Many senior mar­keters have cre­at­ed a sin­gle Insta­gram account for their brand assum­ing that 300 mil­lion Insta­gram­mers are a sin­gle mar­ket seg­ment with com­mon needs, inter­ests, and pri­or­i­ties. How­ev­er, researchers at Penn State’s Col­lege of Infor­ma­tion Sci­ences and Tech­nol­o­gy (IST) have dis­cov­ered that teens and adults use the mobile pho­to-shar­ing app for very dif­fer­ent pur­pos­es.

The researchers detect­ed age infor­ma­tion in Insta­gram user pro­files by using a com­bi­na­tion of tex­tu­al and facial recog­ni­tion meth­ods. In their recent study, they found that teens lever­age social media as a con­ver­sa­tion space and an out­let for self-expres­sion to a greater extent than adults. The researchers also dis­cov­ered that while teens post few­er pho­tos than adults, they are more focused on post­ing pho­tos that attract atten­tion and gain­ing fol­low­ers.

“We found that teens have much high­er lev­els of self-dis­clo­sure on the Inter­net,” says Patrick Shih, a research asso­ciate at the Col­lege of IST.

Shih, along with Kyungsik “Kei­th” Han, a doc­tor­al stu­dent at the Col­lege of IST, Dong­won Lee, an IST pro­fes­sor, and Jin Yea Jang, an IST master’s stu­dent, con­duct­ed a com­par­a­tive study of 27,000 teens and adults who use Insta­gram. They pre­sent­ed the results of their study in their paper, “Gen­er­a­tion Like: Char­ac­ter­is­tics in Insta­gram.”

“We are able to show with real data how teens behave on social media,” Shih says.

The researchers chose Insta­gram for their study because more than 90 per­cent of Insta­gram users are younger than 35. They defined teens as peo­ple between 13 and 19 and adults as those between 25 and 39. They used an appli­ca­tion pro­gram­ming inter­face (API) to extract users’ data, and data col­lec­tion was done between April and May 2014.

In their paper, the researchers pro­pose a nov­el method that lever­ages online biogra­phies and pro­file images with exist­ing APIs:

• They applied tex­tu­al pat­tern recog­ni­tion algo­rithms to parse a list of pat­terns that specif­i­cal­ly describe a user’s age (e.g. “I am a teenag­er”).
• They used an online tool called Face++, which was designed to detect ages and gen­ders of peo­ple depict­ed in pho­tos.

Comparing Adults To Teens on Instagram

The results of the study yield­ed some sur­pris­ing results, accord­ing to Shih and Han. For exam­ple, teens were found to post few­er pho­tos on Insta­gram than adults.

“This may be due to the fact that teens are finan­cial­ly and cul­tur­al­ly depen­dent on their par­ents to ven­ture out­side of their dai­ly activ­i­ties com­pared to adults,” the researchers wrote in their paper.

The researchers also found a clear dif­fer­ence between the two groups in terms of the types of top­ics they engage.

• Teens: More than half of pho­tos post­ed fell under “mood/emotion” and “follow/like” top­ics. Those top­ics aren’t nec­es­sar­i­ly tied to the con­tent of pho­tos, but rather describe one’s emo­tion­al sta­tus or the inten­tion to have more fol­low­ers.
• Adults: A high­er ratio of posts fell under more diverse top­ics, includ­ing “arts/photos/design,” “loca­tions,” “nature” and “social/people.”

While teens may not post as many pho­tos as adults on Insta­gram, Shih and Han say, they appear to be much more con­sci­en­tious about how they por­tray them­selves. Through pho­to con­tent analy­sis and cal­cu­lat­ing the num­ber of pho­tos with tags and those with “self­ie-tags,” the researchers found that teens post more self­ies than adults do. In addi­tion, teens tend to manip­u­late their pho­to con­tent to receive as many “likes” as pos­si­ble, and remove pho­tos with too few likes.

Teens also tend to be more ver­bose in their bios, the researchers report­ed, and to active­ly adver­tise for oth­ers to fol­low them. Teens tend to have more likes, tags, and com­ments and to be more expres­sive about them­selves and their pho­tos. Accord­ing to the researchers, the results sup­port the idea that teens see social media as a place for self-rep­re­sen­ta­tion.

Shih and Han say they could extend their ini­tial study to oth­er social media sites to val­i­date their method­ol­o­gy and com­pare results, since many Insta­gram users pro­vide addi­tion­al social media links in their pro­file. The researchers are also inter­est­ed in exam­in­ing whether Mil­len­ni­als’ social media habits will change when they reach a cer­tain age, or are endur­ing char­ac­ter­is­tics relat­ed to the cul­tur­al moment they grew up in.

Instagram Segmentation

In the mean­time, dig­i­tal mar­keters may want to take a sec­ond look at what some of the most pop­u­lar brands are already doing on Insta­gram.

For exam­ple, Nike has more than a dozen accounts, includ­ing:

• @nike with 16.9 mil­lion fol­low­ers.
• @nikefootball (soc­cer) with 6.2 mil­lion fol­low­ers.
• @nikerunning with 2.4 mil­lion fol­low­ers.
• @nikebasketball with 2.4 mil­lion fol­low­ers.
• @nikewomen with 2.2 mil­lion fol­low­ers.
• @nikesportswear with 2 mil­lion fol­low­ers.
• @usnikefootball with 1 mil­lion fol­low­ers.
• @nikegolf with 308,000 fol­low­ers.
• @nikebaseball with 441,000 fol­low­ers.
• @nikesnowboarding with 285,000 fol­low­ers.
• @niketraining with 143,000 fol­low­ers.
• @nikejane with 20,800 fol­low­ers.

So, mar­ket seg­men­ta­tion is a key part of Nike’s strat­e­gy on Insta­gram. Is it a key part of your Insta­gram strat­e­gy?