Categories
Suggestions

Where do Search Engine suggestions come from?

Search suggestion providers have stated that autocomplete suggestions come from what keyword phrases people are searching on. In fact, Google is has one of the few patents related to search suggestions.

While that is clearly true that Google is using keyword search data, we have speculated there is more to the recipe than just frequency. The general order that many in the business believe Search Engines (eg: Google) use:

  • Frequency: How often is this searched on? Including time-of-day.
  • Popularity: Unique people that have conducted this search.
  • User Behavior: “bounce backs”: How many times was this suggestion repeated.
  • Clicks: How many clicks has this suggestion received.
  • QDF Query Deserves Freshness. A phrase Google coined to indicate a search should surface newer content.
  • Geographic Area: If the search engine can determine your location, it offers suggestions related to where you are at. ex: “best restaurants in” will suggest locations in that city or town.
  • Language: Results are different for non-English languages.

While that may be true for some search services, we do know that it is not entirely the case. Google says:

Autocomplete is a time-saving but complex feature. It doesn’t simply display the most common queries on a given topic. That’s why it differs from and shouldn’t be compared against Google Trends.

Clearly they are not a one-to-one agreement with searches or Google Trends search data.

Filtering the Suggestions

First, they do filter out suggestions based on offensive language (aka: swear words). They also investigate user reported suggestions.

Second, Google personalizes results in various ways. Many have tried to determine just how Google personalizes results, but most people feel it is a hit-or-miss proposition. Google tends to feed you what you have already seen in your space. In fact, one of the statements in Google patents suggest that:

user relative to documents provided to the user in response to previous queries

(- USPOT)

You also can turn off trending searches from your Google settings and then search settings under
“Autocomplete with trending searches” and select Do not show popular searches. That only affects your searches if you are logged into a Google account.

You may also turn off ‘related searches’ from your Google account user settings.

  1. From Google, go to your user Settings
  2. Then Search settings
  3. Then Search customization
  4. Then turn off Search customization.

Monetizing Search Suggestions?

Having looked at an ocean of keyword suggestions, we are slowly starting to believe that suggestions are morphing into new territory – the area of monetization. When we are looking at a group of keyword suggestions, many of those phrases are ending on SERP’s (Search Engine Results Pages) that are advertising heavy and/or Search Engine domain heavy. We believe suggestions are being used to direct users to more profitable SERP’s. It seems the more generic the term, the more the term features advertising. We have not made a final judgement on this yet, but search engines exist to make money.

“Advertising funded search engines will be inherently biased towards the advertisers and away from the needs of consumers” – Larry Page, Sergey Brin The Anatomy of a Large-Scale Hypertextual Web Search Engine

Web Related Links

Categories
Suggestable Suggestions

Discovering User Intent

User Intent refers to the users motivations or goals for performing a search query. Some traditional intents:

  • Navigation or Website query: searching for a website based on keyword
  • Spelling queries: often misspelled
  • Information or Know query: Searching for information or research for any topic
  • Shopping or Do query: Both for product information
  • News: What is new
  • Discovery: Looking for something fun and new
  • Transactional: “best deals on tvs”

You can find more about User Intent with relation to Google, by reading their “Raters Guidelines” that are sent to Google website raters. Section 12.7 Understanding User Intent.

Further Reading on Search Intent