Data and Leadership with Nancy Zhu

Val Buresch (00:01.526)

Welcome to CMBConnect. I'm your host, Val Buresch, and in this episode, it is a pure pleasure to be joined by Nancy Zhu. Nancy is the SVP and Director of Financial Reporting and Analysis at Cornerstone Home Lending, where she oversees financial reporting, analysis, budgeting, performance measurement, and all other internal and external reporting.

So thank you for joining us, Nancy.

Nancy Zhu (00:34.066)

Thank you, Val. It's pleasure to be here.

Val Buresch (00:37.682)

When we were preparing for this conversation, I was impressed by your formidable background in accounting and finance, evidenced by your certifications like CPA and CMA. What was the catalyst that brought you into the mortgage banking industry six years ago?

Nancy Zhu (00:58.12)

Well, honestly, so six years ago, I was working for a consulting company. I was a consulting manager at the time, helping my clients, mostly CFOs and the controllers to automate and streamline their reporting the budgeting processes. So I consider myself having a good accounting background because I have all the education from accounting and finance.

And then I was having exposure in the technology side. So I was familiar with all those ERP systems and reporting, we call BI tool, business intelligence tools. And I was able to combine reporting tools and which is connected to the ERP systems. And I understand the data and the data flow. And I also know how accounting was done, how data was being utilized. So I was able to help my clients, mostly CFOs and controllers.

to streamline their processes. However, my accounting knowledge and I feel like those skill sets are kind of transferable. it's meaning it's spanning across different industries. So I wasn't really looking for a specific industry because my past experiences is always covering all kinds of industries. Now, the job over at my consulting company become later on because it requires a lot of traveling and

my family duties kind of prevent me from traveling all the time. So I was thinking about switching to the next job, which is I wanted to work inside a corporate setting and the industry, wasn't honest, I wasn't really seeking out a particular industry to come to because as I said, my skill sets are kind of transferable across industries. So I was interviewing, I remember I was interviewing for three jobs and I received three offers from the three jobs. And one offer was from a manufacturing company. The other one was from a waste management company. And the third one, which is the last one, was from a mortgage industry. And the reason why I picked mortgage industry was because the people influenced my decision. The first...

Nancy Zhu (03:16.625)

The that really influenced my decision was the CFO at that time. His name was John. He has retired, by the way, and he really liked my skill sets. he told me that I checked off pretty much all the boxes. He has been looking for a person like me for some time, and he was very happy to see me. And as a matter of fact, he moved the whole process so fast.

in the interview process. the beginning, when I received a phone call from the recruiter, until I received the offer, the whole process took about a week. While other jobs, I was talking to different people, the CFOs, the VP finance, they were taking time. But this job, corner store home landing, the CFO job, he pretty much pushed the whole process within a week. So I can tell he really wanted me to help him to

Val Buresch (03:52.182)

Wow.

Nancy Zhu (04:12.419)

streamline the reporting processes. I remember he told me that. So John, is very, he's a very data driven person. He likes to read data. He has an acumen to, you know, to the data. But at the time, the process is kind of manual. So it's pretty much Excel driven spreadsheets, using data dumps, the formulas, sometimes Excel formula is good, but it cannot.

take care of everything. you know, you have to do a manual key in into the reports. And the analysts were stressed out because it's a lot of work, a lot of reports, a lot of demands. And I remember John told me that he commented that Nancy, I wasn't even sure the data that was given to me was accurate or not. So at that time, Cornerstone was in the middle of

think they just implemented an ERP system. So they had a new accounting slash ERP system and they purchased a new software for reporting purposes. So they were looking for revamping the whole reporting process for the company by using a new system and a new BI tool. So he made a decision. He was so decisive. He pushed everything so quickly and didn't even bargain or negotiate with.

Val Buresch (05:24.451)

Mm-hmm.

