About the expert
Host, CXO Loading...
Lorena Lovric
Director of Customer Experience
Lorena Lovric is an award-winning Director of Customer Experience whose work connects AI transformation, agent experience, and customer outcomes. Her agent-first approach starts with the work people do today, not a technology promise in isolation.
She is clear that automation can remove repetitive tasks, but it cannot substitute for judgement, culture, or the difficult human conversation that makes a customer feel understood.
Ready to hear Lorena’s insights?
Listen nowListen now
Episode chapters
- Introducing Lorena: agent-first AI and the leadership work around the technology
- From the front line to CX leadership: keeping the first-hand perspective
- Senior responsibility, real leadership, and why CX should still be fun
- The Roomba warning: clean up the house before you automate it
- Making AI a playground: building with CX and technology, not around them
- Customers are not ticket numbers: speed, attention, and joined-up service
- Happy agents, happy customers: what stays constant as the tools change
- Staying sharp through Women in CX, community, and shared perspective
- Agents first, then the tools: safe experimentation and choosing a platform
- 194,000 customers, 15 agents, and a recall built with the front line
- Avoiding AI scope creep: pick one thing, fix it, then build further
- Creating a safe, high-energy team through fun, visibility, and celebration
- Lorena’s Director-to-CXO advice: stay true, experiment, and learn from mistakes
- The CXO Codex, Women in CX, and closing thoughts
The learning
AI can automate tasks. It cannot create culture.
AI has made it easy to talk about speed. The more useful question for a CX leader is what the organisation will do with the time it gets back. If automation removes a repetitive task, will the team gain more room for judgement, coaching, and better customer conversations , or will a fragile process simply move faster?
Lorena Lovric’s agent-first perspective gives this chapter its test. AI can automate tasks. It cannot create culture, make a difficult commercial trade-off, or tell a customer that the organisation has genuinely understood what has gone wrong. Those remain leadership choices, designed into the operating model before the technology goes live.
AI can automate tasks. It cannot create culture.
Map the process before automating it. Do not scale confusion.
Decide explicitly what your team will do with the time automation returns to them.
Measure employee experience as seriously as customer experience , culture is operational, not decorative.
“Clean up your house before you bring in a Roomba , otherwise it only drags the mess around faster.”
Lorena Lovric, Director of Customer Experience
What research adds
The pattern is bigger than one good story.
Research
McKinsey identifies data and systems integration, process design, change management, and customer adoption as practical barriers to AI-led customer-care transformation , not technology capability alone.
McKinsey · The contact center crossroads: Finding the right mix of humans and AIResearch
A qualitative customer-service study found that trust in AI chatbots is shaped by both human-like interaction and emotional connection, alongside reliability, accuracy, transparency, responsiveness, and data security.
Scientific Reports · Building user trust in AI chatbots for customer service through human-like cues and perceived reliabilityThe Codex reading of Lorena’s agent-first approach is that AI value is created in the design around the tool: the process that is ready to automate, the judgement that must remain human, the guardrails that make escalation safe, and the work leaders deliberately give back to people when time is returned.
A faster service flow is not automatically a better experience. The more useful investment case holds two outcomes together: what improved for the customer and what higher-quality work became possible for the team. If neither can be named before implementation, the organisation may be automating activity rather than improving the system.
Make it actionable
Use The Pre-AI House Check to make the idea real.
Use the Pre-AI House Check before selecting a solution for one service workflow. Make the current work visible, define the exact task you want AI to remove, and decide where judgement, escalation, and the time returned to people will go instead.
The leadership tension
Automation does not arrive in a neutral system. It amplifies the hand-offs, data gaps, unclear ownership, and service habits already underneath. That is why a tool that looks efficient in isolation can make the experience worse at scale: it can remove the moment where someone would otherwise notice that the process is broken.
A human-centred AI decision therefore needs two measures of value. The first is customer value: did the interaction become easier, clearer, or more reliably resolved? The second is human value: did the time returned make room for higher-quality work, or did it simply create a new expectation to do more with less?
Team tool
The Pre-AI House Check
Before automating, make the current work visible and decide what the newly available human time is for.
Map the current hand-offs and name the exact task AI is meant to remove , not a vague ambition to ‘use AI’.
Identify the broken workflow, data gap, unclear decision, or risk that automation could otherwise scale.
Define escalation and quality guardrails, then protect the time AI returns for coaching, judgement, or higher-value problem-solving.
Your next move
Day 1
Choose one repetitive customer-facing task and ask your team: if this disappeared tomorrow, what higher-value customer work would we choose to do with the time?
Build it in 30 days
Run a small AI readiness review for one service workflow: define the customer outcome, process/data readiness, human hand-off, guardrails, and the work reinvestment plan before selecting a solution.
Measure the shift
Track the customer outcome and the human outcome together: resolution quality, repeat contact or transfer rate, agent time returned, and how that time was reinvested.
How this creates movement from Director to CXO
You are building more than a good idea.
The move in Lorena’s chapter is from sponsoring an AI tool to designing the human operating model around it. CXO-level leadership means being able to name the work that should disappear, the judgement that must remain human, and the value the returned time should create for customers and teams. The Pre-AI House Check gives that investment conversation a practical starting point; Track the customer outcome and the human outcome together: resolution quality, repeat contact or transfer rate, agent time returned, and how that time was reinvested.
Additional resources
Keep the useful bits close.
Select one or more filters and keep only the inputs that sharpen the next decision you need to make.
Select one or more filters
Podcast
The Retail Tea Break
Melissa Moore’s conversations with retailers, brands, and industry leaders about the decisions shaping modern retail experience.
Website
Lorena Lovric
Lorena’s writing and work on AI-driven CX transformation.
Research
McKinsey · The contact center crossroads: Finding the right mix of humans and AI
McKinsey identifies data and systems integration, process design, change management, and customer adoption as practical barriers to AI-led customer-care transformation , not technology capability alone.
Research
Scientific Reports · Building user trust in AI chatbots for customer service through human-like cues and perceived reliability
A qualitative customer-service study found that trust in AI chatbots is shaped by both human-like interaction and emotional connection, alongside reliability, accuracy, transparency, responsiveness, and data security.
Save & Share
A printable version of this material, built around the same practical question, research, tool, and next move.
Download guideContinue the Codex
The next useful answer may sit in a different conversation.
Trust is the system
How do you make CX a business-wide leadership system, not a service function?
With Lara Khouri
The accounts that go quiet
How do you spot risk before customers complain, escalate, or leave?
With Abisola Fagbiye