• Sensei-Banner-Customer-Acquisition6
  • Sensei-Banner-Customer-Development5
  • Sensei-Banner-Customer-Relationship5
  • Sensei-Banner-Influence-Marketing5
  • Sensei-Banner-Influence-Marketing-Book4 (2)

Gallup Poll is Correct: Social Media DOES NOT Influence Purchases



Are you a marketer? Are you responsible for social media marketing? If so, you may wish to refresh your CV and LinkedIn profile. You’ll be out of a job soon. A recent study by surveying giant Gallup reports that social media marketing has no effect on the purchase decisions of US consumers despite the estimated $5.1 billion spent on social media advertising by brands in 2013.  The report has made headlines this week on The Wall Street Journal, Time, and Business Week .

Gallup reports that 62% of its respondents said that social media sites such as Facebook and Twitter have no influence on their decisions to purchase products.  Only 5% reported that social media has had “a great deal of influence” on their purchasing decisions, while another 30% said these channels have “some influence.”

Social Media Fights Back

These results are in direct conflict with almost every other study done on the influence of digital and social marketing and advocacy on brand awareness and customer behaviors, such as a Wanderful Media study (of just over 1,000 US adults) that lists a whopping 91 percent of respondents reporting that they “have gone into a store as a result of an online experience.”

AdWeek rebutted Gallup’s findings by pointing to the fact the survey was conducted a year ago and during the last year, so much has changed that the data is no longer relevant. Plus, the survey was only sent to English-speaking Americans and because so many Americans speak Spanish, the data – once again – is flawed. “And they are the most likely to socialize,” adds Marla Skiko, EVP and Director of Digital Innovation at Starcom MediaVest. “Leaving them out raises questions because they are over-consumers when it comes to social media.” (What a crock of shite.)

Facebook (naturally) jumped on the bandwagon in criticizing the report by offering:  “The only thing this poll shows is that self-reported behavioral data is unreliable. For decades, studies that look at people’s actual, real-world behavior have shown that ads on all mediums, including social media, [affect] the things people buy.”   While this may be true, the argument is akin to the childhood threat of “My dad can beat up your dad.”

Others have made a valid argument when referencing review sites such as Yelp: Why are businesses so afraid of negative reviews? Have you ever made a decision to not visit a restaurant or buy a product (or buy it) based on reviews?  The survey did not define “social media” to include these channels and this would certainly skew the results.

Gallup Poll is correct: Social Media DOES NOT Influence Purchases

Now, as the co-author of Influence Marketing: How to Create, Manage, and Manage Brand Influencers in Social Media Marketing, you’d think I would not be supporting Gallup’s findings. And you’d be right. I’m not defending the results;, they’re clearly misleading. However, my disagreement with the results is based on a more important critique.

Did Gallup report the numbers incorrectly? No. We trust the source and don’t doubt that the summary of answers it listed is accurate. The problem is the same problem that has plagued influence marketing in the digital space since its inception.  The issue is that the authors of the survey – like many brands and marketers – don’t understand the nature of online influence and advocacy and, as a result, set up an inaccurate premise from which to conclude any relevant findings.

The question asked of respondents was: “How much does social media typically influence your purchase decision?” I agree, “social media” does not influence purchase decisions; people influence purchase decisions.

Gallup Social Media Influence 1

If this same audience was asked if their family, friends, and/or colleagues influenced purchase decisions the responses would have been almost unanimously “yes.”  The point is, the question was asked incorrectly.  Gallup researchers, like many social media platforms and marketers, are still focused on the technology and their own self-importance. Technology may be a conduit for influence, but actual influence is yielded across personal relationships.

People Influence People, Social Media Does Not Influence People

Gallup’s own survey asked Americans why they use social media and the overwhelming response was: “To connect with family and friends.” We’ve all seen the studies that show people trust people and not brands or their advertising. So if people influence people, and 94% of Americans use social media to connect with family and friends, does it not stand to reason that social media has provided the channel from which friends can influence friends?

Gallup Social Media Influence 2

If this audience was asked: “How much does social media typically affect your awareness of businesses, news or politics, the results would be significantly higher.  If the question was “Does brand advertising seen in social media channels such as Facebook, Twitter, etc. influence your purchase decisions?” the results would be low.

If the audience was asked, “Do the opinions of your family and friends influence your purchase decisions?” the answer would have been overwhelmingly high.

The Real Lesson Learned About Influence Marketing in Social Media

The point here is that marketers have yet to fully understand what drives influence in social media. It’s not the technology. It’s not the platform. It’s not the advertisement.

Influence is a verb, not a noun.

