Inspire EMEA 2022 On-Demand is live! Watch now, and be sure to save the date for Inspire 2023 in Las Vegas next May.

Alter Everything

A podcast about data science and analytics culture.
Alteryx Community Team
Alteryx Community Team

Welcome to a special episode of Alter Everything!

Alteryx’s own Chief Data and Analytics Officer, Alan Jacobson, is our guest host, and is joined by Terry Hickey, Chief Analytics Officer at CIBC in Canada. Alan and Terry chat about their experience driving analytics culture at large organizations. Listen through to the end to hear about how Terry likes to reset in his busy schedule, and where he would like to take his dream vacation.

Also be sure to take our audience engagement survey to share your listening preferences and suggestions to make Alter Everything your dream soundtrack to pair with your analytics lifestyle!






Terry Hickey - LinkedIn, Twitter

Alan Jacobson - @AJacobson, LinkedIn, Twitter




Community Picks





Episode Transcription

MADDIE: 00:14 

[music] Welcome to Alter Everything, a podcast about data science and analytics culture. We have an exciting episode in store for you today with Alteryx's own chief data and analytics officer Alan Jacobson as our guest host. Alan is joined by Terry Hickey, chief analytics officer at CIBC in Canada. Let's get started. [music] 

ALAN: 00:44 

Okay. So hi, Terry. I'm Alan Jacobson. I'm the chief data and analytic officer here at Alteryx, and I am really excited to get to be on with you today. For our guests, Terry Hickey is the senior vice president and chief analytics officer at CIBC and has been with CIBC for, I think it's been over a year, coming from IBM where you were the vice president, leading work with Watson AI. So Terry, can you tell us a little bit about your role at CIBC and what you currently are focused on? 

TERRY: 01:11 

Sure. So first of all, thank you so much for inviting me. Super happy to talk about the things that we're doing here at CIBC. So we embarked on our journey, just as you said, over about a year ago, and one of the first things that we did was we noticed that we were lacking an artificial intelligence, or AI, strategy or advanced analytics strategy for the business. And we went away and we worked with parts of the business to come up with a plan that said, "What are we going to do with artificial intelligence for the [banks?]?" And we came up with these tenets, because what we were trying to figure out was, when will we use AI? When won't we use AI? How will we use AI, and what are the guiding principles behind it? And we came up with four guiding principles. We came up with purposeful, impactful, transparent, and coordinated. And those were the things that we then described to the rest of the organization to say, "Look. Here's how we're going to do it. Our purpose is to help the corporation achieve our corporate goals by augmenting our employees as well." So it's not about removing employees from our business. It's about helping them do their jobs better and becoming more effective in hope of deepening the relationships that we have with our clients. 

TERRY: 02:34 

We've been working on a number of other initiatives inside the organization. One of them has been called Mylo. And Mylo is a tool that we've built that allows us to help the rest of our data scientists inside the organization know what data assets exist within our business. So we have over 120 different data sources within the organization, and it's very difficult for people to find data. And you hear people often talk about drowning in data. We literally could do that if it was possible here [laughter]. So we've got so much information that people just don't know where to go. So the way that we think of Mylo is it's like Google for data inside of our organization. You go to You then type in what you're looking for, and it will come back with data sources. It'll come back with scripts. It'll come back with people that have worked on particular projects that are associated with that particular thing that you're looking for, which enables them to be able to find that piece of information that they might not have otherwise been able to find. So those are just two of the initiatives that we've got ongoing today that are going to fundamentally change the way analytics is performed within CIBC. 

ALAN: 03:56 

Those are great examples. And so it sounds like you're fairly early on the journey. How has the business been doing in terms of adopting this digital transformation journey? I always like to think of digital transformation as two very different words. The digital part, the AI and all the fun data science stuff that you and I get to play with, but there's also that really hard word called transformation. And there's a bit of change [in?] management that frequently goes on. How is the business adapting to these new tools and this new mission that you're now on? 

