AI + MarketingAssociationsCase StudyConsumer PerceptionInternational Market ResearchMarket Research

UNICEF’s Social Media Optimization: The School Cluster (Part 3 of 3)

By 17. February 2020 March 4th, 2020 No Comments
Public transport association

After conducting a baseline analysis of UNICEF’s posts, demonstrated how to run a social media optimization baseline analysis on UNICEF’s Facebook posts. This led to the extrapolation of three clusters from UNICEF’s 2,917 posts. 

In this post, we will focus on the third of those clusters, the School + Education cluster.

Top 3 clusters identified by our embedding models based on top words

Top 3 clusters identified by our embedding models based on top words

To see how associations and optimization approaches vary across clusters, take a look at what we found in the Family analysis and the Refugees + Migrants analysis showed. 

In the long run, our aim is to identify and establish which practices lead to the highest social engagement and success. By knowing which associations drive successful social media posts within the School + Education cluster, SMO can be at its maximum. 

To do this, we will go over the following analyses:  

  • Text analysis of this cluster
    • Text associations that led to HIGH engagement
    • Text associations that led to LOW engagement 
  • Examples of text+associations that were SUCCESSFUL
  • Examples of text+associations that were UNSUCCESSFUL
  • Image analysis for this cluster
    • Image associations that generated HIGH engagement
    • Image associations that generated LOW engagement
  • Examples of Images
    • SUCCESSFUL images
    • UNSUCCESSFUL images
  • Image analysis: Hue, luminosity, saturation
  • Image analysis: Text on image

By considering each of these factors, we will be able to see exactly how certain text and image choices lead to low or high engagement. Ultimately, this knowledge can help you strategically craft your content and social media posts to be as effective as possible. 

Text Analysis for the School and Education Cluster

Then, the posts that were classified in the School and Education cluster were analyzed to identify which associations were invoked in the text and images.

The findings for text associations are illustrated in the image below:

For the School and Education cluster, correlations of text associations appearing in successful posts (represented in yellow bars) and unsuccessful posts (gray bars)

For the School and Education cluster, correlations of text associations appearing in successful posts (represented in yellow bars) and unsuccessful posts (gray bars)

Text Associations with High Engagement 

More specifically, the text associations that were associated with successful posts were:

  • Grateful: thank, thankful, grateful, wonderful, 🙂, awesome, happy, congrats, appreciate
  • Information source: interview, blogs, blogger, podcast, article, excerpt, blog, commentary, author
  • Family: daughter, grandmother, niece, son, aunt, father, granddaughter, cousin, nephew, grandfather
  • Cheerful: wonderful, joyful, wonderfully, fabulous, exuberant, lovely, delight, amaze, cheerful
  • Heartfelt: cherish, beloved, love, yearn, heartfelt, joy, heartbreaking, forever, sorrow, sadness

Text Associations with Low Engagement 

While less successful text for the School and Education cluster was associated with:

  • Consequence: approach, perceive, implication, situation, potential, consequently, outcome, aspect
  • Temperature: cold, warm, cool, spin-dry, soak, melt, water, temperature, wet
  • Critical: importantly, imperative, effectively, crucial, necessary, ability, vital, important, additionally, critical
  • Danger: harm, potentially, dangerous, prevent, vulnerable, danger, susceptible, expose, harmful, cause
  • Trauma: emotional, physical, mental, trauma, anxiety, depression, cognitive, physiological, emotion

Examples of Successful Posts and Associations

Below, you can see how the highly successful associations (like Grateful, Information source, and Family) are identified and measured in real-life Facebook posts categorized within the School+Education cluster:

The green highlights are used to visualize the strength of the goal association and the word. This is made possible via the in-browser NEURO FLASH chrome extension.  The extension works while you are logged in to your account. For your free 3-day trial, sign-up here.