When marketers and businesses start focusing on how customers make purchase decisions and the factors that influence those decisions – instead of which channel provides the largest audience and which people demand the greatest attention from that audience – influence marketing campaigns and such studies will continue to be flawed.

Sensei Debates

Does social media influence purchase decisions? Join the discussion!

Feed Your Community, Not Your Ego
Sam Fiorella

Influence Marketing Banner 1


Join the Conversation

Raz Chorev
Raz Chorev 5pts

Sam, this survey / study you're referring to, just goes to prove what most of us know anyway: you can prove anything statistically - you just need to ask the right audience the right questions. 

Since most people, including marketers, don't understand social media, and its' influence, you'll get many of these rubbish studies. The mere notion of using standard advertising in social channels is wrong, and I'm not surprised it's not working for most businesses. 

tessbabee 5pts

I bought my nokia and bb

because of the info I got from social media

Also, because my family agreed.

samfiorella moderator 5pts

@tessbabee :) Exactly. That's the perfect example of what @dannybrown and I outlined in Influence Marketing. Social media (as a channel) and those who have large followings in it, can drive awareness of a product but it's the closer personal relationships we have in and out of these channels that actually affect - more often than not - the final purchase decision. Thanks for sharing. 

b_WEST 5pts

I actually agree, in part, with the comments from Facebook regarding the accuracy of self reported data. If you were to ask these same people who said that social media didn't influence their purchase decision, what did influence that decision, many would say a TV ad, even though there were no ads for the particular product they purchased. Happens all the time. 

samfiorella moderator 5pts

@b_WEST Sure...advertising affects a person's awareness of a brand or recall of the brand when at point of purchase. To say that the impression of a brand, logo, or ad has no effect - however intangible or difficult to measure - is uneducated.

The point of the study was to determine if purchase decisions were influenced by social media channels but the entire premise (questions, framework, number of questions, etc.) was flawed.  

dbvickery 5pts

Great points throughout, Sam. I know I often draw the ire of people who put a lot of stock in polls and surveys, but I wrote a post over a year ago saying "people are talking about you - and it's not in your survey".

In addition to perhaps not phrasing the questions correctly, I would also argue that the people inclined to even agree to participate in the poll are in a dwindling group that truly would never use social or review sites to make a purchase decision. Meanwhile, my purchase decision is based upon:

- identify need/want

- research (generally starting with Google)

- read professional/influencer reviews

- read peer-to-peer reviews (Yelp, TripAdvisor, Best Buy Comments, Amazon comments)

- poll my own social circles (via Facebook and Twitter, usually)

- Possibly see if the brand, or distributor like a Best Buy, will engage socially

Do the poll of folks across all ethnicities and household income demographic, for ages between 18-45, and your numbers would change without even having to change the questions.

samfiorella moderator 5pts

@dbvickery What strikes me as well is that whatever project manager or researcher that was assigned this job must have seen the "only 5 percent report that social media influences purchases" and said what?

"Yeah, that's right...let's publish this!"  

Seriously? Forget empirical evidence, there's enough anecdotal evidence around that should have made this person stop and say "hold on, we need to rethink how we asked the questions" or something before posting the results. 

BilalJaffery 5pts

Smart smart post. Most marketers are still struggling with the transition to digital.

danielghebert 5pts

You've pointed out the problem with a lot of self-reported studies: The framing of questions can greatly skew results.

In this case, they asked if "social media" influenced purchase decisions. Right away, people would respond no. Most people won't admit that any type of advertising or marketing affect what they buy, no matter how subtle the influence might be.

Framing questions correctly in any type of studies is key to collecting the right data. Gallup didn't do it right.

samfiorella moderator 5pts

@danielghebert Agreed but, for me, it's more than just a poor question. I'm giving Gallup the benefit of the doubt (as a leading surveying firm) and assuming they know how to ask questions for such studies. I'd say it's a lack of understanding or insights on the subject matter that made them not realize the question was being asked incorrectly.

@dannybrown and I went into great detail on this subject in our book, and (frankly) it's not rocket-science. Purchase decisions are - and have always been - influenced by the people we're in relationships with. 

AmyMccTobin 5pts

EXCELLENT post Sam. This reminds me of the 'guns don't kill people' argument - social media is not the 'imapacter,' the PEOPLE on social media are.

samfiorella moderator 5pts

@AmyMccTobin Heh, didn't see the "guns kill people" comparison coming.  Thanks for sharing. 


  1. […] Click here to read Sam Fiorella’s excellent analysis and rebuttal of Gallup’s findings in full. Share ! […]

Show Buttons
Share On Facebook
Share On Twitter
Share On Google Plus
Share On Linkdin
Share On Stumbleupon
Hide Buttons