TERRY: 04:33 

So the thing that we've been doing, because it's so new, we've tried to take a little bit different approach because of where we are on the adoption curve. So I like to talk to my team and my business partners about snacks not meals. So our goal is to find projects that are smaller in size, maybe 8 to 10 weeks in duration, and they might cost us internally between 100,000 and 300,000 dollars so that we can actually execute them and create this machine, if you will, that we can execute on so that we know what muscles we need to exercise as we get more complex and we get into the bigger things that we might work on a year from now. So that approach has served us extremely well. And the other part that we've attached to this execution strategy, if you will, is that everything that we do has an associated business case. So we're not out doing TOCs for the sake of doing TOCs. What we're doing is we're saying, "Look. We're going to execute this work. It's going to have this business case attached to it. It's going to run. It's going to go into production, and we're going to either grow our revenue, we're going to grow our profit, we're going to reduce our cost, or we're going to reduce the risk that we associate with that particular domain or with the organization." And getting the business to buy in when they're going to see tangible results has really allowed them to jump both feet or head-first in, if you will, with the rest of everything that we're doing. So they're super, super excited. In fact, we have over 130 different use cases that we've come up with inside the organization within the past 9 months and [we'll?] have executed on 40 of them within the next 3 months. 

ALAN: 06:16 

Oh. That's great stuff. With ROI, it sounds like, because you had business cases on them. 

TERRY: 06:21 

That's right. Every single one will have an ROI associated with it. Now that's not to say that every single one of them has been successful. In fact, there's been a few of them where we've failed. We had a hypothesis. That hypothesis didn't come true, and we failed. And when you're doing a project that's 100,000 or $300,000 [or?] 8 to 10 weeks, you can fail. If we took a different approach where we were only doing 3 projects in a year and the projects were going to be $5 million each, you can believe that there is almost no appetite within the organization to fail at those kinds of things, but because it's so much smaller and we are getting real learnings out of this, we celebrate the failure, we learn from it, and we move quickly on to the next project. 

ALAN: 07:07 

Yeah. Certainly learning to fail fast is a tenet of many [laughter] world-class, data-science teams. That's great. What kind of percentages do you think your batting average is right now? I've heard all sorts of different percentages about what percent of data-science projects result in failure or that succeed. I'm curious [crosstalk]. 

TERRY: 07:30 

So I would say, out of the 47 or so that we'll execute this year-- I know my goal was to get to 40. I would say that we would probably have failed at about 3 of them, and we learned something on one of them early enough that we were able to change the trajectory. So I wouldn't necessarily call that a failure. So just a little bit under 10%, which to me is a super good number. 

ALAN: 07:58 

Yeah. That's phenomenal. So you talk about the team that's working on creating these snacks. I love your snack-not-meals wording. How large is your data-science team? Can you talk a little bit about what the composition, the makeup, the skills of the team are and the size? 

TERRY: 08:18 

Sure. So as in most large organizations, so CIBC, they-- just about a 50,000-person organization, and there's roughly 1,000 people inside the organization that do something associated with analytics. And I use that term fairly loosely, because no one likes to be known as a report writer. So even if you're writing reports, you're still considered [inside?] analytics. So we have approximately 150 people on my team that are focused on curating data assets, they're creating insights and dashboards for the organization, or they're focused on data science and machine learning. So I'd say out of my 150 people, it's kind of a third, a third, a third across those three different areas. 

ALAN: 09:08 

Got it. And I saw recently you just posted a new position, a VP of Analytics position, where if I got it right, I think you were looking for someone, no small task, to change Canadian banking. 

TERRY: 09:23 

That's right. Yeah. 

ALAN: 09:23 

Can you tell me a little bit about that [laughter]? 

TERRY: 09:26 

Yeah. There's nothing but high expectations for this role, for sure. So I believe that banking is one of those areas that can greatly benefit from analytics. When we look at what organizations like Netflix or other organizations like that, that have so much data about their clients that they're able to create insights, people like me or whatever it happens to be, I believe as financial institutions, we have a tremendous amount of information about our clients that we should be able to leverage, not to provide them with marketing or those kinds of things, but to personalize the experiences and to deepen the relationships that we've got with our clients to make sure that we're providing them with the highest service possible to help them achieve the goals that they have. And I think that's the kind of, the thinking that we're trying to change our team to have now, and I believe if we were able to accomplish that, that we truly would change Canadian banking for sure. 

ALAN: 10:33 

That sounds amazing. So let me switch to a different level of position. So that's the VP position. I'm sure you also have data scientists that are just starting out in their career. Do you have any tips for our newly minted data scientists that are just starting out? 

