You aren’t a robot. You aren’t an AI specialist. You are more than 2 years old. So what could you possibly have in common with ChatGPT?
The new kid on the block
Open AI, the company that built ChatGPT, is under enormous pressure. For the past 2 years, ChatGPT has been synonymous with generative AI, just as Google is synonymous with web search and Kleenex is synonymous with tissues.
That might be about to change. A Chinese company, DeepSeek, has recently taken the AI industry by storm. They claim to have produced a better AI model than ChatGPT at a fraction of the cost. And only a few days later, Alibaba made a similar claim.
As an aside – it doesn’t surprise me that the Chinese would beat an American innovator at its own game. I spent a few weeks in China recently, and it quickly became clear that China’s technology is first class. We all think that China copies western innovations. The tide is turning.
Back to ChatGPT and Deepseek. Whether Deepseek has trained its AI models as cheaply as it claims is up for debate. But one thing is certain, they have trained their AI models at a significantly lower cost than ChatGPT. This will have huge ramifications for the future of AI and the financial viability of ChatGPT.
And this isn’t the only story relating to ChatGPT’s financial viability to hit the press recently. A lesser-known story (but equally important to anyone who cares about pricing) is how ChatGPT set their subscription price of US$20 per month.
Chat GPT’s approach to pricing
The CEO of OpenAI, creator of ChatGPT, is named Sam Altman. Sam has been a well-known personality in Silicon Valley for years, primarily due to his role as founder and CEO of YCombinator. If you aren’t familiar with YCombinator, it is the best-known startup accelerator in the Silicon Valley. It has springboarded thousands of unicorns including Stripe, Airbnb, Doordash, Coinbase, Reddit, Dropbox, and Zapier.
I spent many hours listening to Sam’s webinars when I first became self-employed. He was very insightful. For example, here is a great lecture series. https://www.g51edu.com/how-to-start-a-startup/. The series covers just about everything you need to know about starting and growing a business. Except Pricing.
You might think that someone with Sam’s business knowledge and experience would have a solid pricing framework covering a range of business models such as subscription, marketplaces, e-Commerce, etc. And there would be no better business to apply this framework, than ChatGPT. After all, ChatGPT has 300 million regular users, of which 10 million are paying subscribers. Sam and his team could have tested hundreds of pricing strategies on multiple user across dozens of countries.
Or perhaps ChatGPT could have used AI to set its prices. But no, they didn’t do that either.
So, have a guess. How did they set prices? I’ll give you a hint. The word to describe their pricing approach appears in the first sentence of the paragraph you are reading right now.
That’s right, they guessed.
Don’t take my word for it, here’s what Sam had to say:
“I believe we tested two prices, $20 and $42. People thought $42 was a little too much. They were happy to pay $20. We picked $20. It was not a rigorous ‘hire someone and do a pricing study’ thing”.
I find it so disappointing that someone with Altman’s intellect, knowledge and resources would simply guess when it came to one of ChatGPT’s most important commercial decisions. They knew that $42 was too much. What if they had tried $30? They might have increased revenue by 50%. And when your revenue exceeds $1 billion each year, 50% is a big number.
Right now, you might be wondering why Sam didn’t engage a pricing expert. The ROI would have been so high! What a missed opportunity. The same might be said for most Build to Rent businesses.
Having identified this problem, did Sam learn from it? NO!
ChatGPT recently launched a new Pro pricing tier, with unlimited usage for $200 per month. Whenever a business launches an unlimited pricing tier, they hope their customers use the service as little as possible. Gyms are a great example. But Chat GPT’s Pro subscribers turned out to be more enthusiastic than the average gym goer. The users love the unlimited ChatGPT service so much that they can’t get enough. And every time they use it, it places more strain on ChatGPT’s servers. As a result, ChatGPT is losing money on this tier. A lot of money.
The implications for the Build to Rent Industry
As a BTR operator, you have more in common with ChatGPT than you might realise:
- Your customers pay a monthly “subscription”.
- Different customer segments have different budgets and requirements.
- You have lots of data to help you optimise what price to charge to each customer.
- The chances are, you aren’t using that data as well as you could, and it is costing you money. A lot of money. Revenue Management is your friend.
If you would like to make better use of your data, Price Wizard is here to help. Get in touch to set up a demo.
