Happy posts about your product or service could lead to better search engine rankings. Image: Flickr / kirinqueen / CC-BY

For businesses, being listed well on search results usually equates to better business from both locals and tourists. A careful reading of patents held by both Microsoft and Google show that the search engines could be using “sentiment analysis” to rank these local search results.

Basics of sentiment analysis

“Sentiment Analysis” is a method of studying the words, phrases, and emotions behind statements that are made about a place, product, or service online. This is a way of studying the way individuals feel about something even when they don’t use star rating or numerical rating systems. Someone saying they had a “great dish at XYZ restaurant” or “ABC service was fast!” could be rated as “positive vote”. Someone saying they had a “terrible experience” or “waiting for hours” would be read as a “negative vote”.

Mining for sentiment

The specifics of how search engines incorporate sentiment analysis have not been shared, and likely will not be. However, it appears from analysis of search results that Google is pulling sentiment keywords from high-ranking sites that link to a particular place – a newspaper review of a coffee shop, for example. This means that sites that are already more highly-ranked by the Google algorithm will have more “power” in how they talk about a particular product or service.

Making use of sentiment analysis

For affiliates and small business owners, sentiment analysis can be very useful. It may be possible to game the sentiment analysis system, but gaming the system would be a proposition with very quickly diminishing returns. Instead, the sentiment analysis system can be utilized in much the same way as review systems. Ask customers to post reviews or responses to your product or service, or even offer a small incentive for the post. Keep in mind that unhappy customers are more likely to post than happy customers, so do your best to keep customers happy. Respond to customer inquiries quickly, and live up to the promises you make.

Read More:

Domain-Specific sentiment classification patent
SEOMoz
Extraction and summarization of sentiment information patent