CRO explains how AI is changing not only processes, but commercial models too
Discover how a CRO uses AI to transform sales workflows, pricing, and forecasting while proving that trust, negotiation, and customer relationships still drive enterprise growth.
Eliana Angelova
CRO at Ison Xperiences
AI Impact: AI enables Ison Xperiences to price based on outcomes, improving client retention and revenue.
Revenue Shift: The company has expanded significantly, moving from no offshore revenue to 20% in two years.
Sales Evolution: Outcome-based sales strategy increases enterprise deal win rates and aligns organizational goals.
Workflow Change: AI improves lead targeting and deal tracking, shifting focus from territory lists to data signals.
Trust Importance: AI struggles with account expansion; relationship-driven trust is crucial for upsell success.
Eliana Angelova is CRO at Ison Xperiences, a customer-experience management company with 1,600 virtual agents in production.
We sat down with her to learn how AI is changing sales and pricing. Here's what she told us.
From selling seats to selling results
I'm Eliana Angelova, Chief Revenue Officer at Ison Xperiences, based in Dubai.
Create a free account to continue reading this article.
Want more from the CRO Club? Sign up to get free access and join modern revenue leaders using proven playbooks, peer insights, and AI-driven strategies to build predictable pipeline and scalable growth.
Have an account? Log In
Ison Xperiences is a Customer Experience Management company, handling the full life cycle of client interaction. With universally available AI tools, we automate extensively internally. We built our own AI IP. That's not something most BPOs do; most buy off the shelf and badge it. We built SenshAI, our omnichannel agent assist platform. We also built hAIve for insurance, Three60 CRM for collections, and iQumex for quality assurance. These aren't prototypes. They're in production, running across 1,600 virtual agents today. This forms the foundation of our hAI framework — human plus AI.
As CRO, the numbers that matter tell the real story of transformation. Two years ago, offshore revenue was nonexistent. Today, it's roughly 20% of our book. We went from one offshore delivery location to seven. We handle over 50 million customer interactions every month, across 25 countries, in 60-plus languages. This isn't just scale; it's the infrastructure that makes us relevant to global enterprise buyers needing multilingual, multi-timezone, outcome-tied delivery.
Thanks to AI, we can compete on the same pitch list as companies ten times our size.
Three revenue pillars
My revenue organization has three pillars: pre-sales and solutions, sales, and post-sale implementation. I deliberately structured it this way because in complex B2B outsourcing, the deal doesn't end at close—implementation builds or breaks trust, and I want the revenue leadership that made the promise to be accountable for it.
Our sales team is geographically led. We cover India and APAC, China, EMEA, Africa and SA, and Germany, and we're actively expanding. Currently, I have five open positions in markets crucial to our growth thesis: two in the US, one in Kenya, one in South Africa, and one in the Middle East. So, this team is in build mode.
Pre-sales and solutions sit intentionally alongside sales, rather than below them. The deals we pursue are complex enough to require a solutions architect and a GTM strategist in the room from the first conversation — not parachuted in at the proposal stage. This function allows us to sell outcomes rather than seats.
From a GTM perspective, we go to market by industry vertical: telecom and technology, BFSI and fintech, e-commerce and retail, media, travel, and hospitality. This is deliberate because the AI transformation story differs in each vertical, and generic pitches don't win enterprise deals.
We are on a trajectory from major contender to leader in the Everest PEAK Matrix. Every hiring decision, market entry, and GTM bet we make aims to prove that move.
How AI can shift the commercial model
AI changed how we price…Once we had enough AI in the operation, we could predict and control outcomes rather than just staffing to volume. The moment you can control outcomes, you can price based on them.
AI changed how we price. Three years ago, the default commercial conversation in BPO was seats: you need 200 agents, here's the rate card, here's the headcount cost. The client bore the volume risk, and the provider had no real incentive to reduce contact. More contacts meant more agents, and that meant more revenue. It's a broken model.
Once we had enough AI in the operation, we could predict and control outcomes rather than just staffing to volume. The moment you can control outcomes, you can price based on them. So we moved deals to outcome-based pricing: tied to AHT reduction, CSAT scores, and first-contact resolution. The client now pays for the result, not the headcount.
The commercial impact of that shift is significant in two directions. First, we now win more deals, especially with European and US enterprise buyers who are sophisticated enough to know that a seat-based contract is just a managed labor bill. When we walk in and say, "We'll price on outcomes and take the performance risk," that's a different conversation. Second, we retain clients differently because if you're accountable to outcomes, you're constantly improving the operation, not just maintaining it. Renewals become much stickier. This ensures sustained long-term revenue.
So, what improved? Win rates on enterprise deals, contract lengths, the quality of client relationships, and internal discipline. When your revenue is tied to your client's KPIs, everyone in the organization is aligned on the same number. AI didn't just change our product. It changed our commercial model.
A real-world logo acquisition workflow
I will walk you through our new logo acquisition workflow today, because it's our most transformed workflow.
- It starts with account selection. We do not begin with a territory list and cold outreach anymore. AI identifies target accounts based on firmographic fit, buying signals, and vertical alignment. The model surfaces accounts showing intent signals: leadership changes, expansion announcements, and technology procurement activity. The team starts from a prioritized shortlist, not a blank sheet.
