e” is entered in the search mask, certain brand names appear. Although the searcher might be looking for something completely different, the fact alone that a brand is mentioned amplifies its level of awareness and potentially makes the user resort to these brands in the future.
Figure 10: Brand names are displayed after only one letter was entered
One of the most important results is the fact that Google Suggest influences the user. This happens consciously – “I’ll just take a look at what Google is suggesting to me” – as well as unconsciously. One reason is that the completion function has become a normal and even integral part of the Google search for most people. The above mentioned examples gave situations in which the auto completion function clearly influences our search behavior. It has become clear that this can be an advantage for some companies and a disadvantage for others. If Google Suggest puts your company at a disadvantage, contact a online reputation management company that has experience manipulating Google Autocomplete suggestions and removing suggested searches from the drop-down list.
Image 1: Screenshot taken 14/10/2013 of google.com
Image 2: “Influence Factors of Google Suggest” by Andre Alpar on 28/05/2013
Image 3: Screenshot taken 14/10/2013 of Google.com/trends
Image 4: Screenshot taken 14/10/2013 of Google.de
Image 5: Screenshot taken 14/10/2013 of Google.de
Image 6: Screenshot taken 14/10/2013 of Google.com/preferences
Image 7: Screenshot taken 14/10/2013 of Google.co.uk
Image 8: Screenshot taken 14/10/2013 of Google.com
Image 9: Screenshot taken 14/10/2013 of Google Search App
Image 10: Screenshot taken 14/10/2013 of Google.com
Image 11: Screenshot taken 14/10/2013 of Google.com
Sourced from: www.searchenginejournal.com