Covid-19 Keyword Trends: How Is Coronavirus Impacting Different Industries?

Covid-19 has been the main topic of conversation in marketing circles for nearly two months now. Are we sick of it yet? Maybe. Is it the reality we're stuck with for the foreseeable future? Definitely.

The effects of Covid-19 on every industry can't be overstated, and those effects are also changing every day as new information surfaces and as we all adjust to quarantine as the new normal. In response, many of us have been looking for data to check our assumptions.

We wanted to help with that.

A study of click trends by industry

At Botify, in addition to crawl and log file data, we have the largest collection of real keywords and click data for some of the biggest sites on the web across a variety of industries.

As many of us wait for our respective industries to recover from the effects of Covid-19, our team has begun work to create a resource that helps paint a picture of how this pandemic has impacted and is continuing to impact consumer search behavior.

View the study here

To do that, we commissioned our friend Ben Goodsell of Tight Ship Consulting. The result is a granular breakdown of click trends in specific *industries since the first media reporting of Covid-19 came out of China.

*While viewing aggregate click data by industry can serve as a helpful benchmark, there's no substitute for looking at your own site's data to see how your unique audience is searching. If you're a Botify customer, visit How to Leverage Insights in Botify to Adapt During the COVID-19 Outbreak to learn how.

What industries does the study cover?

So far, we've been able to organize our data into the following categories:

Travel

• Booking
• Flights / Trains
• Hotels
• Media

Retail

• Clothing
• Outdoor
• Fitness
• Cosmetics
• Luxury
• Essential Home
• Non-Essential Home

Real Estate

• Buy/Sell
• Rent
• Utilities & Construction

Services Industries

• Education
• Health
• Job/Career
• Legal
• Pet Care

Health

• Media
• Retail
• Services

Entertainment

• Internet
• Media Streaming

Food

• Restaurants
• Cooking & Recipes

Weddings

• Media
• Retail

Investments

• Media
• Precious Metals

We plan on adding new categories whenever they'd be possible and helpful.

Why correlate with global cumulative confirmed cases?

Overlaying every click trends chart is a trendline that shows global cumulative confirmed cases as reported by Johns Hopkins.  

But why confirmed cases?

Confirmed cases may not tell us how many people actually have Coronavirus, but it is a good proxy for the prevalence and severity of Coronavirus stories in the media. And as we know, the news cycle can have a drastic impact on consumer behavior.

In some cases, there is an obvious correlation between confirmed cases and an increase or decrease in clicks for certain industries, like with restaurant delivery and recipes.

In other cases, there's not.

We think this information is interesting either way, which is why all the charts show this data.

Why show two versions of every chart?

We created two versions of every chart: one that shows year-to-date (YTD) click trends and another "zoomed in" version that shows data from March onward.

The YTD chart is a great way to see how soon after the first publicized case of Covid-19 we began to see a change in consumer search behavior.

The zoomed-in chart is a great way to focus on the data around the time that Covid-19 began to have the largest impact in the U.S. and other English-speaking countries.

You'll also notice another difference with the zoomed-in charts. Instead of showing data from all the categories, we opted to keep only those categories that revealed the most interesting data, as well as added data on specific keywords in those categories.

For example, here's the YTD chart for health categories:

We wanted to call special attention to what's going on in the health services industry, so in the zoomed-in version of the chart, we're only showing the trendline for the health services industry for March onward, in addition to particularly interesting keywords that are contributing to much of the fluctuation in that category.

Why click data?

Most trend research in the SEO industry uses estimated search volume, but we opted to use click data.

Why is that?

In the SEO industry, we're constantly looking at the search volume of keywords to understand trends (i.e. how many times a particular keyword is searched). Or, we look at Google Trends data (i.e. a term's search interest relative to its highest point of interest for a given date range). In both cases, that only tells part of the story.

We wanted to take it a step further and understand what additional insight can be gained by looking at real click data.

How is real click data different from search volume data?

While search volume tells us how many times a keyword was searched and Google Trends data tells us the relative popularity of a search query...

Adwords keyword tool

Google Trends

...click data tells us how many times a site or group of pages was clicked on by someone who searched that particular keyword.

What's the major difference here? Intent.

For any given query, Google offers a variety of results on the search engine results page (SERP). That's because what you're looking for when you search "outdoor shoes" and what I'm looking for when I search "outdoor shoes" could be two very different things -- while you might want to purchase shoes, I may be researching outdoor shoes. Because Google doesn't know, based on the query alone, what the searcher wants, Google shows both informational and transactional results on the SERP.  

Because Google SERPs serve multiple intents, click data is valuable for understanding where consumer interests really lie. It helps us understand not only how many people searched that keyword, but also what they were really looking for, something search volume can't quite capture.

When it comes to understanding the intent of certain keywords, in many cases, it's down to our best guess. With click data though, we can extract informational interest from what might normally be categorized as a transactional query, or transactional interest from what might normally be categorized as informational.

What to expect from future iterations of this study

We believe this data is important and has the potential to be a resource for many businesses that are trying to navigate a very difficult time. For that reason, we've adopted the mantra "done is better than perfect" for this project, and are committed to making improvements over time.

What improvements exactly?

• YoY Comparisons: Year-over-year comparisons are important for understanding seasonality. We have data for 2019 and are working on incorporating it into our charts so that it's clear what's normal seasonality vs. Covid-related volatility.
• Breakdown by Location: We'll soon be able to see how these click trends differ by country.
• Breakdown by Device: We also have desktop vs. mobile data that will help us understand intent even better.
• More & Better Categories: This study looks at industry categories with more granularity than we ever have before, but we're continuing to look into additional categorizations, as well as improvements to existing categories.
• Annotations: With Covid-19, so much changes on a day-to-day basis. We think it'd be helpful to add markers to our charts to indicate when major events happened so that it's easier to see the correlation between those events and changing searcher behavior.
• Improved Visualizations: Data is only as impactful as our ability to understand it, and a big part of what makes data understandable is how it's visualized. We're working on improvements, from colors to formatting, to make our charts easier to read and gain insights from.