Using the Faces View

Highviz.io uses AI to detect faces in your Instagram content and reveal how different faces - by age, gender, and emotion - impact your engagement.

👀 What Highviz Faces Can Tell You

Highviz automatically detects faces in your posts and breaks down:

  • Who’s in your posts - grouped by age range and gender

  • How they’re performing - including likes, comments, and saves

  • What expressions and styles resonate - based on facial emotion and headwear detection

🔍 What You’ll See in the Faces Section

🧠 Face Summary Table

For every detected face, you’ll see:

  • Thumbnail & ID: Each face gets a visual and anonymized ID

  • Sex & Age Range: Based on visual estimation (e.g., Female 20–29)

  • Engagement Rate: Calculated as (Likes + Comments + Saves) ÷ Impressions

  • Posts Detected In: How many posts featured that face

  • Performance Ranking: Easily spot which faces drive the most engagement

✅ Why this matters: Certain types of faces — like smiling, younger, or specific demographics - may consistently perform better. You can use this to plan future shoots and collaborations.

😊 Expression & Accessory Insights

We also detect:

  • Happiness Likelihood: How likely the face is expressing a smile or joyful emotion

  • Headwear Likelihood: Whether the person is wearing a hat, scarf, or head covering

This can help answer:

  • Do happy faces get saved more?

  • Does headwear improve post performance?

📈 Engagement Calculation (Important!)

Highviz uses a more reliable formula for engagement:

Engagement Rate = (Likes + Comments + Saves) ÷ Impressions

This shows how compelling your post was once people saw it, not just how many people it reached.

💡 Tip: Reach is the number of unique people who saw the post, while impressions count total views. A single person scrolling past twice counts as 2 impressions.

🔎 Face Details View

Click on any face row to open the Face Details Page, where you’ll see:

  • The posts featuring that face (with thumbnails and captions)

  • Breakdown of performance:

    • Impressions, Reach

    • Likes, Comments, Saves

    • Engagement Rate for each post

  • Happiness & headwear likelihood for that specific post

Example: You might find that a 20–29-year-old female face with a high happiness score led to a higher engagement rate than prevopis posts - that’s a strong signal to use that type of imagery more often.

✅ Final Takeaways

  • Use Highviz’s face analysis to understand who your audience engages with

  • Prioritize faces that drive higher engagement per impression

  • Mix and test different demographics, expressions, and visual contexts

  • Combine this with color and object data for deeper insights

Was this helpful?