Editorial
The Gist
- AI impact. Generative AI revolutionizes customer success, enhancing service quality and efficiency.
- Data strategy. The success of AI in CS depends on the richness of data available for training.
- Cost efficiency. AI reduces the need for large CS teams, balancing cost and customer satisfaction.
Customer success (CS) seems to be right in the bullseye of what this first wave of generative AI is likely to impact. Some recent indications of this are Klarna’s incredible results deploying AI customer support agents, and the number of startups going after this space (most notably, Brett Taylor’s Sierra).
A big reason for this is customer success’s role as the translation layer between product knowledge and the customer needs, a job description ripe for large language models (LLMs). Solutions architects translate between customer workflows and product configurations. Success managers translate between product roadmap and upsell opportunities. Support agents translate between customer pain points and product documentation.
At the risk of oversimplifying, CS’s North Star is its translation quality, i.e., the goodness of the fit between the product output and the customer need. And in no part of customer success is AI upleveling on translation quality faster than in support.
Let's explore some aspects of AI in customer success.
AI Support: Beyond Human Capabilities
The basic support job-description can be done surprisingly well by LLMs today for the simple reasons that they’re available 24/7, have perfect knowledge coverage of product documentation and roadmap, they have access to all customer interaction history and can learn on an ongoing basis from human input (whether they be customers or product managers). There are known shortcomings of course (like hallucinations), but I would take the bet that these are solvable.
Beyond doing human jobs better, AI support agents today can also do what humans struggle with: summarize hundreds of customer support interactions across phone/email/chat to help the product team prioritize what to build next, track all features being shipped and proactively inform the cohorts of customers who ever expressed a related pain-point, etc.
We’ve only been talking about support so far, but it’s not hard to envision a natural leap to success manager agents (akin to an AI concierge service) and even solutions architects co-pilots guiding customers through product deployment and configuration. The "agent-ification" of customer success.
Related Article: 8 Ways AI Can Elevate Your Customer Experience
Building an AI-Native Customer Success Function
What does this mean for how you build an AI-native customer success function? Let’s first understand CS as it exists today. Other than for the highest tier enterprise white-glove customer, CS is largely a BAND-AID for missing product features and usability friction. Meaning, it is a cost-center that should be aggressively limited in its usage to keep your margins healthy.
Said another way, in its terminal state, 80%+ of CS headcount should be replaced by AI, and the remaining <20% (reserved for the hairiest issues and highest LTV customers), heavily augmented by it (the metaphorical super-hero cape).
Related Article: Dear CX Leaders: Are You AI-Ready? AI in Customer Experience Is Here
AI Enhances Customer Success, Cuts Costs
Customer success leaders pre-AI solved for this cost-center issue using a variety of creative solutions like setting restrictions on number of hours a mid-market/SMB customer could leverage CS and billing for overage, growing the number of customers assigned to each CS rep, building online help-centers and certifications for customers to self-serve, creating customer communities in Slack/Discord to answer each other’s questions, etc.
These “scaled programs” were valuable but never quite replaced the quality of 1:1 CS interaction and left customers dissatisfied. This inevitably led to unreported hours by the CS team because a dissatisfied customer is a renewal risk…and as you know, every renewal is precious. This tradeoff of having a large CS headcount on the books or risking poor customer satisfaction is a false dichotomy in this emerging generative AI paradigm.
For example, you can deliver customers of any tier unlimited high-quality 1:1 support interaction via AI agents (like Klarna is attempting). Looking more closely at success management, your AI agents can take on higher order tasks like building custom dashboards and metrics for your buyer to help them understand how their organization is adopting the product, etc.
One level up from that is facilitating an upsell (i.e., being able to recommend certain features to a customer based on their usage profile) and completing that motion all the way to invoicing. Meaning, for a small marginal-cost, your company can serve the next customer to the fullest degree with minimal full-time employees or effort by utilizing AI in customer success.
Related Article: Customer-Centric AI Strategies and Why You Need One
Data Quality Key to AI Success
It's very hard to predict the timing of this — progress will appear slow at times and disorientingly fast at others. What is clear is AI in customer success will only be as good as the quality of data your AI is trained on. What will differentiate your AI in customer success function from that of your competitor’s is not so much the caliber of CS talent you can hire as it is the quality and richness of data you have available for your AI model.
Anyone will be able to ask GPT-5 to come up with a generic upsell deck using basic product and customer information. However, having the model know the context behind the purchase from the Salesforce opportunity, their usage metrics from your Segment instance, and their engagement with your campaigns through your Hubspot data can create game-changing upsell content.
Related Article: AI in Customer Success: Companions for Enhanced Engagement
Data Strategy: Foundation for AI-Driven Success
A lot of your AI in customer success strategy is simply your data strategy. To that end, the best thing you can start doing today is ensuring that critical customer and product data across channels is recorded and stored in line with enterprise data best-practices (i.e., single source of truth for each data-source, unique IDs for customers across tables, clear data provenance, etc.).
Your AI agents should be able to read through past support tickets, product documentation, product roadmap, pricing tiers, customers' invoicing data, etc., and figure out how to service that relationship. Like with all things, the path is windy and will start with small leaps from hard-coded logic in your support tool to co-pilots helping humans get more done to agents doing the work for humans who simply set the strategy and continuously train and refine the model.
Related Article: Gathering Consumer Data That Matters for Measuring Customer Success
Webinar
Sep
25
CMS Briefing: A Live Look at What’s Next in AI-Driven Platforms
Learn how leading organizations are using AI‑driven tools to publish faster, personalize smarter and stay secure.
Register
Webinar
On demand
Ready or Not: How Data-First Organizations Are Unlocking Agentforce Potential
Learn how to cut through the noise, activate Agentforce and build a Salesforce AI strategy that actually delivers.
Watch Now
Webinar
On demand
AI in Customer Service: Faster Resolutions, Happier Customers
Don’t let rising demand burn out your team. See how to build a smarter, more resilient support org.
Watch Now
Webinar
On demand
From Hype to High-Impact CX Strategies That Actually Scale
Turn buzzworthy AI and outsourcing trends into measurable CX wins with fresh 2025 data.
Watch Now
Webinar
On demand
Insights to Action Rethinking the Contact Center for Real Business Impact
Join our exclusive webinar to hear CX executives share their innovative strategies for transforming service delivery.
Watch Now
Webinar
On demand
Accelerating Healthcare Ops with AI
Sign up to see how your peers are putting AI to work — and making it stick.
Watch Now
Final Thoughts on AI in Customer Success
To wrap up, I want to borrow from Benchmark's Matt Cohler, venture capitalist and a member of LinkedIn's founding team, “Our job is not to see the future, it’s to see the present very clearly.”
I can’t tell you what AI in customer success will look like in five years, let alone 10, but a clear-eyed view of the present makes me confident in saying it will be dramatically AI-impacted: customers will be happier, business margins will look better and the product team will have the best translations they’ve ever had of customer needs.
Learn how you can join our contributor community.
