In a recent video analysis titled "GitHub Copilot Pricing Controversy Demystified," pricing expert Ajit from Monetizely breaks down the recent changes to GitHub Copilot's pricing structure that have sparked significant customer backlash. The video examines how GitHub is navigating the challenging balance between maintaining healthy margins on their GenAI product while preserving customer satisfaction.
GitHub Copilot has been a remarkable success story, now accounting for an impressive 40% of GitHub's $2 billion annual revenue. However, their latest pricing adjustments have created friction with some of their most loyal customers.
The History of GitHub's Generous Pricing
GitHub has historically positioned Copilot with notably generous pricing compared to competitors:
"They started offering things for, let's say, $10 per user per month where the same plan was being offered and started at $20 by OpenAI or by Anthropic. They have been historically very generous in their plans," explains Ajit.
This aggressive pricing strategy helped GitHub rapidly build market share. However, as inference costs and resource requirements for advanced AI models remain significant (despite dropping 99.7% in the past two years), GitHub has been forced to implement new restrictions.
The Controversial Changes: Premium Requests and Overage Fees
The core of the controversy centers around GitHub's implementation of "premium request" limits and the introduction of overage fees. In the Pro plan ($10/month), users now get 300 premium requests per month, while Pro Plus ($39/month) offers 1,500.
What exactly are premium requests? Ajit clarifies: "A request is any interaction where you ask Copilot to do something for you, whether it's generating code, answering a question, helping you through an extension. Each time you send a prominent chat window, you're making a request. Premium requests are where which require more advanced processing power or advanced features."
The most controversial aspects include:
- Monthly vs. Daily Limits: "The criticism is valid. You could use up the request in one day and you are offering this for one month and that is an issue," notes Ajit.
- Model-Specific Multipliers: Different models have different "costs" in terms of premium requests. GPT-4.5 has a 50x multiplier compared to other models, meaning it depletes the monthly allowance much faster.
- Identical Overage Fees Across Tiers: Users in the higher-priced Pro Plus tier pay the same overage fee rate (approximately 4 cents per premium request) as those in the lower Pro tier.
A Pricing Expert's Critique: The Three-Part Tariff Problem
Ajit analyzes GitHub's implementation of what pricing experts call a "three-part tariff" (base fee + included usage + overage fees) and identifies several flaws in their approach.
When breaking down the effective cost per premium request:
- Pro plan: $10 ÷ 300 = 3.33 cents per request
- Pro Plus plan: $39 ÷ 1,500 = 2.6 cents per request
- Overage fee: Approximately 4 cents per request
The issue becomes apparent when examining the Pro Plus tier: "The problem with Pro Plus is that they've kept the overage fee the same even for the Pro Plus users. That is part that I actually think is a little bit unfair."
Ajit contrasts GitHub's approach with Eleven Labs, which has a more traditional implementation where:
- The inferred price (base fee divided by included usage) is always cheaper than the overage fee
- Higher tiers have proportionately adjusted overage rates
How Should GitHub Fix Their Pricing?
Based on his pricing expertise, Ajit suggests several improvements:
- Add More Premium Tiers: "I think GitHub needs to do something similar to help customers scale with them. I definitely think that they are limiting their upside because people are going to start to watch the meter on their overage. But if they provide more top tier plans, developers will be more comfortable paying."
- Adjust Pro Plus Overage Fees: The current structure makes the Pro Plus tier less attractive once users exceed their monthly limits.
- Reconsider Usage Limits: The controversy around using up all premium requests in a single day suggests a need to reassess how usage is allocated.
The Broader Lesson for SaaS Pricing Strategy
GitHub's situation highlights the unique challenges of pricing GenAI products compared to traditional SaaS offerings. The balancing act between adoption and margins is especially difficult when dealing with variable backend costs that depend on model selection and usage patterns.
"They are not a regular SaaS product and making sure the math is done for each and every model and then transposed into their pricing page and making sure the pricing page is simple enough is no easy feat," Ajit acknowledges.
Conclusion
While GitHub's implementation of usage limits makes business sense to protect margins, their execution has several flaws that undermine the value proposition of their higher tier. By failing to properly segment users and create appropriate premium tiers, they risk alienating power users who would otherwise be willing to pay more.
As Ajit concludes: "And that is my experience as a pricing consultant. I hope they can consider with me. But in any case, I hope you learned a little bit about how AI companies should price and how they're pricing today."
For SaaS executives looking to implement usage-based pricing for AI features, GitHub's current situation offers valuable lessons in the importance of tier structuring, overage fee alignment, and the careful balance between simplicity and fairness in pricing design.