Nancy Zhu (05:35.657)

whatever I was requesting. And that was the first person who really influenced my decision to come to Cornerstone Home Landing. And the second person that influenced my decision was actually the receptionist at Cornerstone Home Landing. Her name was Terry, and she has also retired. She was so warm and so nice and so welcoming. I remember I came to Cornerstone for interview and I was nervous because I'd never been here, you know, and as an interviewee.

naturally we are nervous. And she greeted me with such a smile and such a non-judgmental way. I instantaneously felt calm. And I feel Terry, the receptionist, even if no matter how significant her role is, I think she is the face of the company. If the face of a company is such a warm and kind person,

I just think that this company will be worthwhile to come to work for and make contributions for. And that's the reason why I came to Kuala Lumpur-Ledin. I wasn't picking the industry at that time, but it turned out that it's such a nice decision and I never regret a day that I came here, came to the mortgage industry. Now, when I started working for mortgage industry,

I got fascinated. I did not realize how fun it was because of the complexities and the dynamics of the industry. So before, like I mentioned, I know I have strong accounting background and I'm pretty strong in technology. I'm pretty savvy to...

identify which database to go for what information and put them together and do it quickly streamlines the process. So it's all it all makes makes everything easy and fast. So that way we can spend more time analyzing the numbers instead of gathering them like before, know, the status quo before. So I was I was very, very enthusiastic about the

Nancy Zhu (07:41.66)

the job I was doing because I knew I'm putting into use all my skill sets in accounting, in technology, in using tools and in how I understand the databases and the structures and the schemas and the, you know, the fields and how we link the fields together, the keys. So I was very happy enough. And then after I came over into the mortgage industry, I realized that, you know, I have a finance background. I have a master of science in finance.

And after graduate from my degree, I was mostly working accounting and auditing and financial analysis. I wasn't really putting my finance knowledge into use. as you know, finance is all about investment, capital market, primary market, secondary markets, bound pricing and hedging and derivatives. But you don't use that in your day to day accounting job. So after I came over to

to the market industry, realized that, a second, I have another piece of puzzle that I can utilize it. you know, honestly, the finance background helped me greatly in how I understand secondary market, how capital market was doing, how they were performing the hedging, and it helped me greatly in how understanding how the data, you know, the reports, the hedging reports coming from the hedging system.

So it's really, really instrumental in helping me to understand the number because eventually all the data will translate into financial numbers because we need to book accounting entries and you need to able to analyze the numbers. Although I'm not an expert, I'm not working in the capital market, but really it helped me to understand what they do and helped me to bridge the gaps between accounting and the capital market.

Val Buresch (09:18.925)

Yes.

Val Buresch (09:30.755)

Yeah.

Val Buresch (09:34.828)

Yes, you were able to speak their language, basically with your knowledge and that fusion of skills and background and understanding. Yeah. Yeah. What, you know, what leaped leaped leapt at me as you were talking is, is the fact that you were hired and, just going back to your story, how you started in cornerstone home lending, at a time when

Nancy Zhu (09:37.561)

Right, somewhat.

Nancy Zhu (09:43.375)

Right, right.

Val Buresch (10:02.466)

That company was going through technology transformation, seems, with the application of this new system inside the lender. But what they realized is that the human element cannot be replaced. So they were looking for someone like you who can speak.

data and technology and also who can relate to other human beings in the company and make that connection, make the adoption of that technology faster. Would you agree that that's what was happening, that you were in the right place in the right time at that time?

Nancy Zhu (10:34.918)

Bye.

Nancy Zhu (10:39.978)

yeah, absolutely, absolutely.

Val Buresch (10:43.348)

Yeah, well, I wanted to ask you just to elaborate a little bit more about your fascination by the complexity of the mortgage industry. Can you pinpoint a specific moment or perhaps a concept that you made you realize this was the career for you apart from the human element and that wonderful story about Terry and John?

Nancy Zhu (11:11.047)

Right, yeah, mean, so the so more industry has, you know, the whole lifecycle starts with production, right, it has a production process, which is already very dynamic and complex. And then later on, once the loans are closed, the funded, it goes through a most IEBs will go through a selling process, selling into the secondary marketing. And it comes with

hedging and from accounting perspective, because this is the lens that I'm looking at the whole thing, from accounting perspective, we recognize fair value. So we do a lot of fair value recognitions in the accounting world. So accounting event is when the loans are locked, technically, it's an accounting event, we need to put a value to the lock, which most people don't probably they're not aware of that because

Val Buresch (11:51.51)

Yes.

Nancy Zhu (12:06.795)

once we commit to the borrower and provide a relock to a borrower, we have a commitment. So we need to recognize that commitment because it's when the interest rates changes with the market, we still need to fulfill that commitment. So if the interest rates turns to the other way, then it's going to be a potential loss for the company. So it's an accounting event that we need to recognize the fair value for the interest rate law.

the time of lock. So that's a new concept. And it's, you know, the whole process of calculating the value for the lock pipeline is itself is a very, complex calculation, because it's all based on assumptions and based on past historical assumptions and data. And you, you know, you look at the fear of how much those potential locks is worth in the market. And you also look at the whole

Val Buresch (12:47.724)

Mm-hmm.

Nancy Zhu (13:04.741)

What is the cost that will help to accomplish the process from lock to the end of the production cycle? So we have to calculate the estimated cost component to it. So that process is complex. And then after that, when the lock turn into the actual loan, it's a completed loan and it's waiting to be sold into a secondary market. So that completed loan.

Val Buresch (13:21.027)

Yeah.

Nancy Zhu (13:31.398)

we call warehouse loans, we also need to put a fair value to the warehouse loans. the warehouse loan fair value calculation is somewhere, it's somewhere similar to the interest rate log cut calculation, but it's still different because the warehouse loans, a lot of the cost components are already accomplished. So whatever the cost that put into the warehouse, it's pretty much done. Like the discounts we have already.

already received all the appraisal expenses it's already realized. So the warehouse fair value has a lower component or less component of costs compared with the lock pipeline. The lock pipeline is really sitting in the pipeline so it's waiting to invest in a lot of the costs and expenses associated with the loan production process. So that's a second piece of complexity to put a fair value for the warehouse loans.

And then the complexity just continues on. Later on, the loans are waiting to be sold into the secondary market. And the hedging piece, because every month we need to recognize the hedging gain or loss. That's mostly for the pipeline. And the hedging piece is another complex piece because believe me, a lot of the accountants, have no idea what hedging is. What is the...

the scale of hedging, what are we doing hedging for? And let alone you need to understand those reports that's mostly coming from the capital market and translate them into accounting numbers. So that part to me is also a very complex piece. So, and move on to the cycle and then later on.

Val Buresch (15:00.291)

Mm-hmm.

Nancy Zhu (15:19.151)

Suppose we sold the loans into the market to the investor, but our responsibility does not stop here. So we still need to, a lot of times like Hulustone, we retain a lot of the loans, even if we sold the loans to Fannie, Jeannie, and Freddie. So we have a servicing portfolio. And according to GAAP we need to put a value for the MSR, servicing rights, and the whole fair value for the mortgage servicing rights.

Val Buresch (15:42.442)

Right. Yep.

Nancy Zhu (15:46.807)

is a very, very complex process. So as you see from the...

Val Buresch (15:50.602)

It depends a lot also driven by CPR, right? The conditional prepayment rate. Yes.

Nancy Zhu (15:53.979)

Right, CPR is a lot of assumptions. And although I don't calculate the value, but I know because from the finance perspective, it's all discount cash flow model. So we are projecting the future. What is the future cash flow looks like and take away what is the future costs and based on assumptions, how much should be realized and then discount back by utilizing a type of discount rate. And then you calculate a fair value. So that process is complex, even if I'm not

Val Buresch (16:07.341)

Mm-hmm

Nancy Zhu (16:23.836)

calculate the fair value but we need to understand their reports. What is being put into the assumptions? Are the assumptions reasonable or not? Why do they change their assumptions? So it's a lot of deep topics involved in getting those calculations.

Val Buresch (16:37.699)

Yeah.

Yes, yeah. And you're one of the few people who understand that. mean, most, even in mortgage banking, most people go day in and day out without thinking about those issues. So thank you for lifting a little bit the cover and allowing us to see the complexity of mortgage banking from that perspective, especially the accounting perspective.

Nancy Zhu (16:53.266)

Right.

Nancy Zhu (16:57.608)

Yeah.

Nancy Zhu (17:04.838)

Yeah, basically I was lucky enough because I'm dealing with all the reports and I was tracking regional performance and I was looking at the consolidation, I was analyzing different aspects of the financial reports and I was also in charge of publishing the servicing packet for the company on a monthly basis. So I was able to slowly.

Val Buresch (17:05.783)

Yeah.

