Covid-19: Data Analysis in Times of Crisis

The crisis in which we are currently living generates a lot of stress within our organizations. In addition to this, user behaviour has drastically changed in the past few weeks, forcing us to be more agile to quickly identify their new needs and to take advantage of the resulting business opportunities. Using this knowledge, I want to share some tips for those whose role is to monitor and analyze data on a daily basis. 

The current situation has divided businesses into two key groups: those for which sales (mainly online) are exploding, and those for which consumer interest in their products and services becomes lower every day. For each of these groups, here are our general observations as well as the key questions you should be asking yourselves in order to adapt your business strategy to the new behaviours of your users.

The first general change that we see concerns the distribution of traffic sources to the different digital properties of businesses. Given that users are in an emergency situation with a specific goal in mind, more will arrive on your site through search engines (organic source) or by accessing it directly through the browser. Social networks are also driving more traffic to websites as the world is spending more time on social media channels. 

The types of platforms used will also change. Mobile is very prominent right now, but as more people are working from home, they will be much more present on websites using their computers rather than their mobile devices. 

It will be very interesting to see the change in behaviour regarding internal searches made on your website. What are the keywords you see that are emerging most often? Are there trends? 

All organizations that are seeing an increase in site visits (notably those in the food or pharmaceutical industries) will see a significant change in the behaviour of their users as well as a wave of new users who will discover their site for the first time. 

Here are a few key elements to observe in user behaviour.

For e-commerce sites

• Part of your users who primarily made their purchases in-store will now make them online. Will this increase in online purchases be equal to the volume lost in terms of offline purchases? In most cases, the answer is no. There will often be a loss of users during this transition.
  • For in-store purchasers who have not yet made the move to online, be sure to contact them by email or through social media; if they do not become purchasers during this period, it will still be important to keep them active/engaged with your brand in order to keep them as customers. 
  • For new online customers acquired during this period, this is the time to focus on their loyalty so they become permanent clients rather than temporary ones. 

•  
• New users and new clients will likely have more questions during their journey, being new to your site. Therefore, they’re more likely to try to contact you or will visit the FAQ pages available on your site. 
• There will be an increase in searches for information such as delivery services or return policies. 

For non-e-commerce sites (information/customer service/lead generation)

• Informative content (FAQ, client accounts, customer service) should be consulted much more; 
• New users will surface. They are possibly already customers that were not used to coming to your site. These users will come for information related to the services they’re subscribed to or wish to subscribe to. 

In your particular situation, start making hypotheses and make a list of your questions on the new behaviours you expect from users at this time. This should always be your first step – don’t dive into an analysis without a goal in mind! 

For e-commerce sites

• Check the frequency of purchases: what is the number of purchases made by a user?
  • Start by isolating the number of users across a segment who make a purchases 
  • Associate this number with the number of transactions, and you’ll have the average number of transactions per user 
• Check your average cart value: Observe the value of the average cart and its evolution compared to the previous year. Is it increasing or decreasing? 
• Check the average number of products per transaction by dividing the number of products purchased by the total number of transactions: are your users purchasing more products? 
• Which products are users most interested in? Your top categories have likely evolved. Check which categories and products are most prominent in purchases. 

It would also be interesting to compare the behaviour of users who are already used to your site with the behaviour of users purchasing from your site for the first time: do they make more regular purchases? Is the average cart value higher? Use segments to isolate these two types of users. 

For non-e-commerce sites

• Which pages are most consulted at this time? Is the FAQ and, more specifically, questions related to customer account cancellation? Information on purchasing? New pages or sections on COVID-19? 
• Isolate multiple user types across segments and observe if there is a difference in the content that is most consulted: 
  • Known users/customers vs. new ones;
  • Regular users vs. users visiting once or twice in this period;
  • Users who consult COVID-19 content vs. those who do not;
  • Desktop vs. mobile users 

All of this information will allow you to generate insights that will help improve your e-commerce site performance and user experience. I recommend you consult my recommendations below on how you can use this new data. 

Whether you have an e-commerce site or not, user visits on your site are not related to an essential need. Among the users who still come to your site, you’ll need to ask yourself the right questions to identify new behaviours and adapt accordingly. Here are a few examples: 

• Are these users regular visitors or new ones? 
• Which content is consulted most often? 
• If purchases are made, which products? Are these products that are necessary now, or is it for future use? 
• Has the frequency of visits changed? Is it only a panic visit for a user who got the information they were looking for, or is what the user needs something they can consult your site more regularly for? 
• Is the user navigation experience seamless? Are there any roadblocks to get to relevant content or to take a required action?

Above all, I recommend adapting your measurement tool for the current crisis. 

• Create annotations within your tool to give context to specific dates, such as the beginning of COVID-19, the closure of schools, closure of non-essential businesses, etc. This will help you better understand behaviour changes through these events, and will save you time trying to identify the chronological order if you’re looking at this data months later. Context is key! 
• Create segments of your users split by regular and new visitors: from your analyses, you’ve identified a few key behaviours. Isolate these criteria into different categories in order to make a comparison and observe the changes. Do these new user behaviours end up becoming regular ones once this situation subsides? Do these new users stick around?

Once you have all the answers to your questions and you’ve identified your user needs, you can then implement actions in order to help them more easily access what is important to them. 

In this stressful time, we want to minimize user frustration and make their experience as simple and pleasant as possible. In some cases, the goal will also be to reduce volume on call centres which are experiencing a significantly higher volume of calls. 

Here’s how to action the insights that this data will help you generate: 

• Highlight your most popular content :
  • Integrate it on your homepage or other key pages; 
  • Amplify it through an email communication 
• Make it easy for customers to communicate with you: make your phone number, chat service, forum, or FAQ clear and simple to find. 
• Create content that is missing on your site. By looking at searches made on your site, you can identify opportunities for content that is searched for but that doesn’t exist on your site. 
• Make navigation or action easy on your site, if you’ve noticed an issue in your analysis 
• Edit the text on your site to have a more reassuring tone. 

Armed with these new optimization ideas, I hope this helps you with your analysis! The mots important thing is to stay positive in this situation and to take the time to find opportunities, optimize your time and efficiently contribute to your business objectives. 

If you’re looking for support, our dedicated Analytics team is there to help you analyze your data and define the key actions to put in place to adapt your service offering in these strange new times. 

Take care of yourselves, your families, and your measurement tools!