TERRY: 10:56 

So we actually do quite a bit with a number of schools across Canada and some in the US, actually, and we're trying to grow those relationships, because we're trying to find those talented people while they're still in school before they hit the street looking for a job. We're actively out there looking in their programs. We're sponsoring programs. We're sponsoring analytics days where they can come and spend time with us. So I would say that look for opportunities to work with organizations in co-ops or in the summer where you can come in and learn about, not just the tools, because you can learn a lot about the tools inside programs or inside schools. But I think that what they need to be doing is learning about the business and being able to translate what the businesses are saying into what the data science languages, whatever tool that you're trying to use, Alteryx as an example of course, to be able to execute on. I think that that's a skill that you really have to learn when you get into the real world versus in school, where a lot of that work is done for you already. 

ALAN: 12:12 

Great advice. So as you built your team, can you share a bit on your views around diversity and inclusion? 

TERRY: 12:18 

Yeah. Absolutely. One of the things that I hold very high is exactly what you were talking about, diversity and inclusion. And I stood up in front of my team, I guess it was about nine months ago, and I came up with some tenets and said, "Look. I'm not going to wait for HR [and?] come by and tell me that I should be diverse and inclusive." I said, "Here's the things that I'm going to stand up in front of 400 people and say, 'Here's what I'm going to do.' I'm not going to have a HR program that's going to dictate it. I'm going to proactively go out here, because if I don't change it, then other people won't change it. So I want to make sure that we're all aligned in creating an environment that's diverse and inclusive." So some of the examples of that are I have said that all resumes, as we bring them into the organization, will be scrubbed for gender. So we won't know the gender of the people that we're interviewing with. We are also trying to minimize and simplify job postings before they go out because certain people are attracted to certain roles the way that they are written, and we have found that if we can simplify it, we have a better chance of getting a more cross section of people responding to it versus one more dominant gender or skill set. 

ALAN: 13:52 

I love that. Were there certain words that you found, or--? 

TERRY: 13:55 

Yeah. So there's certain words. There's a couple of tools on the market that you can go online and look at, and there are certain words where when certain people look at it, they look at it and it could be a more male-dominant word versus a female-dominant word. So we tried to find those words, and we tried to either remove them or neutralize them so that people didn't think that they were reading into it that we were looking for a male or a female, because we're not. I mean, we're obviously looking for the best candidate for that particular role. And there's not just one word. There's hundreds of words that we look for in the job postings that we're putting out. 

ALAN: 14:38 

Interesting. So can you tell me a little bit about your background? I saw that, I think in your time at IBM, probably working on Watson, you received a patent on, if I'm not mistaken, blockwise extraction of document metadata. I'm curious to hear a little bit more about that and kind of how you transitioned from a more technical job into this one. 

TERRY: 15:00 

Yeah. So I've had a fairly diverse background going back a couple of decades, and it didn't start out necessarily technical in the analytics space. I've worked in contact centers. I've worked in digital. I've worked in financial services and large outsourcing. So I've kind of done a number of different things, and I remember sitting with one of my leaders at IBM and she was asking me, "So what's next? You've done a lot. What do you want to do next?" And I distinctly remember telling her that I wanted to change the world. And she said, "Well Terry, here at IBM, you've got one of the best opportunities to be able to go off and do that." And that was when I was given the opportunity to lead AI and advanced analytics for IBM Global Services around the world. And I was literally working with 5,000, 6,000 people around the world, trying to help them as we were integrating and deploying solutions for IBM's clients around the world. So it kind of gave me that background that I was looking for. 

TERRY: 16:06 

As part of that, one of the opportunities that I had was with my team to create the solution-- at the time, we internally dubbed it SmartPages, where we were working on the ability, and IBM has since then made it into a full product solution, but what it allowed you to do was to scan a document with no context about what that document was, and it would then extract the relevant pieces of information from that document. So it would recognize that, "Oh. I've just scanned an invoice, and on invoices I need these 5 following or 12 following kinds of pieces of information to be able to do something with." It encapsulated it into a JSON message and then allowed a downstream process, whether it was a robot or some tool like that, to be able to go and execute-- maybe it was pay that bill or something along those lines. So that's where the blockwise extraction came into play, where we were looking for patterns around how were things like addresses or how did words relate to each other without knowing what those words were in advance of each other. 