Val Buresch (17:21.773)

Mm-hmm.

Nancy Zhu (17:30.736)

a peek into different aspects of the companies. You know, from the production side, because I don't work in production, so I don't have a lot of the knowledge on the front end of the business. But we were also tracking some of the key KPI metrics, cycle time, pull through, you know, and underwriter touches.

A lot of times if we cannot get to the numbers, we will rely on people who are subject matter experts to get the numbers, but we do collect all those information and put it into a nice packet and we're comparing across different areas and regions or we are looking at by channels to see which channel is more profitable and which channel is less and the...

how we allocate the costs, things like that. So it's a lot of fun. I mean, it's very complex and took me a while to really understand the different bits of the industry, of the company. But you know, think really the complexity, the more complex they are, the more motivated I am to debunk the myth.

Val Buresch (18:44.95)

That's wonderful. Well, there's also one more thing that's very interesting about you. And that's also fascinated me as being a first generation American. But you particularly, you are interesting because you've lived in four different countries. So can you tell us a little bit about that? And also how has that global experience shaped your approach to business and leadership in such a industry like the US mortgage?

Nancy Zhu (19:02.074)

Right, right. Yes.

Val Buresch (19:14.528)

market that is location based.

Nancy Zhu (19:17.468)

Why? Right. So I was born and raised in China and I came to the US about over 20 years ago. I came to the US, but after I arrived, I also went to Australia and I was there for about two years. And after Australia, I went to Indonesia and I was there for about four years. And then we moved back to the US. Houston is where I'm currently living. So every

country I've lived has its uniqueness. Sometimes they can be drastically different. For example, Australia and Indonesia, where I lived two and four years respectively, they are completely different countries. Australia was it's located in the southern hemisphere. So as you can imagine, we have different seasons, right? Their summer is our winter. So we had

Christmas in Australia during hot summer days. We were wearing shorts and they drive on the left side of the road. So it took me a while to get used to that because I'm always worried about being hit. And the weather was awesome. It's very sunny. It's sunny all year round. In the winter, it's sunny and windy, but in the summer, it's sunny and hot.

UB Ray was really really strong and everybody has to put on sunblock and supposedly they have a lot of high rate of skin cancers so statistically by age 70 70 % of Australians get skin cancer so the reason why is that that's what I heard was they have a missing ozone or they have a thinning ozone layer so and also the

Val Buresch (21:12.197)

wow.

Nancy Zhu (21:13.798)

the air was clear so the UV ray just hit directly to the surface. So therefore, if you live there, you know, in schools, everywhere, in public, you know, it's strongly advised that people need to wear sunblock everywhere you go. And in addition to that, it's multiple layers of protection, including wearing a hat, wearing put on sleeves and stay under the shade and, you know, don't forget put sunblock.

Val Buresch (21:37.07)

Hmm.

Nancy Zhu (21:43.945)

So it's all a unique experience to me. The food was awesome, weather was good, a lot of beaches and they spend their leisure time on beaches most of time. And it's just a phenomenon place that they have a lot of fresh produce and everybody's independent. So meaning if you want something to be fixed, you have to do it yourself.

Otherwise you have to pay a lot of expenses to hire somebody to do it for you and everything is expensive of course. Indonesia is the extreme opposite. It's a developing country. It's always located in close to tropical areas. So it's always hot and humid and they have two seasons. They have rainy season and a non-rainy season but it's hot and humid all year around.

So when when we go outside we don't instead putting sunblock we put mosquito repellent Because it's very humid. It's very easy to get mosquito bites The places that we hang out in Indonesia is not anything like a beach or outdoor Unlike Australia we go to the malls and that's pretty much where we go for leisure time and and and the lifestyle is fairly easy because it's

Val Buresch (22:47.758)

Yes.

Nancy Zhu (23:07.836)

The labor there was very affordable. So average middle class families, they can afford one or two mates. And when we were living there, we even had drivers. So because we were advised not to drive because the traffic was terrible. It's always congested, lot of motorcycles on the road and a lot of madness on the road. So we would...

Val Buresch (23:26.573)

Yeah.

