This is just one use case of the Search Console API. Despite the organic traffic dropping significantly, the search trends stays relatively consistent, indicating a drop in keyword rankings to the pages that drive high-volume of traffic – these rankings then pick up towards week 33. In the graph below there is clear divergence between the two metrics in week 24. The graph below shows example data with the organic traffic on the left axis and the search trends on the right axis. Does it look like you are moving up or down Google’s search results page? There may be points where the lines deviate from another, indicating a change in keyword rankings. What you will likely see is a positive correlation between them, with traffic increasing as search popularity increases. We recommend overlaying the Search Trends data and organic traffic on a graph over time using a right and left axis in order to assess the trend. The aggregation of the Search Trends data means that we can see the relative popularity across all search queries over time, providing an insight into the changes in market interest. Now these two datasets can easily be overlayed in a graph to analyse (see graph below for example). In our final line of code above, we have merged the datasets together into 3 columns: week start date, organic sessions and relative search trend. Merging the Datasets and Gaining Insights # Merge Search Trends and Organic Traffic # Time_trend$keyword % group_by(week) %>% summarise(sum(sessions)) Trends = gtrends(keywords, gprop =channel,geo=country, time = time ) Therefore, the change in popularity of one has no direct impact on that of another. We advise doing this if you are assessing the overall popularity of a collective of search terms, as this loop runs each keyword separately before aggregating each of their weekly changes. The gtrendsR package has a limit of 5 keywords at a time, so if you want to run multiple queries then you can use a loop as shown below. Time represents the start and end date of the request.Geo represents the country based on their two digit ISO code – however if left blank this will default to worldwide.Gprop represents Google’s properties, namely news, images and YouTube – however if left blank this will default to web.Keywords = c(“Soft drinks”, “Fizzy drinks”, “Pepsi”, “Lemonade”, “Fanta”)ĭata1 = gtrends(keywords, gprop = channel, geo = country, time = time)Īs shown in the above code, additional variables can be included in your request: gprop, geo and time: # Load Libraries and Set Up Query Conditions # Selecting the desired keywords can be difficult – we recommend looking at the top non-brand terms in Search Console ordered by impressions. To start, simply install the package with install.packages(“gtrendsR”) and load up the library with library(gtrendsR). This data is the easiest to obtain, mainly because there are no API keys required. Google Trends API – How to Pull Search Trends Data We will use the example of a company that sells soft drinks and is wanting to understand the popularity over time of products they sell. The example code provided should be straightforward to copy/paste and adapt to suit your needs. For Google Trends we will use the gtrendsR package and then for Google Analytics we will use the googleAnalyticsR package, which we covered in our previous article – Using the Google Analytics API with R. There are R packages available for both Google Trends and Google Analytics which are pretty easy to use for those with a basic understanding of programming. If your organic traffic is beginning to slip, is this due to a drop in market interest or because you aren’t ranking so highly for your top search queries? This blog provides such an example by showing how you can overlay Google Trends data with your website’s organic traffic – this can be used to understand SEO performance. Overlaying this data with that of other sources can provide insights that would otherwise be missed. The popularity of search queries over time can be analysed to show market awareness. Google Trends is a key feature of many digital marketer’s toolkit. These days, Google is essentially the world’s sentiment barometer and the best thing about Google’s search trend data? It’s free. Unsurprisingly, studies have been carried out to forecast rising COVID-19 cases by region. For example, search trends data can recognise disease outbreaks far before they are announced by the media by seeing increases in search queries related to specific symptoms. With billions of searches each day, Google can understand a lot more about our society and the sentiment of populations than governments. We all know that Google captures a frightening amount of data about us.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |