Google is constantly refining its search engine algorithm in an effort to root out spam and improve the search experience for users. Over the years, this refinement process has largely rewarded those employing ethical search engine optimization (SEO) practices, as it is no longer easy to “cheat” the system with manipulative, blackhat SEO tactics.
Yet there is one area where search itself still seems very primitive and inequitable: autocomplete.
Autocomplete (or “autosuggest”) refers to the search suggestions that appear as you type a query into the search box (shown above).
According to Google, these suggested searches are “a reflection of the search activity of users and the content of web pages.” This is a bit vague, but it would be reasonable to assume that if searchers are frequently searching for a term, or if a term is frequently appearing on the web, then that term is more likely to appear in autosuggest.
The search engines have been opaque as to whether clicks on autocomplete terms also factor into their algorithm. Since these clicks initiate a web search, I think there’s sufficient reason to believe they do influence autocomplete suggestions — and if that’s the case, then having terms appear in autocomplete becomes something of a self-perpetuating cycle.
As you might imagine, keyword combinations like [brandname scam], [brandname complaints], and [brandname lawsuit] can have a negative impact on a business’ reputation. Businesses that get bad press, receive a negative review, or have been the victim of a defamation campaign, can be dogged in autocomplete by terms that seem geared to reinforce anything negative about them — true or not.
The problem is that negative results — in autocomplete or in SERPs — can often be self-perpetuating. While it is unclear whether or not search rankings directly impact autocomplete, one could assume that content with higher visibility on search engine results pages (SERPs) is likely to have a greater number of searches associated with it.
Let’s say, for example, that a new piece of content is introduced about your business — it can be a Yelp review page, a court document about a lawsuit launched against you, or a page from an anonymous rant site.
Google frequently tests the “interestingness” of new content by randomly popping it up on the first page of search results to see if it gets more clicks than stuff already there. So, when a scandalous-looking page pops up there, you’d better believe it gets the clicks!
Once negative content begins getting the curious clicks from the rubberneckers on the information superhighway, then it tends to stick up there where all can see, attracting even more clicks — a snowball effect.
No matter how old, or false or incorrect the information may be, the curious clicks can propel it into staying visible. Once there, it can then attract people to search for it again later — thus influencing future autocomplete suggestions.
Certainly, in an open marketplace, it’s generally good for consumers to have access to more information so they can make informed decisions! (Indeed, there are high-profile examples of people who would like to be forgotten, yet which may not merit it from the perspective of public interest.)
But, it’s frequently not a level playing ground, as everyone knows. If the information dogging you includes the lies of a competitor, an ex-employee with an axe to grind, or some resolved past history from a decade ago, then it’s not really informing the public about who you truly are, right now.
Frustratingly, search engines seem unwilling to acknowledge that negative autocomplete suggestions can often create or fuel an unfair prejudice against a business or individual, as there’s no real guidance on how to influence autocomplete results.
In fact, search engine representatives state categorically that they consider it “abuse” if you attempt to “artificially” manipulate autocomplete results.
To be fair, Google and Bing both have suppressed a number of terms from appearing. Words like “nude” (both), “scam” (Google shows much less frequently now), racist and obscene terms. (See “Sex, Violence, and Autocomplete Algorithms: What Words do Bing and Google Censor From Their Suggestions?“)
And for Autocomplete suggestions that violate Google’s policies around hate speech, racism, religion, sexual orientation and such, you can report offensive suggestions here. Otherwise, they make no mention of any other recourse.
Still, this hasn’t gone far enough to really fix the issue. It’s great that they removed racist epithets, but what if it regurgitates defamatory claims like “violence,” “herpes,” “arrest,” “scandal,” “lawsuit,” “class action,” etc.?
Where listings in the search results are concerned, the general guidance from search engines is to work on promotional efforts to help positive content outrank negative content.
Additionally, search engines also allow individuals to send in takedown requests for damaging materials — you can provide a court order establishing that some content was harassing or false defamation or somesuch, and Google and Bing frequently will remove such items from their search results. This is actually quite generous considering that they aren’t legally required to do so.
But, there doesn’t really seem to be a takedown request for autocomplete, even if removal of damaging content from SERPs does begin to help in a process to displace such stuff.
So, what can one do about it?
WARNING NOTE: There’s apparently no method of influencing autocomplete that’s acceptable to the search engines at this time.
While it’s unclear what, if any, penalty there could be if they were to detect you attempting to manipulate results, imagine possible worst-case scenarios: they might freeze the autocomplete results involving your name so that any negative stuff is permanent; they might opt out your name from appearing in suggested results; they could even penalize your website for a period of time.