Nancy Zhu (23:37.171)

told not to drive, so we had a driver to drive us around. However, the place that we live is called Jakarta, the capital city of Indonesia. It has a lot of experts living there, so food-wise, their food was very spicy, but they have lot of varieties of different kinds of international food, especially Japanese food, Korean food, American food.

and all kind of other kind of foods. It's just so, so abundant and you'll be surprised. So yeah, mean, it's just different life. so back in China, I've always been the good student. was always the top. So we kind of ranking everybody in school all the time. And I was always the top ranked student in the school. was ranked number one for

Val Buresch (24:13.528)

Sounds wonderful.

Nancy Zhu (24:35.506)

for six consecutive years in my middle school and high school. So I was always studying, What I can remember when I was living there growing up was I did not remember anything else. I was studying and I was a good student. I was a top student. And I graduated from one of the top universities in China. after I came to the US 20 years ago, I realized that

I really had nothing. mean, the language, they spoke English so fast, even if I have been learning English back in my hometown. But the English I learned was different from the English that they spoke. And I could not even follow what they're saying 20 years ago. And of course, my education did not count. Even if the university I came from, I graduated from, was one of the top five universities in China, but nobody knew about this.

Val Buresch (25:21.234)

Hahaha

Nancy Zhu (25:33.673)

And my experience working for Ernst & Young in the Shanghai office, I was an auditor there. But because we had different gap, accounting gap, the US gap is unique in the world. It's different from European gap. It's different from China gap. It's different from Australia gap. But their gaps are somewhat similar. They are all kind of following the European or international standards. But US gap is totally different. So what I did,

Val Buresch (25:49.996)

Right?

Nancy Zhu (26:02.92)

My working experience in Erciyán did not count. My education did not count. My English was terrible. So when I came to the US, I had a complete reset. I had to start from zero. So I remember I went to UH, University of Houston for my master degree in finance. And I was struggling in understanding the professor. I mean, I could understand about 80%. By 20%, I wasn't.

fully digest those and I was so worried that I, whether I could catch up with my studies. But I tried, I mean I tried very hard. So I graduated with a 4.0 GPA from U of H, which I'm very proud of. So yeah, every country is different, it's unique. So one thing I learned is that,

Val Buresch (26:50.53)

That's amazing.

Nancy Zhu (26:56.392)

I've seen a lot, this global experience of living in different countries and every country, their culture are different, their lifestyles are different, their weathers are different. But one thing in common, in my opinion, is the people. Everybody is living their life. Everybody is honestly earning their lives. are earning their living. Everybody has their troubles, their struggles, and their vulnerabilities.

Val Buresch (27:22.36)

Mm.

Nancy Zhu (27:24.038)

So when I now when I see people, I don't see the differences. I see what's common because, you know, we all wanted to be respected. We all wanted to have a life, have a purpose. We all wanted to work hard for our next generation. We all want to protect our children. No matter what background they're coming from, they all have they all want to have a good life. They all wanted to improve themselves. They all want to be good people.

So that's what I see when I look at people from different backgrounds.

Val Buresch (27:55.532)

This is so powerful, Nancy. This is an amazing story and relating it to that global experience to the US mortgage market. I couldn't agree more with you. I had similar experience being at top, top, top in your own country of birth and then coming here starting from zero. Nobody cares where you come from and what you have learned, what university you have graduated, how prestigious it is.

Nancy Zhu (27:57.801)

Thank you.

Nancy Zhu (28:13.511)

Right.

Val Buresch (28:24.862)

you have to reset when you come to the US. what a... No, I just wanted to say what a powerful perspective you bring of that mutual respect for other people and for acknowledging that everybody's trying to do their best and just seeing that and bringing that perspective with you to our market is incredibly powerful.

Nancy Zhu (28:27.398)

Right. And I'm like, yeah, sorry, go ahead.

Nancy Zhu (28:50.512)

Right, and I just want to add is I realized that no place is perfect. There's no paradise. Because a lot of people ask me, know, out of all the countries that you have lived, what country do you like the best? And I always say that the country where I'm currently living is the best. Because in my heart, I think there's no paradise. Just like the people, we are not perfect. Nobody is perfect. No place is perfect.

But what we can do is how we can make use of what we have and turn them into the best use. So I enjoy every place that I have lived. I like the people around me where I live. And I just wanted to make the best use of what I have and turn it into something that's meaningful and helpful.

