Speaker: John Oppenhimer, Market Researcher & Community Manager, QuestionPro
Top questions & answers from this webinar
Q: Are there any specifications for which image types can be uploaded in the survey setup e.g does it need to be jpeg or eps etc?
Answer: These are the file formats for images for Heatmap and HotSpot question types that are allowed for uploading and use: JPEG, PNG, GIF, and WEBP.
Q: Can you break out the analysis by subgroups?
Answer: You can! Because you can review the Heat Maps both from the main dashboard and by navigating within the HeatMap Analysis section, you’ll be able to view and download Heat Maps across different audiences by setting filters the way you typically would using QuestionPro’s Surveys platform - the Analytics main dashboard.
Q: With the drag and drop hotspot method how do you avoid borders of boxes overlapping causing false positives? Same for heat mapping, it seems like edges of selections overlapping will create the darkest areas, not the “center” of what the user has selected.
Answer: For Heatmap analysis, if you decide to use drag and drop, a good practice is to provide instructions in the question text to let your respondents know that they should drag and drop in areas that they ONLY like or dislike (depending on what the question is asking). That way, only what gets members attention is represented in the Heat Maps. If you want to further avoid that, then you might be better off using clicks rather than drag and drop so that the areas that get the most attention will provide you with more exact information of what got the most attention among your audience.
For HotSpot image testing, a best practice is to avoid overlap. But when it comes to visual representation, you’re better off resorting to the total number of likes/dislikes/etc., and the percentage among your survey participants - since our platform has the capability to quantify that information.
Q: Can you have only 1 choice on a Hot Spot? Can you make the borders of hot spots not visible? How do you get rid of the text boxes if you don't want them.
Answer: While you can, it may defeat the purpose of evaluating just one aspect of an image when there is a border drawn around it. In a case like that, you are better off presenting the image to your respondents and then ask them if they like/dislike/are neutral/etc., towards a specific area or item on that image. So HotSpot image testing exercises should have multiple areas/items within an image to test, so you can compare areas/items across each other to understand from a relative standpoint of what you audience or targeted market values.
Q: Is there a use-case that uses both question types in the same survey?
Answer: You certainly can set up both question types in the same survey. Though typically, heatmap and Hotspot testing aren’t used on the same images within the same survey, since Heat Mapping provides generative insights whereas Hot Spot testing is more evaluated. Though it’s common to send a follow up survey that focuses on the same image(s) but uses the other exercise that goes around.
Q: For concept testing, is it better to be using larger audiences? Or only your brand’s top promoters?
Answer: For concept testing, you’re better off using larger audiences. That way, you can understand what your bottom line values when doing the analysis. Plus you can filter your audience and compare Heatmaps between your promoters and the rest of your audience or just from total in case you notice any themes.
In the case where you have multiple images to test, you can randomly assign each respondent to just one image, and then ask a follow up question that presents all of the images and asking which one is the most appealing to each respondent. By knowing which image is the most popular among your total audience, you can then analyze what about this image was the most appealing (and likewise for things you audience did NOT like about your least popular images - if asked).
Q: Does it help to follow up image testing with qualitative questions?
Answer: Absolutely - especially when you need to collect that qualitative data (e.g. verbatims, quotes, etc.) to support your data reports.
