Paul Siroky is Using Machine Learning to Predict Your Next Purchase at TD Ameritrade
The Innovator: Paul Siroky, Director of Client Marketing Analytics at TD Ameritrade
Innovation Superpower: Harnessing the power of machine learning to predict customer purchasing behaviors.
You probably didn’t know that Paul almost became a restauranteur instead of a marketer.
Tell us about your current role and responsibilities. Why did you choose to join your current company?
Today I lead the CRM analytics function in the Marketing Department at TD Ameritrade. My team is responsible for finding ways to curate, analyze and strategically apply all of our client data to making better, more informed decisions. Being an online focused financial services company, we have a lot of information about our clients - from account information to platform interactions - and we are constantly thinking how we can use the information to ensure the solutions and services we are presenting to our clients are relevant and timely to create an easy and helpful client experience.
TD Ameritrade has a reputation as an innovator and technology-focused company, and as someone who relies heavily on data and technology to be able to solve problems, that is really appealing. The investments that our clients trust with us have significant importance in their lives, which leads to a really engaged client. This creates a considerable number of opportunities and feedback for improved interactions and experiences which makes for more exciting and impactful problems to solve.
What current developments in marketing are most inspiring to you? How will they affect the future of marketing?
I think the confluence of machine learning and AI, the volume of data that marketing departments have to work with, and the technology to personalize and really create customer-centric experiences is creating a great opportunity to get much closer to a true 1:1 interaction. Marketing departments have been trying to apply data and analytics to decisions making for a long time, but I am really starting to see the ability to impact real-time interactions and client experiences at a speed, accuracy, and level of personalization that we haven’t been able to execute at before.
Executing a few large campaigns is the past; the future entails a transition to a strategy of executing multiple smaller, customer-specific campaigns in concert and with low latency. Consumers have come to expect a personalized and real-time solution to their needs. Consumers lives are spent interacting with so many devices and platform that the number of messages we receive daily continues to grow, and the human brain can’t really process all that information. It is our job as marketers to work hard to ensure that those messages become more personalized and relevant to make our clients’ lives easier.
What are you working on now that you think is innovative?
We are working on using machine learning to choose the most relevant messages and offers for our clients. We are working to quantify the likelihood that a client is interested in specific content, services, or products as well as their preferred channel of interaction. Whether it’s an article about retirement saving options, an educational course on how to trade an option, or a conversation with a financial consultant about estate planning, we are working to use all the information we have on our clients to match them with the right solution in an analytic-driven, optimized way.
Tell us about your career path and how you ended up where you are right now: What big learning moments have you had in your journey? Did you have any notable mentors?
I started my marketing career in market research for a large retailer. One of my early projects involved analyzing survey responses and trying to validate what the customer told us to what we were seeing in the transaction logs. This was my first foray into customer databases and I was really intrigued by all the things we could learn about our customers by looking at their purchase behavior from a what, when, and which price point perspective. It was really interesting to me that instead of asking consumer about their spending patterns, we could get a more accurate view from our own data.
As I learned more about machine learning, I found that we could not only use this information to profile what a customer had done, but begin to predict what they might do or need next. This has been my focus for the past few years, continuing to find ways to improve our knowledge and foreknowledge of our clients and apply it to our strategic decision making.
Tell us about your teams. How do you pick and develop the talent on your team? How do you ensure there is collaboration?
One of the things that I look for in a team member is that they are naturally inquisitive and thoughtful, and enjoy solving problems. It is also important that analytics understand the language of the business (and vice versa) to ensure valuable collaboration and that everyone is working toward the same solution. In my career, the most successful analysts are always the ones who have strong technical analytic and data skills as well as a deep understanding of the business problems we are trying to solve.
What one thing do you need from your CMO to help you be successful?
I think the thing my CMO provides that allows me to be successful is that she is a believer that data and analytics will lead to better decisions and a more impactful marketing practice. Because there is so much foundation that needs to be laid to really start getting value from analytics including training, infrastructure, a robust and organized data environment, and smooth integration with other technologies; it requires a true commitment to the capabilities. It is often difficult to get true actionable quick wins, given the upfront cost of establishing analytic best practices and capabilities. The support to build that type of environment allows me to successfully use data and analytics to make us a better marketing department.
What advice would you give to marketers who are just starting their careers?
Be comfortable with failure. Most of the marketing you work on in your career won’t have the impact you were hoping for. Be comfortable with that, learn from that, and use that feedback to make your next, better, decision. There is no end state of a perfectly optimized campaign; you will always be looking for ways to improve upon what you’ve done in the past. Your job is to make smart, informed decision and test, test, test.
Where do you see yourself in five years?
PS: Continuing to improve interactions with our clients using data and technology. As more and more data is collected and our ability to synthesize and implement it into actionable strategies improves there will always be new problems to solve and new ideas to test.
Favorite place to vacation? Phoenix in the winter
If you were a superhero, what would your special skill be? Having the energy and enthusiasm of a toddler (so I can keep up with my kids)
Name something that most people don't know about you. I started my career in restaurant management
If you weren’t a marketer, what would you be? An Architect, I enjoy designing and building things
What's the best thing you've read/listened to/watched recently? Everybody Lies by Seth Stephens-Davidowitz