I want to make it clear that I am not endorsing such tactics. I report upon this as an interesting aspect of online reputation, search marketing and local search marketing.
As far as I’ve been able to tell, the following are the only options you might have for influencing Autocomplete that do not explicitly violate Google’s Terms of Service:
This is probably the least-objectionable, or most “white hat,” option. In fact, the practice is considered acceptable in and of itself — it’s your motive that defines whether this would cross Google’s or Bing’s terms and conditions.
You’ve probably heard audio and video ads that said something like, “Search for us at _____!” or, “Google _____ to find us!” Many businesses find this easier than spelling out a website address, and the bonus for reputation management cases is that you can craft a custom search query that’s bound to get some level of search volume.
This option has the added advantage of being beneficial in terms of promotion regardless of any effect upon autocomplete. In your ad, you could ask interested consumers to search for your desired keyword phrase with your business name included at the end. This would theoretically start building in local prominence for autocomplete to suggest your business name on the end of a search phrase when people begin typing it in.
For example, let’s say you’re an attorney, and when people in your area begin to type in a phrase like [personal injury], the first phrase autocomplete suggests is [personal injury shyster & shyster]! Imagine [florists smith], or [restaurants jones cafe]. This is why I call autocomplete optimization “hijacking” — those queries could predispose the search results to display your website and marketing content first and foremost.
For individuals experiencing serious online reputation issues, this may be the best option, assuming the total volume of searches for your name is very low — you could ask family and friends to help you by searching for positive keyword phrases associated with your brand name (for good measure, have them click on search listings for properties you own, like your website and social media profiles).
Companies could ask their employees to go home each night and conduct such searches, too. Remember, it could take some time for the pattern and volume of searches to have an effect.
On that note, make sure you stop conducting negative searches! I’ve looked into a few cases where I believe that an individual or company contributed to negative terms appearing for their name searches because they (and sometimes their attorneys) kept searching with those phrases to locate bad stuff their haters might be saying about them. You just can’t do that!
Your searches, including negative, terms may be feeding the data that keeps those negative terms prominent in Autocomplete/Autosuggest. If you have to keep tabs, consider using some monitoring software — Google Alerts, mention, trackur, Radian6 and others can all provide various keyword monitoring solutions.
If you don’t have enough family, friends or employees to displace the search suggestions, you could crowdsource it through a service like Mechanical Turk, just as Danny Sullivan has reported previously in “How Google Instant’s Autocomplete Suggestions Work.” It should be noted that this may violate Mechanical Turk’s terms of service.
Use the Übersuggest tool to research the top Autocomplete/Autosuggest terms associated with your name. It’s likely that nudging up terms already associated with a keyword phrase would be easier than trying to introduce something new and alien into the wild.
You can also read Brian Patterson’s article, “How Google Autocomplete Can Affect Your Brand’s SEO & ORM Strategy,” which lists some of the top terms associated with company names, such as “careers,” “jobs,” “locations.” For local businesses, I know that the city name or city and state abbreviation are often in the top list, as well. Example: [argent media dallas tx].
In general, I’m uncomfortable with putting forward any sort of suggestions that contravene search engines’ guidelines. In a majority of cases, playing by the rules is going to be more sustainable over the long term, and best practices often have other ancillary benefits that are not merely search-marketing related.
But, what does one do when there simply is no other recourse? Google and Bing have provided no real guidance as to how one might unravel negative terms or to position one’s self best where Autocomplete/Autosuggest are concerned. At least, that’s the present situation.
With the EU applying increasing pressure around the right to be forgotten, I think we’ll probably see more laws enacted that limit what and how items will be suggested in conjunction with the names of businesses and individuals. It would be great if the search engines would dial this back more in some way, or perhaps allow those with distinctive names to opt out of the treatment, perhaps.
Leveraging Autocomplete / Autosuggest SEO for promotional ranking benefit is less acceptable compared with merely attempting to sanitize one’s name of defamatory attacks.
However, there is still a gray area involved, because it may be a good idea in some cases to attempt to solidify positive suggest terms proactively to reduce vulnerability to potential reputation-damaging attacks – or even to proactively use the means to jockey for position in the wake of recent changes before other aggressive competitors due, such as some who could be motivated by the recent Pigeon Update.
Regardless of where you find yourself on the scale of black hat vs. white hat, it appears some companies are using these means to promote themselves at some level, and in a niche where there’s no alternative option provided by the search engines aside from “deal with it,” companies will have to decide whether they may need to “fight fire with fire” versus keep their hands perfectly clean.
The post Using Autocomplete To Hijack Local Search Results & Improve Online Reputation appeared first on Search Engine Land.