CX as ROI
AI ROI is one of three customer-facing levers, or it is burning tokens.
What this lens looks for
AI investments earn ROI through one of three customer-facing levers. AI that does not move at least one of them is burning tokens. This lens forces every announcement, deal, or deployment into a three-way decomposition before the verdict lands.
The three levers
Cost-to-serve. Same service, lower unit cost. Either fewer humans per transaction, less infrastructure per workload, or less downstream rework per outcome. The buyer-facing question: does the customer get the same thing for less of our spend?
Capacity reallocation. Same team, more customers served. Either faster cycle times, broader hours of coverage, or higher concurrent throughput. The buyer-facing question: did we open economically viable work that we could not staff before?
Customer-felt functionality. A customer can do something materially better in the service experience than before. Not internal productivity, not training completion, not seat adoption. Externally visible performance improvement in the journey. The buyer-facing question: does the customer notice a difference they would describe in their own words?
When we apply it
- An announcement claims ROI without specifying which lever
- A deal headline uses “transforms CX” or similar instead of naming the operating change
- An internal-productivity metric (hours saved, tickets per agent) is being read as customer-experience proof
What the verdict looks like
Articles through this lens map every named metric to one of the three levers, name the lever that was actually pulled, and call out the levers that were claimed but unevidenced.
Where the lens comes from
The framing is rooted in three decades of customer-experience work at scale and is the load-bearing strategic frame for this publication.
Articles through this lens
-
Smaller Models Can Preserve Performance When the Loop Is Good
The right loop lets teams route simple work to cheaper models without giving up quality. That is how AI systems lower cost while keeping the user experience steady.
-
Burning Tokens Without CX Proof Is a Budget Smell
Lower inference cost can improve a model budget line, but it does not prove customer value. Fund only when one CX lever has a baseline, owner, and measurement window before rollout.
-
What Counts as CX ROI Evidence Before Funding an AI Project
A three-lever funding gate for AI projects. Cost-to-serve, capacity-reallocation, and customer-felt functionality are the only places a project can cash out. If the vendor pitch does not disclose evidence for at least one lever with a pre-declared baseline, the request is positioning, not operating proof.
-
What Andy Crowder Actually Bought: A Three-Lever Read of the PwC-Advocate Health Deal
The May 14 PwC-Advocate Health expansion through the CX-as-ROI three-lever filter. Capacity is the lever pulled; cost-to-serve and customer-felt functionality remain claimed but unevidenced.