TERRY: 17:17 

And it's really interesting. And what really comes to home is when we knew we had something that was special, that we were able to take documents that were in Italian or German or French that we were able to put in front of the machine and in front of people, and it was able to extract those pieces of information. Even though it didn't understand the words on the page, it knew that this word and that word, because of their proximity to each other, because one was caps and one was underlined or whatever it happened to be, that they were related to each other. So that's what that particular patent was for. And there were, I believe, six other patents as part of that product that we created as well. 

ALAN: 17:59 

Yeah. I love that moment where you see it actually come to life and work. We frequently call that the thrill of solving, and it sounds like you certainly had that moment in that example. 

TERRY: 18:11 

Oh. Absolutely. It's a great feeling to look at this thing that you've created, and it's not that it's thinking on its own, but it almost looks or it seems like it is. It was fantastic. 

ALAN: 18:22 

So when you look out over the horizon, are there any kind of new technologies that you think will be the next big thing, whether it's for banking or AI in general? 

TERRY: 18:31 

I think that there's a lot of technologies that we already have today that we haven't grasped the entire functionality of inside of our own organizations. So I'm sure that there will be new technologies that come over the next little while, but I think that we still haven't gotten our hands around what we need to be doing with what we've got, whether it-- 

ALAN: 18:53 

Yeah. I totally agree. 

TERRY: 18:53 

--whether it's natural language processing or visual recognition. I think that those environments or those capabilities will continue to expand, but we're only scraping the surface as it relates to those today. I do think that there will be transformational things that come along in the future, and the ones that are going to make the biggest impact are the ones that are tailored specifically for a particular industry. So we've seen large vendors, whether it's IBM, Microsoft, Google, etc., create solutions that are trying to be generic that can cut across domains, and I think that some of the adoption of technology that we've seen related to artificial intelligence or machine learning haven't taken off as quickly as they could have had those organizations built tools or products that solved a specific problem in banking or solved a particular problem in credit cards or something along those lines. I think when we start seeing more organizations coming out with those purpose-built tools, we're going to see an exponential growth in the capabilities in artificial intelligence and advanced analytics as well. 

ALAN: 20:03 

Yeah. I couldn't agree more. I think we see that from our vantage point as well. So when you think about kind of your day, are there any things that you do during the day, either at work or outside of work, that you view as being some of the keys to your success? 

TERRY: 20:22 

I think [laughter] that this is a good question. I think that having some sense of reality is important or perspective is important, I think. I know some people talk about-- the term work-life balance has kind of gone away, but I've always [strived?] to have some sort of work-life balance throughout my entire career. Coaching my kids in a number of different sports as they were growing up. And I think that taking time off and disconnecting is important to be able to get that new breath of fresh air, if you will, or reinvigorate yourself to come up with the next great idea. So I think that the way that I do that today [inaudible] is by getting on my bicycle and riding for 50 km or 100 km, something like that. 

ALAN: 21:13 

100 km? 

TERRY: 21:14 

Oh. Absolutely. I mean, I just got back from-- 

ALAN: 21:16 

Oh. My. 

TERRY: 21:17 

I just got back from France three or four weeks ago, where I rode 700 km in a day as part of a charity ride for veterans with post-traumatic stress disorder. So that's really how I clear my head. 

ALAN: 21:30 

Oh. That's amazing. 

TERRY: 21:31 

You get on a bike. You head out into the wilds, so to speak, and the ideas, they just start flowing and it's such a liberating feeling. 

ALAN: 21:39 

Okay. And you didn't wire yourself up with GPS and spend the next week doing analytics on it? 

TERRY: 21:45 

Well, I do all that too [laughter]. So I've got a computer that I attach to my bike. So don't worry. I've got all that covered [laughter]. 

ALAN: 21:50 

Oh. That's great. So we like to have a bit of time for what we call Community Picks. And so we'd love to know, and I think our listeners would love to learn of anything that you're doing or enjoy doing in kind of the technology, nonprofit, leadership space. It could be books, anything. 

TERRY: 22:11 

Yeah. Absolutely. So as I mentioned, bike riding is a big one for me. We just got back from France where we were commemorating the 75th anniversary of D-Day. So that was a big one for me from a riding perspective. I sit on the board of a couple of nonprofits, one called Renascent, which deals with people who have addiction problems, whether it's alcohol or drug. There's a bit problem around the world, as you know, [on?] those kinds of things. And then we're involved in, from a CIBC perspective, in stopping human trafficking. Human trafficking is a major problem that occurs around the world, and we all assume that it's someone from overseas that gets brought to our country and indentured into the sex trade. In reality, most of the people that are put into this are 13- to 16-year-old girls that come from every walk of life across our country. They could be daughters of lawyers or policemen or policewomen. They could be bankers. It is nondiscriminatory, and we're working a lot with the police and other organizations to try and stop human trafficking because of the impact it's having on the general public [at?] whole. 

