The Power of Search Trends

Search engines are powerful sources of data. Obtained by analyzing user queries during a certain period of time, information is extracted by studying the repeated patterns in search engine’s input keywords and can tell us a lot about trends, fads or specific moments in time.

Virtually everything that occurs nowadays is registered on the Internet. Users typically use search engines as their starting point to navigate through the network. They rely on them to find relevant information in order to satisfy an immediate need. This immediate need will vary from time to time and can be (in general words) news, entertainment, pleasure or knowledge.

Analyzing top search engine input keywords allows identifying prevailing queries at a certain time. Google’s Hot Trends and Yahoo’s Buzzlist show the most popular search terms during the past hour. Popular terms can also be requested for a desired day. The fact that they are listed as popular does not mean that these are the terms with the highest amount of queries, since generic terms like “weather” would always pop-up, contrarily it means that these terms have suffered an abnormal rise in traffic. These lists are updated hourly, providing a good way to stay tuned, practically in real time, to what society is demanding.


If we look at this information during a period of time (e.g. a year) we can clearly see how the popularity of a term rises or decreases over time. This method can be used for anything that we may imagine: products, services or even people. If it has a name, it probably has been searched in a search engine before. By using Google Trends it is possible to obtain an estimation of the number of queries for a certain word throughout time.

Obtaining this information with surveys is very expensive and at the same time may reflect false results since people are not always willing to tell the truth about certain aspects of their life. There are certain things that people are unlikely to admit in a survey, situations in which we commonly do not respond frankly. Surveys often come up with an additional difficulty:  knowing what a large group of the population is thinking at any given time is much more complicated than simply asking them what they think. The question and the sample of people must be selected correctly. This is something that can be overcomed by analyzing search engine trends. For instance, consider people suffering depression. Before going to the doctor or telling anyone, most of us would normally hide it and search for help on our own. Google is a place where anyone can ask a question without the fear of being judged or having to reveal  our identity. By analyzing the amount of queries for the word “depression” it is possible to estimate the months in which people are more likely to suffer depression. However, results are not always reliable. If we search for a pattern with just the word “depression”, we will obtain a graphic that varies with the school year. Every year we can see that during the holiday season the amount of queries decreases substantially. This is due to the fact that many students search for “great depression” during this time of the year. In order to get accurate results we must combine the word “depression” with a common antidepressant treatment like Zoloft or Prozac. Between October and November is the time of the year when more people seek help from Google, although, Christmas also stands out. A query with the word “depressed” returns similar results.

     
     Scale is based on the average traffic of "depressed".

Search engines allow understanding human behavior and provide excellent information for market research. However, it is important to keep in mind that this data is not always fully trustworthy since it is gathered using estimations based on the number of queries and may not necessarily represent an insight into what we are looking for.

When do you think you are more likely to workout surrounded by a lot of people? In other words, when is a gym going to increase membership sales? A research into the word “gym” will return the answer: right before the beginning of the year and before summer.

    Scale is based on the average worldwide traffic of "gym".

Let’s look at the reasons behind this to better understand why this occurs: It makes reasonable sense that people want to look better during summer time, since they go to the beach and swimming pool and they expose their bodies publicly. What could be the reason for this behavior at the beginning of the year? Could it be New Year’s resolutions and intentions? Not surprisingly, keyword “stop smoking” shows similar results. This chart might change in the future, as people are increasingly aware being fit is a matter of taking care of oneself during the whole year.

Aggregate search behavior shows us when consumer demand peaks for various goods and services. Business can start planning their demand with just the click of a button. Which month of the year has the highest demand for wedding dresses? January. In which time of the year are people more likely to be looking for sex? Summer and Christmas; this is probably due to the fact that it is the time of the year when people usually have more free time. What about the day of the week? Saturday. When does the demand for karts increase dramatically in Spain? In Easter Holidays and during the summer season.

Research can be conducted per desired period of time or region. In fact, it is possible to compare terms in different regions to analyze the popularity of a product in different parts of the planet. Data is normalized by the total traffic of each region to cancel variables such as Internet usage statistics or demographics. The fact that two regions show the same percentage for a keyword does not mean that their absolute search volumes are the same. This way, data from regions with heavy differences in search volumes can be compared equally.

By searching for a brand name we can obtain information of brand awareness for a particular product or service, or even measure the effectiveness of a competitor’s marketing plan. Relating various brands will tell us how popular they are to one another. The following figure shows a worldwide comparison between Siemens, Cisco and Avaya, three world leader brands in Unified Communications.


     Scale is based on the average worldwide traffic of "Cisco".

The same comparison in the United States, where Siemens’ portfolio (not only in Unified Communications) and brand awareness is much smaller, throws up different results.


    Scale is based on the average traffic of "Cisco" from United States.

Another interesting example of a brand awareness measurement is Lenovo’s purchase of IBM’s laptop division including the famous “ThinkPad” brand. By analyzing search trends of the keywords “IBM ThinkPad” and “Lenovo ThinkPad” it is possible to evaluate Lenovo’s marketing efforts to generate brand awareness and link the purchased brand to its company name. This could be done with a survey; however, the process is expensive and much slower. Traditional survey-based brand studies will probably take several months before coming up with good results. On the other hand, search engine data is updated daily, reaches a much bigger population (instead of being just a sample) and is inexpensive.

   Scale is based on the average traffic of "IBM Thinkpad" from United States.

Watching consumer’s online behavior is crucial to create better marketing plans. Furthermore, changes in consumption habits can be quickly spotted within search volume research.