Guidelines For Conducting Research During COVID-19
April 21st, 2020 | Joe Spencer, VP Innovations & Research Solutions, Hotspex
The leadership team at Hotspex has been carefully monitoring the COVID-19 situation and after thoughtful consideration of the context, research respondents and client needs, has prepared the following guidance on conducting market research during the crisis. These are guidelines only and are not necessarily absolute truths.
Our field partners are informing us that survey demand (clients fielding projects) and supply (respondents participating) are holding steady, with no loss of response quality. Our own experiences fielding online research remains unaffected – including fielding in China at the peak of the crisis, which likely bodes well for fielding in the coming weeks elsewhere in the world.
- Focus on A/B testing. Be cautious testing where the decision criteria rest on historical norms. If possible, garner that insight by using a control leg that recreates the existing market as your baseline. Including a control leg is the most prudent strategy to protect validity. Control vs. Test is much more reliable, especially in times of uncertainty.
- Categories of concern should proceed with caution. These currently include travel/hospitality, cleaning and disinfectant, and household paper such as facial tissue, paper towels, or toilet paper. Research to focus on the current COVID-19 event is encouraged. Research to shape brand strategy over the coming years for these categories, be cautious.
- Avoid research methodologies and approaches that require participants to be in a place that is either not safe or not recommended.
- Avoid designs requiring in-person interaction (e.g. no CLT, shop-along, or focus groups) when social distancing measures are in place.
- For internet responses, be aware of any class of respondents that would be more likely to require shared resources to access the survey instrument. Examples include taking the survey from a library or internet café. In geographies where home internet penetration is not universal, consider mobile.
- Consider the social acceptability of the topic. Now is not the time to test a concept for a new cold and flu product, finger foods, or public tournaments, for example.
- Set the context appropriately. Use scenarios to allow respondents to get into the correct mindset. This allows a “safe haven” for respondents to reflect in an environment of normalcy – and it is relatively easy to do.
- Lean on behavioural and true implicit, non-conscious methodologies for confirmation and insights. Conventional explicit question-and-answer methodologies are more likely to be influenced by the current state of the world, so it’s more important than ever to get a richer, unbiased view of consumers – which the implicit measurements uniquely bring.
- Consider increasing the volume and frequency of sampling for trackers. Results during this time period are real, and likely meaningful for your brand. With the unpredictable and fast-moving nature of public sentiment, it is vital that you have sufficient sample size to see shifts in response to unexpected developments. Don’t get caught unable to cut your data by “before vs. after COVID-19”.
- Do not simply dismiss unexpected results. Again, insights garnered during this time represent the current state of the world. If you have followed all of the above rules, but get unexpected or undesirable results, dig in. If you can conclude that it is an artifact of the current environment, then explore whether the results are likely to persist once normalcy has returned. Is this a short-term or a long-term change?
- Celebrate productive skepticism. It is incredibly important to have an assigned “devil’s advocate” to try to come up with alternate explanations for all of the insights used to make decisions. It takes extra time, but it helps ensure that you aren’t overlooking something that could impact the generalizability of your results.
This article was originally published on the Hotspex blog.
Joe Spencer, VP Innovations & Research Solutions, Hotspex
Joe joined Hotspex in February 2011 and has progressed through multiple departments to become the Vice President of Innovation and Research Solutions. His focus is to continually innovate and add new capabilities to the Hotspex toolkit. Joe’s career includes forays into information technology, statistical modeling, academia, and consulting.