How valuable are social tags as research tools?

Source: Instagram/@vip_manfashions
Feb 28, 2018
Tom Ryan

Providing a revealing look at consumers’ perceptions of products and brands, social tags have the potential to take the place of much of traditional market research, according to researchers at the University of Maryland.

Examples include images and photos tagged and shared on Pinterest, Instagram and Facebook, videos tagged on YouTube, and tweets tagged (via hashtags) on Twitter.

In their research, professors P.K. Kannan and Yogesh Joshi from the university’s Robert H. Smith School of Business compare conventional techniques brand managers use to determine consumer perceptions, such as brand concept maps and text mining.

More than 7,000 social tags were then aggregated to show the added value in the more unconstrained, open-ended information they contain.

Previously, for instance, marketers would build so-called brand maps in market research. Consumers would be asked in interactive sessions, “What comes to your mind when you think of this brand?”

Their responses — “good quality,” “cheap,” “trendy,” etc. — would be collected to create a narrative that describes consumers’ overriding sentiments about a brand, with the most-repeated terms dominating the narrative.

“This can be a very intensive process,” Prof. Kannan said in a statement. “We argue in this article that these social tags are basically a snapshot of what consumers are thinking about brands. They are giving marketing researchers a lot of information, for free.”

Marketers can use social tags as a complement to automatic keyword identification in text mining to compile a start list of keywords to mine.

The information from social tags was also found likely to be a more accurate depiction of consumers’ feelings because they’re not prompted by researchers, as in conventional approaches. Social tags can also show how brand associations vary across segments, how topics associated with a brand change over time, and how brand perceptions change over time.

The study states, “While similar information can be obtained using other methods such as text mining and primary research methods, it is the ubiquitous nature and type of tagging data that leads to its several advantages over the other methods.” 

DISCUSSION QUESTIONS: What advantages do you see, if any, with social tags in gauging brand perceptions? Will findings from social tag-based approaches amplify or replace traditional approaches such as sorting, personal interviews and surveys?

Please practice The RetailWire Golden Rule when submitting your comments.
"Any source of unprompted perceptions is a good way for marketers to obtain cues to real feelings that shoppers/users have."

Join the Discussion!

11 Comments on "How valuable are social tags as research tools?"

Sort by:   newest | oldest | most voted
Chris Petersen, PhD.

Pinterest and Instagram are the new go-to search engines. One of the most powerful things a brand can do is to get appropriate images on visual search engines so that customers not only discover products, but also see how customers are using products in their daily lives. Even with the classic search engines like Google, images and videos play a huge role in search and discover. However, images have no SEO value without being properly tagged.

Brands have always been concerned about how they are being labeled by customers. Sentiment descriptors are an outcome of branding not a means. The most important branding tags today are those that drive SEO.

Charles Dimov

I agree with this perspective. It is easy to get lured into believing that residing on the social platforms is enough. Yes they are great and will definitely drive traffic and demand. However, always remember to bring the branding tags back to align and help drive SEO.

Kim Garretson

I think it’s important to consider the motivation of those using tags in the social landscape, especially considering Instagram. In the image example here, there are 30 tags with this post. Why would someone take the time to embed 30 tags? Sure, everyone is interested in expanding the reach of their posts, but in this case, and with most tag-heavy posts on Instagram, I think there is more than ego gratification going on. The heavy posters likely are looking for paid gigs from the brands to do sponsored content and other promotional work, or hope to make money in other ways, from expanding their blog reach for more advertising, etc. So I would put a filter on these results around true opinions of the brands, and what is behind the time spent on tagging.

David Weinand

Good point Kim — but I think the same can be said for surveys and focus groups. Seth Stevens-Davidowitz points out this bias in his book “Everybody Lies.” People’s responses are unconsciously biased based on their attempt at saying what they think wants to be heard. Social tagging can definitely add great context to what people really think (in situations where they’re not using 30 tags, as you point out).

Ken Lonyai

People lie — well that’s a bit harsh (but there is a book called “Everybody Lies” about this), so I’ll say that when subjects are interviewed about a product or experience, many tend to say what they believe the inquirer is looking for or go with the group consensus. Research is highly dependent on subject selection and the construct of the questions and process. Done well, there still is some value, but most of the flaws are eliminated when the “subjects” organically share their conscious thoughts and signal unintended messages. That’s the value of reading social tags and product reviews: it’s structured by the user/consumer and in most ways is more authentic.

That said, there are two issues:

  1. Only a percentage of people are motivated to take actions like using social media and writing reviews, so a crucial segment of users may not be represented;
  2. Data is data and once acquired, can be bent to fit any narrative intentionally/unintentionally.
Anne Howe

Social tag tracking does have potential to amplify other approaches and I believe there is a value add to using the learnings to augment the brand perception mapping. Any source of unprompted perceptions is a good way for marketers to obtain cues to real feelings that shoppers/users have.

Dr. Stephen Needel

Rely on this data at your own peril – this can’t possibly be a representative sample of your brand’s users (or your brand’s non-users). A biased sample is a biased sample, whether you are asking them a question or they are spontaneously tagging things. Garbage in – garbage out.

Meaghan Brophy

There’s value in tracking social tags, especially because they are unsolicited. And utilizing this user-generated content definitely eliminates a lot of legwork, and gives marketers a good starting point.

But like others have pointed out, social tags won’t give you a complete picture. Many customers don’t post about brands on social media. But many people who haven’t interacted with a brand firsthand are basing their perceptions of you off of social posts. So while social tags don’t form a complete snapshot, they make a very relevant one.

Ralph Jacobson

Social tagging is proving to be huge for the innovators with whom we’re working. Additionally, the digital commerce leaders in these organizations are now implementing “Intelligent Tagging” with true artificial intelligence to begin the process of linking external social tags with internal online merchandise image tagging. This could be a completely new way to drive brand value and loyalty.

Vahe Katros
Vahe Katros
1 year 6 months ago

It’s a coincidence that Palantir’s CEO, Alex Karp, appeared on CNBC today. Palantir is expert at deriving “aboutness” from information. The interview is here. #AlexKarp is a valuable tag. #interesting, not as much.

Kai Clarke

The greatest issue with social tagging is first being aware of the social tag you prefer to use. Oftentimes it requires being part of the awareness for a service or product to know the social tag, and then choosing the “correct” tag to use in your post, search or blog. Until we get more sophisticated labeling, branding, and simple chatting, the research will remain secondary in its ability to accurately drive proven results on a regular basis.

"Any source of unprompted perceptions is a good way for marketers to obtain cues to real feelings that shoppers/users have."

Take Our Instant Poll

Do you agree that social tags provide an accurate "snapshot” of how consumers perceive brands?

View Results

Loading ... Loading ...