AI x CIO: How CIOs Drive AI Adoption That Delivers Real Business Outcomes

I’ve been thinking a lot about why so many AI initiatives, despite incredible advances in technology, still fail to deliver real business value. And what keeps coming up in my conversations with CIOs and technical leaders is that AI doesn’t fail because the models aren’t good enough. AI fails because organizations struggle to adopt it in meaningful and sustainable ways.  

To see what was really going on with our customers who are actively solving this problem today, I sat down with Chris Colangelo, AVP of Business Technology & AI at Verizon. He’s a leader who has been on the front lines of building and scaling AI in one of the world’s largest enterprises. During our discussion, Chris shared what really separates successful AI initiatives from the ones that die on the vine.  

At the heart of our conversation was a simple truth: AI can do almost anything, but without a clear problem to solve, it rarely drives real impact.  

Start With the Business Problem, Not the AI 

Chris believes that AI breaks down when teams treat it like a shiny tool. He said it’s like buying a brand-new buzz saw and then walking around asking, “What can I cut with this?” That mindset is exactly why so many AI initiatives fail to deliver value. 

At Verizon, the approach is the opposite. Chris and his team start by putting AI aside and asking a few disciplined questions:  

• How big is the problem we’re trying to solve?  
• Is AI uniquely suited to solve it?  
• And will solving it create a real competitive advantage?  

If the answer isn’t clear, it doesn’t move forward, no matter how exciting the technology looks. 

Verizon has cracked the code on how to best support sellers, especially new ones or those working with highly technical products. Instead of expecting every rep to master a complex portfolio, Verizon uses AI to help sellers respond faster and more intelligently, by leveraging AI to surface relevant knowledge about the prospect, draft personalized responses, and handle the heavy research behind the scenes. 

So what do they end up with? Better customer conversations, at scale. All because they focused on using AI on a clearly defined business problem. 

Outreach steps in to help here by operationalizing this mindset. Outreach helps customers anchor AI to specific revenue outcomes, whether it be pipeline creation, deal progression, or seller productivity. Our AI Agents are designed around those clear business moments in the revenue workflow that our customers care so much about, not just generic “AI features.” 

Adoption Is the Hard Part and Most Teams Underestimate It 

Chris was very clear about this: adoption is the hardest part of AI, and anyone who says otherwise hasn’t done it at scale. At Verizon, adoption is factored into the process from the beginning. He calls this adoption mitigation: evaluating, before a project even starts, how likely it is that people will actually use the AI. If adoption risk is high, the project drops in priority, no matter how promising it looks on paper. 

Outreach helps Chris’ team by embedded AI directly into required sales workflows. A seller might be creating a sequence or reviewing a call, and Outreach’s AI capabilities are right there assisting the motion. Sellers at Verizon are getting value without needing pesky reminders to opt in or change behavior.  

Adoption doesn’t happen by hoping or throwing hundreds of training sessions on the calendar. It happens by design, when AI is embedded directly into the workflows people already depend on.

How Verizon Uses AI to Augment Sellers, Not Replace Them 

Verizon’s goal with AI is augmentation, not automation. The intent isn’t to replace human sellers. It’s just to make them better at the parts of the job that actually matter.  

In practice, that means giving sellers AI-powered access to knowledge, especially when products are highly technical or when reps are still ramping. AI helps read incoming customer emails, research account details, troubleshoot issues, and draft thoughtful, personalized responses. Humans never leave this process fully though. They’re firmly in the loop, reviewing, refining, and sending the final message. It’s a philosophy we share at Outreach as well: let sellers do what they do best, while AI takes care of the rest. 

For Outreach customers, this means sellers stay firmly in control while AI handles the heavy lifting. The result is higher adoption, faster execution, and more time for sellers to focus on what only humans can do: building trust and advancing customer conversations.

Build vs. Buy: How CIOs Think About AI Strategically 

The build-versus-buy question is one I’m all too familiar with. Every CIO faces it when evaluating AI. Chris and his team approaches it with pragmatism. Most organizations will do some of both, but the key is knowing where differentiation truly matters. When strong solutions already exist and the capability is largely generic, buying is the smarter, faster choice. There’s little value in reinventing what the market already does well. 

He reframes the build-vs-buy debate around strategic time horizons. Verizon buys when AI is deeply tied to first-party data, proprietary workflows, or processes that create competitive advantage. He aims for a six-to-eighteen-month strategic edge, because this window is often enough to shift outcomes.  

Outreach helps customers in securing time advantage with pre-configured, workflow-native AI agents that provide immediate leverage without years of internal development. But flexibility is at our core, so we make sure our agents can integrate seamlessly with any proprietary data or systems.

Why Pilots Fail and What CIOs Must Do Instead 

Chris shared a story that really stuck with me:. To get a pilot’s license, you can read every manual, watch every training video, and even sit in the simulator, but until you actually get in the cockpit and fly, you don’t really know what it takes. The same is true for AI pilots. Proofs of concept are easy to build and quick to launch, but production at scale is in another stratosphere. 

Outreach is built for production from day one. In fact, Verizon has leaned on Outreach to embed AI into live revenue workflows to be measured against real outcomes, not just demos.  

The lesson for CIOs is clear: don’t treat pilots as an end in themselves. AI must be integrated into workflows from the start, with a plan to scale quickly once value is proven.  

Workflows Are Where AI Either Wins or Dies 

One of the clearest lessons that came through from Chris’s experience at Verizon is that AI fails when it asks people to change how they work. Even the most advanced models stall if they live outside the flow of daily activity. 

Verizon prioritizes what he calls “background AI.” This is intelligence that runs inside required workflows, activates automatically, and delivers value without any additional steps. When AI is embedded into the workflows people use everyday, adoption becomes guaranteed by design. 

That’s the philosophy foundational to us at Outreach and why we’ve been able to help Verizon so seamlessly embed AI into their processes. Outreach doesn’t layer AI on top of sales tools. It’s also not just a chatbot where you get a quick answer to a question. Outreach's AI Agents are fully autonomous agents working in the background on behalf of sellers, helping sellers be more productive and enhancing the quality of their work. This is how AI moves from promise to performance and delivers real ROI. 

Outcomes Matter and ROI Is the Real Bar for AI 

If there’s one thing that separates successful AI programs from the rest, it’s that outcomes matter more than outputs. Chris’ experience reinforced what I see across enterprise AI everyday. Building tools is only half the story. The real test is whether AI delivers measurable value for the business. Verizon tracks ROI closely, making sure every investment in AI drives impact, not just experimentation. 

It’s why I’m seeing a broader shift toward outcome-based partnerships. Vendors should be accountable for results. In Q4 alone, our internal AI agents supported hundreds of meetings booked and contributed to over $16M in created opportunity, with dozens of autonomous agents operating continuously in the background. These results reinforce a simple truth: AI only delivers ROI when it’s operationalized inside workflows and held accountable to business metrics. 

For CIOs, that means focusing on solutions that integrate into workflows, deliver real metrics, and enable teams to achieve their goals faster and smarter.  

Watch the Full Conversation 

This interview is part of our AI x CIO series where we explore how enterprise leaders are driving adoption, scaling AI, and realizing real business outcomes. To catch up on earlier episodes, check out the resources below.