- From there, we personalize outreach at scale. By the time we make first contact, we already know the account's likely pain points, their tech stack, and who the relevant stakeholders are.
- When an account enters the pipeline, we continuously update its score based on engagement. If a deal stalls, the model flags it. The team does not discover a dead deal at the end of the quarter. They see the signal early and can act.
- At the proposal stage, the presales team uses that intelligence to build an outcome-based commercial model: the KPIs that matter to this client, our delivery platform's commitments, and optimal pricing. And everything is grounded in data, not intuition. The solutions team works with AI that has access to every previous deal we won or lost, every delivery model we ran, every pricing structure we used, and every client outcome we committed to.
- And after close, we track post-sale implementation against the KPIs we committed to. If we are drifting, we see it before the client does. That is what makes renewal a different conversation.
AI is not magic. The magic happens when you stop letting information live in silos.
Why revenue leaders must learn the difference between AI tools and AI solutions
Early on, we confused AI tools with AI solutions.
MIT published research this year that names it precisely. A generative AI tool helps individuals work faster by summarizing notes, drafting emails, and preparing for calls. A generative AI solution integrates into a process, changes how work gets done, and measures business-level impact.
Most organizations start with tools and call it transformation. We did the same. People used AI to do their existing jobs slightly faster, and we mistook that for strategic progress.
The question that would have saved us that time is simple: "Does this change the output or just the speed of the input?" If the answer is only speed, it is a productivity tool, not a business initiative.
Start with the business outcome you need to move, and work backward to the AI capability that moves it. Not the other way around.
The delineation between human work and AI work
AI drives four things for us right now:
- Lead generation: Identifying the right target accounts, surfacing intent signals, and feeding the top of the funnel so the team is not prospecting blindly.
- Pipeline prioritization: Scoring where to spend time based on fit and engagement.
- Forecasting: I will not take a number to the board unless the model validates it, because sales leaders are optimistic by nature.
- Churn risk: Tracking signals across the existing client base so we can get ahead of a problem before it becomes a conversation.
What stays human?
- Pricing on complex deals: The final call involves relationship dynamics and strategic judgment that a model does not have.
- Negotiations: Always.
- Territory planning: We are expanding in the US right now, and that was a strategic decision, not an algorithm.
- People and hiring decisions: That is always human.
The simple principle is this: AI handles what is repeatable and data-rich. Humans handle what is complex, contextual, or high stakes.
How AI presents new revenue challenges
The good results of AI are many…The challenge is that AI creates a confidence gap in the team. I also underestimated how much client education is required.
The good results of AI are many. Lead generation quality improved significantly because we target better accounts earlier. Forecast accuracy increased because we use model-validated numbers rather than the sales team's gut feel. The shift to outcome-based pricing made our client relationships stickier because when we are accountable to the client's KPIs, renewal becomes a very different conversation.
The challenge is that AI creates a confidence gap in the team. When a model tells a salesperson to deprioritize a deal they believe in, it creates friction. Building trust in the output takes time, and the team sometimes ignores the signal and gets it wrong. That costs us.
I also underestimated how much client education is required. Outcome-based pricing sounds compelling in a pitch, but shifting a client's mindset after they have bought seats for ten years requires a real commercial conversation. AI is changing our model faster than the market is ready to absorb.
Why AI falls short on account expansion
And I'll say this, too: AI has fallen short of our expectations. We expected AI to help us identify upsell and cross-sell opportunities inside the existing client base more systematically. Surface the signals, prioritize the conversations, move faster. And the signals are there. The data tells us when a client is growing, when their volumes are shifting, and when there is a logical next service to introduce.
But we find that data does not trigger expansion conversations in enterprise BPO. Trust triggers them. And trust lives in the relationship, not in the model. The clients who expand with us do so because a senior person picked up the phone, sat in a room with them, and had a conversation that was not on any sales script. AI cannot replicate that, and it cannot accelerate it either.
How the profile of a great salesperson is changing with AI
The best salesperson is no longer the one with the biggest network.
In enterprise BPO, the prevailing wisdom has always been to hire grey-haired veterans with deep rolodexes, because relationships close deals. And relationships still matter. But AI has shown me that a data-literate, AI-enabled salesperson who enters a conversation having already mapped the account's pain points, understood their buying signals, and built a commercially precise proposal will consistently outperform the veteran who is relying on rapport alone.
The profile of a great revenue professional is changing, and holding onto the old definition means hiring for the past, not the future.
Why data and collaboration are the biggest predictors of AI success
I have two pieces of advice:
The first is to get into the data before you get into the tools. MIT research from last year found that 95% of enterprise AI pilots failed to deliver measurable business impact. The reason is almost always the same: The data underneath the tool is inconsistent, incomplete, or untrusted. Clari found that 67% of revenue leaders do not trust their own data. If that is you, no AI tool will save you. Fix the foundation first.
The second is to get aligned with your technology leadership. Clari's research shows that CRO and CIO alignment is the single biggest predictor of AI revenue impact. The CROs who are winning are not buying tools independently. They are building a shared data strategy with their technology counterparts and making AI a company-wide capability, not a sales department experiment.
Overall, use AI to sharpen your thinking, not to replace it. Stay curious, stay close to the data, and never stop making judgment calls.
Follow along
You can follow Eliana Angelova's work on LinkedIn.
More expert interviews to come on The CRO Club!
