Skip to Main Content

The Nonprofit Email Report: Data-Backed Insights for Better Engagement

Browse The Report

  1. Introduction
    1. A Note from Neon One
    2. Who’s Represented In This Report?
    3. Why Is Email So Important?
    4. Anatomy of an Email
  2. Nonprofit Email Deliverability & Engagement Benchmarks
    1. Introduction
    2. Nonprofit List Sizes
    3. Ask The Expert: Is List Segmentation Really That Important?
    4. Nonprofit Email Bounce Rates
    5. Nonprofit Email Open Rates
    6. Nonprofit Email Unsubscribe Rates
    7. Nonprofit Email Click-Through Rates
    8. Nonprofit Email Fundraising Performance
    9. Nonprofit Email Performance by Date & Time
    10. Email Sender Superlatives
  3. A Data-Driven Approach to Subject Lines & Preview Text
    1. Introduction
    2. Convey Positive Emotions in Subject Lines
    3. Ask the Expert: How Did You Use AI for Subject Line Sentiment Analysis?
    4. Words to Include (or Avoid) in Your Subject Lines
    5. Experiment with Emojis in Subject Lines
    6. Write Compelling Preview Text
    7. Words to Include (or Avoid) in Your Preview Text
    8. Put It All Together — Performance Benchmarks & Word Usage
  4. Creating Effective Emails
    1. Introduction
    2. Ask the Expert: What Should I Keep In Mind When Creating Compelling Emails?
    3. Tip #1 — Include Imagery in Your Emails
    4. Tip #2 — Pay Attention to Salutations
    5. Ask The Expert: Do Salutations Really Make a Difference?
    6. Tip #3 — Use the Word "You"
    7. Tip #4 — Make Your Message Scannable
    8. Ask The Expert: How Do I Create a Great Call to Action?
    9. Tip #5 — Include Great Calls to Action
    10. Put It All Together - Build Clear, Compelling Emails
  5. Lessons from the Most Engaging Email of 2022
    1. Introduction
    2. Top Engagement Email Dissection
    3. Go Build More Engaging Emails
  6. Data-Backed Insights for GivingTuesday and Year-End
    1. Introduction
    2. Who’s Included In This Data?
    3. GivingTuesday Fundraising Totals
    4. GivingTuesday Email Data
    5. End of Year Fundraising Totals
    6. End-of-Year Email Data
    7. Use These Data-Backed Best Practices to Nail Your GivingTuesday and Year-End Goals
    8. Now Get Out There And Write Some Emails
  7. Methodology & Appendix
    1. Methodology
    2. The Dataset
    3. Terminology and Definitions
    4. Metrics Definitions
    5. Analysis by Mission & Organization Income
    6. Analysis by Time & Date Sent
    7. Analysis by Content Sentiment
    8. Data Privacy & Security
  8. About Neon One
    1. Learn more about Neon One

Analysis by Content Sentiment

For the purpose of this report, a dataset of subject lines from nonprofit emails was collected and analyzed using an AI-powered sentiment analysis tool. The task of sentiment analysis is crucial for nonprofits looking to automate the assessment of subject lines, social media text, and messages as positive, negative, or neutral.

However, this traditional method of sentiment analysis is limited, as it only categorizes emotions into three types. A more advanced approach would be to identify specific emotions, such as anger or disapproval in negative text and caring or admiration in positive text.

Previously, the process of understanding the emotions behind the text was time-consuming and relied on human empathy. But, with the advent of large language models, nonprofits can now quickly and cost-effectively classify emotions in the text through automation.

For the purposes of this report, we classified each subject line using OpenAI’s Davinci-003 model as one of the following 27 emotions: Admiration, Approval, Annoyance, Gratitude, Disapproval, Amusement, Curiosity, Love, Optimism, Disappointment, Joy, Realization, Anger, Sadness, Confusion, Caring, Excitement, Surprise, Disgust, Desire, Fear, Remorse, Embarrassment, Nervousness, Pride, Relief, Grief, or Neutral (not counted as an emotion)‍.

GPT-J provides a better opportunity for emotional analysis and an updated finding will be shared when these tools can more easily and quickly ingest the volume of data provided. Davinci-003 did create some noise in the classification that slightly impacted the findings. For example, if the completion output was “The sentiment is love” instead of “love,” that may have slightly impacted the data interpretation. It is possible our natural language query of the database excluded this result. However, this was a very small percentage of the overall output.