ALAN: 23:36 

Yeah. What a great cause to support. 

TERRY: 23:38 

Yeah. Absolutely. Absolutely. 

ALAN: 23:41 

So I'd be curious if people wanted to get involved in one of those orgs, where might they find out more about it? 

TERRY: 23:50 

So all of them have an associated website. You can go to You can go to, and for human trafficking, you can look at Covenant House. There's one other one that we're working on from a CIBC perspective, and we've kind of coined it Data for Good. And I know there are organizations out there that have a similar kind of title, but that's where the people on my team and across the CIBC organization are looking for organizations where we can go and help them. I say that my team - and this is not just my team. It's obviously people with the analytic skills that are listening to our podcast today - that I believe that our teams have super powers. They are the modern-day super heroes that can fundamentally make a difference for organizations, and I want my team and the people on their teams, I want them to go and make a difference for other not-for-profits. So what we're doing is we're actively going around looking for organizations that are looking for people that have the analytical skills. They've got all this data, but they don't know what to do with it. They don't know who their donors are. They don't know where they live. They don't know how to get a hold of them. They don't know all these kinds of things. And if we can come in with a team of four or five people and sit with them for an afternoon and give them these insights that are going to help them make a difference for whatever cause that they're doing, we think that that's absolutely well worth it. And that's one of the big things that we've got our team working on this year right now. 

ALAN: 25:26 

Yeah. Certainly analytics can be fun, but analytics with purpose is [on?] a totally different level. That's great to hear. 

TERRY: 25:32 

That's right. 

ALAN: 25:32 

Great to hear you and your team are focused on that. So we'll end with something maybe a little bit lighter. Can you tell us something that maybe very few people, or maybe nobody, knows about you? 

TERRY: 25:44 

I would say that one of the things that most people don't know is that I grew up all over the world. So I lived in Venezuela. I've lived in a couple of different parts of Spain. I've lived in a couple of different parts in Switzerland. I lived in the UK for bit. So as a child, I had never lived in the same house-- actually, up until the time I got married, I had never lived in the same house for more than two years from the time I'm born. So we were constantly on the move, and no, my parents weren't hunted by the police or anything like that. My father worked for General Motors at the time, and we just ended up going to wherever they needed him the most somewhere in the world. 

ALAN: 26:23 

So if you could take a vacation anywhere in the world for your next vacation, where would it be? 

TERRY: 26:29 

So our favorite place on the planet is Hawaii, and we've been there quite a bit, but the one thing that I haven't done that I really, really want to do is go on a safari in Kenya. That would be the one. 

ALAN: 26:44 

Oh. I'm there with you. I've got a little one that loves giraffes. I can't imagine if he was surrounded by giraffes [laughter]. So how old did you say your kids were? 

TERRY: 26:53 

My son is 20, and he's studying actuarial science. And my daughter is 18, and she is going into first-year data science this year [laughter]. 

ALAN: 27:02 

So I was going to ask you if they're doing anything with analytics [laughter]. I think you've nailed it on both counts. 

TERRY: 27:08 

That's right. That's right. 

ALAN: 27:10 

That's awesome. Well, hey. Terry, thank you very much for spending time with us today. It was great getting to know you, and I look forward to hopefully spending more time with you in the future. 

TERRY: 27:18 

Thank you so much. I really appreciate it. Thanks, Alan. [music] 

MADDIE: 27:24 

Thanks for tuning in to Alter Everything. Continue the fun and share your thoughts on Twitter using hashtag #altereverythingpodcast or leave us a review on your favorite podcast app. You can also subscribe on the Alteryx community at And hey. While you're there, go ahead and fill out our audience engagement survey. The first 100 people to leave their feedback will be entered to win one of five pairs of Bluetooth headphones. You can also join us in-person at Inspire London this October. Use the promo code INSPIREPODCAST, all one word, for 15% off your Inspire registration. Hope to see you there. 

This episode of Alter Everything was produced by Maddie Johannsen (@MaddieJ).