Pepsi's disastrous advertising foray into race relations was near-universally reviled the instant it hit social media last week.

But a new poll suggests that the public at large may not have been as put off by the ad as the internet outrage might imply.

Media and survey research firm Morning Consult asked more than 2,000 Americans to rate how their opinion of Pepsi changed after watching Kendall Jenner broker peace with its product.

A surprising number of them liked what they saw. Forty-four percent said they saw the soda maker in a more positive light after watching the ad, while only a quarter saw it as a turn-off. (The remaining 31 percent didn't have an opinion.)

The findings are a world of difference from those brand research firm Amobee gleaned from social media data last week. Of the thousands of mentions of Pepsi in the days following the ad's release, the firm said 31 percent described the ad as "tone-deaf," and 10 percent included the words "worst ever."

YouGov, one of the most widely cited trackers of brand sentiment found that Pepsi tumbled a whopping 20 points in its "buzz" index in the week after the ad dropped among an unspecified number of online respondents in the UK.

Why do these two sets of people appear to be living on different planets?

One reason might have to do with who the Morning Consult asked. The researchers chose a sample pool that matched the demographic makeup of the country's population. That meant an overwhelming majority -- 80 percent -- were white, and a similar portion were age 30 or older.

Meanwhile, outside of parent-friendly Facebook, social media skews towards a much younger and more diverse crowd, according to the latest Pew Research survey.

While the black people and young people (aged 18 to 29) polled actually seemed to like the ad slightly more than the group as a whole, the firm ultimately only spoke to 246 and 409 people from each group respectively (giving it a margin of error of about 7%).

Reactions didn't seem to cut along political party lines either. Despite controversy over the ad's perceived exploitation of liberal political struggles, members of both major political parties tended to like the commercial more than their Independent counterparts. The ad made around 60 percent of Democrats and 55 percent of Republicans see the brand in at least a somewhat more favorable light as opposed to 48 percent of Independents.

Real-time opinion tracking throughout the duration of the ad reveals a bit more insight into these numbers.

Sentiment among Republican viewers plummeted to its lowest point when a woman wearing a Hijab appeared on the screen then rebounded once they glimpsed the line of riot cops. The introduction of Kendall Jenner prompted a bipartisan dip towards the negative.

The questioning was conducted during the latter half of last week -- after the news cycle around the ad had largely blown over. Researchers didn't say whether respondents had seen the ad or followed the media's coverage of it.

Perhaps these overall mild responses are indicative of what Pepsi's marketers might have seen from the focus groups they conducted before foisting the commercial on the hapless public.

Since it's hard to construe the debacle as a win for Pepsi regardless of what any data might say, the poll may be more of an indictment of the practice of focus group testing than a vindication of the soda giant's marketing strategy. Some ad agency execs are skeptical of traditional focus groups in part because of something called the Hawthorne Effect -- that is, people tend to act differently than they might otherwise when they know they're being observed.

Brand advertisers generally put more faith in that type of research than the agency creatives who make the ads. But Pepsi eschewed an agency in favor its in-house marketing arm.

A 2015 article in Digidaydetailed the array of tech tools Pepsi uses to gauge how young people feel about it and its ads. The brand told the site it only gives commercials the green light once they've passed through multiple focus groups.

Like this new poll, those guardrails evidently failed to register the ad's overwhelming cringe factor.

So if surveys can't be trusted to reliably anticipate social media sentiment, how can companies predict such backlashes?

If Pepsi's marketing division looks anything like the rest of the industry, very few people of color were involved in the decision-making process. Maybe execs can start there.

（作者：产品中心）