Welcome to the latest edition of our weekly newsletter. This week we talk about:
CS is still experimenting with AI
We shouldn’t overlook traditional machine learning
Let’s get into it…
CS is Miles Away from AI Adoption
As four guys we have been implementing AI projects at some of the world's biggest brands over the last number of years, and we always look at others' use of AI with interest, particularly in the Customer Success space, where we have 60 years of combined experience. Notably, many CSMs and leaders are sprinting headfirst into adopting AI, and it's for good reason:
We're being asked to do more with less
We're asked to increase revenue per head
Stable growth currently rules over mad growth
Personal fear of being left behind
With full respect to our colleagues in the CS industry, but a list of "5 AI Prompts to use in CS" isn't really going to move the needle towards AI maturity. And this is perhaps reflected in some statistics from EverAfter's recent CX survey:
Only 32% run even a single live AI use case
Just 3% describe their AI deployment as "extensive"
27% cite data quality as their top barrier
Nearly half the market remains stuck in "exploring" or "piloting" mode
This demonstrates that most of the industry is still trying to work out where AI fits into their current workflows, let alone if it's delivering results.
AI is More than Personal Productivity
From our perspective, CS needs to step back and look at the broader challenges, go beyond just personal productivity hacks, and look at the real use-cases where AI (including traditional Machine Learning, not just generative AI) can add efficiency and insights. This requires a deeper understanding of AI, beyond just its surface applications. But of course, this isn't easy to do. How many CSMs know the difference between K-Means and Random Forest models or the importance of monitoring model drift? This is perhaps why Gen AI is so attractive, it's a mostly no-code solution that's very flexible and accessible, but it isn't always the right solution.
I'm sure you could ask ChatGPT to predict which customer account is most likely to churn, but a machine learning classification model would be a better solution.
There’s a Big Gap Between Desire and Implementation
At Barre, we currently see a significant gap in CS between the desire to use AI and the actual implementation of any production use case. This gap underscores the urgent need for a more strategic approach to AI adoption in Customer Success.
And if that's where you find yourself, why not contact us and let us know what challenges you are trying to solve with artificial intelligence?
And that’s it for this edition. See you again next week!
Were you sent this newsletter? You can receive it yourself every Tuesday by subscribing here.