How to prioritize, build, and measure AI: Insights from the NYC Domonumx Customer Panel

More than 350 software leaders gathered at the NYC domonium to build a community and learn how to navigate the AI ​​era. A prominent session featured pendo customers who shared real world insights from their AI implementation journey.

Including panelists Alo Mukers (SVP Product Management, Bank US), Kelly Delaney (VP, Learning Experience Design, Savvas Learning), and Mike Bartels (SVP Workflow & Insights, Nasdaq).

Here are the main insights and practical takeaways for your own AI strategy.

How many companies have you invested in AI? How do you balance it with other software investments?

Mike Cartels (Nasdaq): We consider ourselves as a technology company, so AI is the center of our strategy. For internal operations – the efficiency of engineering and customer success – we are in 10 out of 10. Every software investment considers AI strategy and value creation. We are currently around 5 or 6 for product features because we are still sending existing road map items. We are in product integration 8 or 9 for AI.

On the product side, we are closer to five or six in terms of actual investment, just because we have the rest of the product road map to be given. Not everything that our customers asked were AI first. But we were in eight or nine when we thought about road maps after the next few quarters and integrated AI into our products.

Kelly Delaney (Savvas Learning): We are very focused on AI – 30% or more of our road maps may be based on AI features – but the actual investment in software will not change. We are considering how to make our product experience more efficient for customers. Their needs have not changed; We only find a better way to serve them.

Alo Mukers (US Bank): We are in the industry with heavy regulations and compliance, so making AI approved approved is very difficult, let alone use it in practice. We have focused more on cases of internal use and efficiency, which are more easily raised. On the customer’s side, we have a special group that focuses on AI, more on the data side, trying to find different ways to eventually provide customer value.

How do your customers feel about AI?

Alo Mukers (US Bank): Our customers handle data and financial transactions, so they do not want their privacy or data to be distributed. But they Do Want Speed ​​- They expect us to do things – if faster, whether to develop features faster, test faster, or provide faster value. They want benefits as long as we protect data.

Mike Cartels (Nasdaq): We drove AI’s conversation with customers twelve months ago. Now, they expect AI as the first or second agenda item in a road map discussion. High curiosity, and hopefully in 12 months, we will discuss who does not use it.

Kelly Delaney (Savvas Learning): About a year ago, we collected kindergarten panels through the teacher reading fifth grade. 15 out of 16 used AI regularly, but not the startup AI you expect.

They use Canva AI and other tools that have been embedded in their workflow. Its insight is that they find AI tools in the workflow they have faced, rather than going to a tab or tool that separates.

Do you use AI to solve existing or new problems?

Mike Cartels (Nasdaq): Both of them. We found a new way to help customers solve the problems they initially bought software, but we also saw other things they did outside our products and carried them.

For example, we find customers spend a lot of hours summarizing board material before the meeting. We built AI’s ability to complete this, bring the workflow to our products to increase sticky.

Alo Mukers (US Bank): The problem is the same for small businesses and restaurants, but expectations around speed for different solutions. What is changed is the possibility of a solution that we can offer.

We always talk about providing proactive insights to help customers save time and money, but that seems to be a dream of a previous pipe. Now, this is a significant possibility – we have the potential to provide insights that can be followed up to customers who will be transformational for their business.

Kelly Delaney (Savaas Learning): The problem is really the same, but the problem is truly difficult to do that we want to be part of the solution to now is within reach.

Collecting different forms of student data and provide significant feedback to teachers, or create curriculum depth to serve neurodivergen students – things that will require a very high pile of textbooks – now possible.

How do you use Pendo for Discovery on your AI trip?

Kelly Delaney (Savvas Learning): First, I have to say product discovery course Changing the way we think about ourselves as a product organization. We have seen funnels to understand the natural workflow, then determine where we can add AI solutions or the right product solutions in the natural path through our curriculum offer.

Alo Mukers (US Bank): Pendo is very important to us product discovery. This allows us to do faster things because this is a partner tool, helping us overcome the rules. We use it on the side of support and customer support for data about customer questions and all our tracking data, and we will immediately use Listen. Because Pendo has more AI features, this helps us become more advanced in our practice.

Do you measure AI features differently from traditional features?

Mike Cartels (Nasdaq): We use traditional metrics such as utilization, frequency, and how often the features are used, but we Really Trying to track the time that is stored specifically. Time is a valuable currency for any profession.

This is important because we have strong discipline to measure investment returns. We try to find out the monetization path. Is it good, better, or the best price? The price of the increase over time? Mandiri Module? We want to connect different revenues with our AI investment.

What are the suggestions that you will offer to the organization on their AI trip?

Mike Cartels (Nasdaq): First, regulating governance earlier, especially for regulated industries. Second, the power of technology is in the exponential path – things that you try three or four months ago that could not work may work now, so back and try again. Third, you will not find ideas to use AI if you don’t use it yourself. Experiment with chatgpt, confusion, or any llm in your personal life, then translate the experience to the customer’s value.

Kelly Delaney (Savvas Learning): Learn with your team and what kind of learning model it is. Share what you learn – Learning is collaborative. Prepare a chat room, task force, and share. Don’t be afraid to try new things and modeling lack of fear. We are all in the starting line, so surround yourself with people with intellectual curiosity and study together.

Alo Mukers (US Bank): Focus on where customer value is, even in the case of internal use. When Ai feels extraordinary, focus on what will help customers, and the rest will come.

This panel was held on June 17, 2025. Read more from the nyc domonumxor Explore everything in the summer release ’25.