Google Trends as a use case for Digital Analytics
Google Trends is a public web facility of Google Inc., based on Google Search, that shows how often a particular search-term is entered relative to the total search-volume across various regions of the world, and in various languages.(Source:Wikipedia.com)
Google Trends is one of the best available source for online sentiment analysis and also a simple but efficient tool for Digital Analytics. There are a lot of interesting implications of the data that Google Trends throws up, for marketing strategy and business plans.
For example if I am interested in the two-wheeler automotive industry then I would want to search on the term scooters and motorcycles as they are the two most important product lines for this particular industry. In the Google Trends home page, I compare the search terms scooters and motorcycles and under the region tab, I select India and further under the categories tab, I select Autos & Vehicles and simply press enter- and Voila!! a neat graphic emerges. The results can be interpreted under the following sections:
i) Change in interest over time: On the x-axis we have the time values and on the y-axis we have the numbers for the search interest which represent search interest relative to the highest point on the chart for the given region and time. A value of 100 for the search interest is the peak popularity for the term. A value of 50 means that the term is half as popular. Likewise a score of 0 means the term was less than 1% as popular as the peak.
For the period 1 Apr 2012 to 1 Apr 2017 the graphs for the scores of the search terms for both scooters and motorcycles have shown a decreasing trend and importantly this is corroborated by actual sales data- the scooter and motorcycle sales were growing at a very fast clip in 2012 in percentage terms which has slowed down in the recent years. The internet search volume on Google is directly correlated with actual sales numbers within the particular product line.
The scooter graph leads the motorcycle graph except for a few points and this has important implication for the product mix in the two-wheeler automotive category. The percentage of scooters sold for every hundred two-wheelers has been constantly rising in the past five years and now scooters would form around 30% of the total sales pie whereas earlier it would have been in the ratio of around 15%. This is quite significant for an industry which has seen a tepid growth except for the segment of scooters. (It is a different matter that for worldwide search the motorcycle graph would lead the scooter trend and again this depends on the particular regional consumer preferences.) Also you can get to know the sales seasonality by studying the peaks and troughs- in the festive season of Diwali around the month of October you will find the peaks. Also a peculiar Indian phenomenon is that around the month of December you will find a trough as consumers generally don’t buy during this period as the model year changes in the coming January so the perceived resale of the vehicle comes down!! Also if you want to analyse the effects of the recent demonetization in India, just scroll over to November 2016 and you will find one of the lowest troughs for the internet search volume for both scooter and motorcycles.
ii) Regional interests:
The states of Goa(100), Kerala(90), Puducherry(75), Meghalaya(72) and Karnataka(63) have very high search scores for scooters which is corroborated by the actual sales data- all these states have a very high proportion of scooter sales in the total two-wheeler sales.
iii) Related Queries: Under the related queries section we have the option to see the top searches or rising searches. Top searches are the terms that are most frequently searched with the terms one enters during the same search session within the chosen region and category while rising searches have the most significant growth in volume during the requested time period. Both top search and rising searches have different marketing implications.
The top search data gives a rough idea of the market shares of the different products of the manufacturers. Under related queries if we see the top 10 entries for scooters, Honda and Yamaha stand out which is also seconded by their sales numbers for scooters.
In the rising searches list for scooters, all the products have ‘breakout status’- it means that the search term grew by more than 5000%. The rising search data is closely related with the marketing effort by the company- generally the companies whose marketing campaigns have been a success score highly in this list.
So in a way through Google Trends you can measure the effectiveness of your marketing campaigns also. For example, the high scores of TVS Jupiter scooter can be attributed to the successful marketing campaign for the product with Amitabh Bachchan, a leading filmstar of India.
If we see the motorcycle data for top searches and rising searches, the list of top 10 terms appearing in the rising searches does not correlate with the searches in the top search list, whereas for scooters the two lists do match. The top searched product in motorcycles is UM motorcycles Renegade model whereas this model does not have the maximum market share.
This throws light on the difference between the consumer buying behaviour between the two products of scooters and motorcycles. Whereas the majority of motorcycles sold in India is of entry-level commuter type vehicles for which the customers generally do not search on the internet but for scooters the customers who are searching the product are also importantly buying the product. This has an important bearing on the planning of the media mix for the products- scooters should have more of a digital content whereas for entry/commuter motorcycles the media mix should be tilted toward more ATL activities like newspaper ads, hoardings etc. For motorcycles for premium brands like UM Motorcycles, Harley etc the advertising should be mostly through digital media.
Remember those questions from the case interview preparations? The analysis above can be used to answer the case questions on industry analysis and entering a new market (assuming the interviewer allows you to use your mobile along with wi-fi !!!) and by extension can be applied to any real time situation also-
- What is the growth rate of the industry?
-Based on current and past data, does the future of the industry seem worth entering into?
-Region wise/market wise entry potential
-Seasonality of the sales cycle
-Where is the industry/product line in its life cycle stage of development( Emerging, Mature, Declining)
-Who is our competition and what size market share does each have?
Also important insights on the effectiveness of the marketing blitzkrieg and consumer behaviour can emerge from sifting the data judiciously by doing sensitivity analysis by varying the timelines and regions. Further, by downloading the Google Trends data in a .csv file and using R programming, the data can be sliced in various other meaningful ways.
In Google Trends we have a freely available important data set for all of us to analyse in this digital age- as one has rightly said that digital marketing is nothing but marketing in the digital age — the core marketing strategy fundamentals remain the same, only the medium changes.
"Data source: Google Trends (www.google.com/trends)."