Last week, Google’s marketing department published an article detailing their experiences and findings during the current Covid-19 (Corona) pandemic. In this article, Google also tells what currently works for them, and what does not. I would like to share the 4 most important learnings from this with you.
- Continue to focus on long-term KPI’s;
- Analyze cross-channel key influences;
- Pretesting of advertisements;
- Be cautious with A / B testing.
Continue to focus on long-term KPI’s
Even though your longer-term goals are still relevant, in these corona times you might be tempted to change your KPI’s to short-term KPI’s. Ofcourse it is important to focus on ROI KPI’s such as the CPL or CPS of campaigns, but other short-term KPI’s are no longer relevant and feasible in these unique times.
Analyze cross-channel key influences
Where in stable times the effect of your marketing efforts on your results is almost predictable, this does not apply now. Not only are you dealing with unpredictable behavior of your target group, other factors also influence your results. Think of news in general or news about your company or brand, other campaigns or other channels that you or your competitors are now deploying, strong fluctuations in both prices and performance of channels such as organic and paid search and social, etc.
That is why it is important now to quickly find the factors that have the most influence on your marketing results. You can easily do this with a key influences (regression) analysis with tools like Power BI, R or Python. Contact me if you want to know how to do these analyzes.
Pretesting of advertisements
In uncertain times like these, people seek help and comfort, both practically and emotionally. Your target audience may not be waiting for your message, however much you would like to let them know. If you are going to approach them, it is now even more important to know how your ads are perceived. This way you avoid missing the plank, or worse, hurt your image and reputation.
Ad pretesting doesn’t have to be difficult or time consuming. You can submit your advertisement with a few specific questions to a representative sample of your customers, prospects or suspects with tools like SurveyMonkey or Hotjar.
Be cautious with A/B testing
In their article, Google goes deeper into market testing, which is now difficult due to the differences in social distancing protocols of individual states in the US and thus the unpredictability of user behavior. If you are not targeting the US you are dealing with one protocol in your country, but you are still facing unpredictable user behavior.
For the latter reason, I am now reluctant to set up and run A/B tests. Of course, if they are set up properly, they are representative, but user behavior fluctuates too much so that the results of your A/B tests for the longer term are not representative. Regardless of whether A/B testing should be a priority now…
The full article of Google Marketing can be found here.