Inbound Marketing Techniques

3 Flash Poll Examples: Amazon, Facebook, and Google


In this post, which continues our recent theme of Flash Polls, what they are and what insights they can deliver, we examine how Amazon, Facebook, and Google utilize Flash Polls.

Amazon Asks

amazon_asks_example_1

This curious question is displayed after the user scrolls to the bottom of the home page in the Amazon Prime app, in a section they call “Amazon Asks”. The Flash Poll is inserted within the flow of the page, which minimizes the interruption to the user experience.

The results of the Flash Poll are revealed after the user answers. We expect users will answer the poll because they are curious to compare themselves to other respondents. We also like that the Flash Poll appears at the bottom of the page. There’s no other place to go, so why not participate?

Amazon can use information gleaned from these polls to either advertise more specifically to each user or to trace the profile and shopping pattern of certain groups. But the questions are so random that it’s hard to know if the data is truly actionable. It seems that most questions, like the one above and the example we cited in our last post are frivolous but that’s not always true. In one recent example, Amazon Asked a more pertinent question to their business model:

When choosing a product, which of these matter the most to you?
Price, Sustainability, Brand, or Customer Reviews

amazon_asks_example_2

Facebook Advertiser Sponsored Poll

facebook_flash_poll_example

Like the Amazon Asks example above, Facebook’s Flash Poll is inserted into the flow of their iPhone mobile app so that it minimizes interruption to the user experience. Anyone who doesn’t care to answer can keep scrolling down.

Facebook is forthright in disclosing this particular instance of a Flash Poll is paid for by a sponsor. Of course, they won’t say who sponsored it. It is fun to speculate who placed this ad (our guess: Microsoft) but we don’t know for sure. In any case, this particular Flash Poll asks:

Which, if any, of the following voice-activated home assistant speakers have you heard of? (Select all that apply)
Google Home, Amazon Echo, Siri, Cortana, None of the above

It’s a basic market research question designed to gauge brand awareness. Facebook actually provides a service called Brand Polling Solutions to its advertising partners so they can measure the effectiveness of their ad campaigns. “Delivering real business results is important to every campaign. Our brand polling solutions help advertisers see how their ads resonate with their audience. That way, we can measure how effective a brand campaign really is.”

We like the application of taking a proven market research method and applying it to web advertising but we wish there was some type of incentive or explanation that would draw in more respondents. A direct financial incentive might skew results but it’s hard to fathom why anyone would participate in this poll when it’s so easy to scroll down instead. Is there a way to design such polls to increase opt-in from users?

Google – How Satisfied Are You?

google_flash_poll_example

Occasionally, Google will ask it’s users to rate its search results on a scale from “Very Satisfied” to “Very Dissatisfied” as they do in this example. This Flash Poll appears in the lower right hand corner of the page. The placement provides the best of both worlds because there’s no chance it could antagonize the user and yet it is very hard to ignore. It does not interrupt or obscure any content. At the same time, it grabs your attention.

Unfortunately, upon answering this question, the dialog box disappears. There is no opportunity for the user to explain what might improve the results or why they gave such a rating. Since Google went through the trouble of asking the question in the first place, we’re surprised they don’t try to probe more deeply.

We think this example could be improved by making a more personal appeal to the user so that Google can understand why a particular result resonated with the user and/or how the search results can be improved. It is not hard to imagine a day in the not-too-distant future when Flash Polls collect feedback from users and this feedback improves a machine’s Artificial Intelligence.