MarTech Interview with Scott Michaels, Chief Product Officer at Apply Digital

Scott, welcome to MarTech Cube. Can you please share your professional journey and how it has shaped your vision as Chief Product Officer at Apply Digital?
My career has followed a pretty unconventional path. I studied English, not technology, and started out as a technical writer at InfoNet in the late ‘90s, documenting large industrial systems. But instead of just writing manuals, I realized it was faster to fix the bugs I was documenting than write about them. That led me to learn programming myself and move into the tech side of the business.

But from the start, my focus was always on the user; understanding their needs, frustrations, and what they were willing to pay for. That instinct for product thinking shaped my career. For example, I worked in the Apple ecosystem before the iPhone existed, porting software to Mac, and was fortunate — in part because there were just so few of us working in that space back then — to help develop some of the first third-party iPhone apps for brands like The New York Times and ESPN.

After that, I worked in startups before co-founding Apply Digital nearly a decade ago. We saw a gap in the market: plenty of IT and development firms existed, but few could build full-fledged products. Today, we’re a 750-strong team, working across industries to develop platforms and experiences that drive enormous business value.

You have lots of experience working with innovation labs. What role do they play in digital transformation?

Large companies often reach a point where they become consumed by maintaining and iterating on their existing products. This creates a natural barrier to innovation. Dedicated labs — which have a variety of names, so call them what you will — are structured teams dedicated to exploring new ideas outside the constraints of daily operations.

But most labs fail. They turn into think tanks or hackathon hubs, producing ideas that make incremental gains if at all, but not serious revenue. This is usually because they lack clear targets and don’t set the bar nearly high enough. One business we work with only pursues innovations projected to generate at least $100 million. That focus changes everything. It forces teams to think bigger and measure success rigorously.

The key is interdisciplinary teams, not just tech or marketing, but also finance, operations, and compliance. Without this, ideas remain conceptual rather than commercially viable.

Where do you see AI making the most significant impact in fostering genuine innovation within the labs?
AI is hugely valuable at both ends of the innovation cycle. At the start, it speeds up research and market analysis. Work that used to take weeks or even months can now be done in hours. At the other end, AI is excellent at helping to optimise products post-launch, improving performance through rapid data analysis, refining user engagement, and enhancing marketing effectiveness.

The real shift, though, is AI becoming a fundamental part of the products themselves. There’s almost no project today where AI isn’t embedded in some way. It’s no longer an optional feature but a core component of innovation.

How do you avoid the “type trap” in an innovation lab, where teams might overfocus on specific technologies or methodologies?
Generally speaking, by setting ambitious targets. As I mentioned, many teams get stuck in small-scale optimisations, like trying to shave milliseconds off load times or automating minor workflows. That’s useful but not transformational.

Another risk I’m increasingly seeing regards the provenance of AI. Many companies limit themselves to the tech stacks of their preferred cloud providers. If you’re locked into Microsoft, you focus on OpenAI. If you’re with Google, you prioritise Gemini. But these models vary dramatically in strengths and weaknesses. The best companies test ideas across multiple AI models – regardless of cloud vendor alignment – to find the best tool for the job. The reality right now is the AI landscape is moving at a pace that hasn’t previously been seen in our industry, and vendors are regularly leapfrogging each other. Consequently, it’s a huge mistake to not look at the entirety of platforms you could use in your AI endeavors.

Outside of innovation labs, how can businesses move from unrealistic AI dreams to tactical applications with immediate productivity gains?

AI adoption is following the natural curve of how employees engage with any new tool. Some are early adopters, the honey badgers who dive in and champion AI-driven workflows. Others need structured enablement.

The best approach is to create a safe, approved AI environment where employees can experiment without compliance risks. Most companies struggle because employees don’t trust where their data is going or they have become scared in the wake of hearing about data leaks or inappropriate use. But if you provide a secure internal AI workspace — whether for writing, data analysis, or workflow automation — employees will discover use cases organically.

Another key tactic is trumpeting good ideas internally. When someone finds a productivity win, they should be encouraged to share it widely. Change management isn’t about forcing adoption — it’s about letting real use cases spread through peer influence.

With 2025 set to be a big year for agentic AI, what do you anticipate its impact will be on MarTech and e-commerce?
Agentic AI — which can learn and make decisions on your behalf — is really going to reshape personalisation. In MarTech, the goal has always been marketing to a ‘cohort of one’ rather than broad segments. Agentic AI brings that much closer, dynamically crafting and delivering personalised messaging, ad creatives, and even buying decisions in real-time.

For e-commerce, the bigger shift might be AI-driven shopping agents. Already, platforms like Perplexity allow users to ask, ‘What are the best noise-cancelling headphones?’ and receive not just recommendations but buy buttons. If this trend grows, it means e-commerce platforms need to rethink discovery. For example, companies that create ‘fast lanes’ for AI shopping agents — offering structured, high-trust data on inventory, pricing, and reviews — will end up capturing more sales as AI becomes a new proxy consumer.

What advice would you give business leaders implementing AI in their digital transformation strategies?

Think of AI as a change management challenge, not just a technology rollout. To embed AI meaningfully, you need broad internal buy-in. That means working not just with IT, but also compliance, HR, and legal to create policies that encourage experimentation rather than restrict it.

Leaders should focus on ensuring employees feel safe using AI in their daily workflows. Many companies have the technology but lack an adoption strategy. If employees hesitate to use AI tools because they fear security breaches or job risks, the initiative will stall.

Also, avoid building novel AI models in-house unless it provides clear competitive differentiation. Many AI capabilities will soon be embedded directly into core enterprise software — Word, Excel, Google Docs, CRM platforms; you name it. Focus your internal efforts on unique AI applications that create real business value rather than duplicating what vendors will soon offer as standard.

For more expert articles and industry updates, follow Martech News