Now, let’s take a look at a few examples of what successful posts within the School + Education cluster look like:

Grateful Association

The following three posts were significantly correlated with the grateful association:

Grateful association

Grateful association

Grateful association

Grateful association

Grateful association

Grateful association

Information Source Association

The following three posts were significantly correlated with the information source association:

Information source association

Information source association

Information source association

Information source association

Information source association

Information source association

Family Association

The following three posts were significantly correlated with the family association:

Family association

Family association

Family association

Family association

Family association

Family association

Examples of Unsuccessful Posts and Associations

Below are examples of texts and associations that showed low engagement scores. They represent the following associations, in order: 

Critical association

Critical association

Danger association

Danger association

Trauma association

Trauma association

Image Analysis for the School + Education Cluster

When it came to the images, the following images were positively and negatively associated with the School and Education cluster:

 For the School and Education cluster, correlations of image associations appearing in successful posts (represented in yellow bars) and unsuccessful posts (gray bars)

For the School and Education cluster, correlations of image associations appearing in successful posts (represented in yellow bars) and unsuccessful posts (gray bars)

Image Associations with High Engagement 

For successful posts in the School and Education cluster, the images were associated with:

  • Renewable energy: energy, renewable_energy, renewable, solar, generator, fuel, biomass, biogas, power
  • Public transport: passenger, car, lorry, taxi, commuter, roadway, pedestrian, transport, tram
  • Colorful: yellow, purple, color, blue, dark, bright, shade, red, pale
  • Delicious: yummy, delicious, sweet, deliciously, concoction, savory, flavor, deliciousness, tart
  • Nutritious: nutritious, meat, veggies, eat, food, poultry, meal, foodstuff, veggie

Image Associations with Low Engagement 

For unsuccessful posts in the School and Education cluster, the images were associated with:

  • Internet: internet, app, mobile, user, device, network, smartphone, access, software
  • Workshop: internet, app, mobile, user, device, network, smartphone, access, software
  • Sport: olympic, soccer, gymnastic, olympian, football, olympics, sport, championship, volleyball
  • Authorization: copy, correspondence, receipt, text, acknowledgment, note, letter, correspond, authorization
  • Age: 3-year, 7-year, 4-year, 5-year, 8-year, 6-year, 2-month, 9-year, 9-month

Examples of Images 

Successful Images and Associations

Below are examples of images that were associated with high engagement and their respective associations: 

Unsuccessful Image Posts and Associations

Below are images, and their respective associations, that showed to have the lowest levels of engagement:

Image Analysis for School + Education Cluster: Image characteristics

Also, we assessed whether the following photographical elements were associated with post success:

 

  • Hue

 

  • Luminosity
  • Saturation

 

 

For the School and Education cluster, images that were darker performed significantly better than images that were bright.

For the School and Education cluster, images that were darker performed significantly better than images that were bright.

Luminosity was a significant predictor of post success with darker images performing significantly better than brighter ones.

Image Analysis for School + Education Cluster: Text on image

We also measured whether a relationship between:

  • The number of words in the pictures and success
  • The area of text used in an image and success

And found that the more words that an image had, the more successful the post was: 

 For the School and Education cluster, the more words that there are in an image, the more successful that the post is.

For the School and Education cluster, the more words that there are in an image, the more successful that the post is.

And, the larger the area that was occupied by the text, the stronger the level of engagement:

For the School and Education cluster, the larger the area that the text occupies on the image, the more successful it is.

For the School and Education cluster, the larger the area that the text occupies on the image, the more successful it is.

Conclusion

Using big data insights and machine learning tools, we were able to pinpoint exactly what led to high success for UNICEF’s posts. By isolating the associations and stylistic factors that are associated with success, UNICEF’s SMO  strategy will automatically improve.

Now, we know what factors contribute to the success of a social media post. For the School+Education cluster, specific guidelines can be established, defining which associations for texts and images are most effective.

Therefore, piece by piece, a detailed social media optimization strategy is forming which outlines which factors to use and avoid. 

For more info, check out these related blog posts: 

Top 3 clusters identified by our embedding models based on top words

Top 3 clusters identified by our embedding models based on top words

Jonathan T. Mall

Author Jonathan T. Mall

Jonathan is a cognitive Neuropsychologist turned entrepreneur, obsessed with understanding what and why people think. He founded NEURO FLASH to help communicators find the best words, phrases, and images to connect with their target audience most effectively.

More posts by Jonathan T. Mall

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.