Val Buresch (29:42.188)

Yes, that's beautiful. Really beautiful. Well, I want to pivot a little bit back to the power of data and automation because it's where my heart is and where I think our industry should be because of all the data we have. But specifically, I wanted to ask you about your experience. You are...

an accountant that's financial reporting, this is where you live. So reporting is the keyword there. But you mentioned to me that you are focusing on data-driven insights. You're using automation to focus on high value thinking. That's where human brain is much better, is when you can look at patterns and begin to synthesize data. So with that said, I wanted to...

to ask you to walk us through an example of a process, and a process that you automated, and the strategic value that was unlocked as a result of that automation.

Nancy Zhu (30:47.657)

Yes, so six years ago I came to came to cornerstone and I as I mentioned at that time the process was pretty manual a lot of reports a lot of them are in Excel and it driven by Excel formulas So I came over I started to revamp the reporting process So what I do is, you know, I have I have the knowledge about the software the BI reporting tool that we have

and I also have the knowledge about the ERP system, the cross modular information in the new ERP system. So I basically redesigned the whole process by using the software we call Jet Reports. So I use Jet Reports, which is connected to the ERP system, and I grab the information from different pieces, and I write logic in those reports. So now...

Val Buresch (31:39.214)

Mm-hmm.

Nancy Zhu (31:41.894)

It took, of course, it sounds easy, but it took me pretty much a full year to revamp all the reporting process, redesign all the logics from scratch, including all the affiliate companies. We had about 15, 16 affiliate companies, and every company has a reporting package. We also have regional reports, regional reporting packets, and we also have management, the executive team packets, and we just have different kinds of packets.

Val Buresch (31:51.584)

Mm-hmm.

Nancy Zhu (32:11.689)

pretty much redesigned all those reports by utilizing the software. And sometimes some information cannot be automated because it could come from a a flat Excel file that's provided by an external company. In that case, you know, the only way that we can do is using Excel formulas, but we can still do something about it because we can, you know,

designing an input area. reporting wise, I'm just speaking the reporting perspective. Reporting wise, in order for the reports to be automated and streamlined, we need to be smart about it. So usually there is an input area. If you need to provide input, you want to collect the input area in one central area. And then you use, you you flow the input area into the reporting, into the reports. So that way, if one of the inputs changes, you don't need to go back.

Val Buresch (32:46.765)

Yes.

Nancy Zhu (33:11.187)

to each individual report to change that change. So you come to the input area where you have a central repertoire of all those information that coming from somewhere else, you have that centralized location, you just change the inputs directly, and then everything else will flow into the reports. So that's the kind of design in the back end of all those packages, right? Because we want to be streamlined, we want to be efficient. So that way,

We have to think about, what if things change? How we maintain those reports? How we make changes when things change? How we make changes as fast as possible? know, ideally we want to have maintenance free reports. So that way all you need to do is just click the button, say, okay, run this for this period, run that for, you know, rolling 12 months, run that for a year today, and then everything will take care of.

Val Buresch (33:53.506)

Mm-hmm.

Val Buresch (33:57.966)

you

Nancy Zhu (34:06.919)

But sometimes things may happen that may make it impossible to fully automate a certain reporting process. And that way, the only thing that makes it better is to make sure that there are sections. There's a section that's just for input. There are sections just for assumptions. And there's a section for the reports. So that way, when any of the element changes, it will flow into the reports. So I kind of redesigned.

all those reportings either automated fully or partially by leveraging different levels of automations. And then the other thing that I did after I came over is that, you know, in the multi-industry, we're not good at budgeting because things change too fast and we cannot predict too far into the future. So we didn't have a formal budgeting process when I came over.

Val Buresch (34:57.738)

Right. Yeah.

Nancy Zhu (35:05.321)

Sometimes, you know, some of the warehouse banks may require some budgeting for the next year. And we may do that on a high level, just based on high level assumptions. And we will come up with a top-down approach budget. And then we send to the banks and the bank says, okay, then we are good. We move on. We never did a bottoms up, detailed budgeting process. So I came over to Cornerstone and we...

You know, the CFO at that time wanted us to go through a bottoms up budgeting process. And that's just an experimenting because we never done that before. And we went through processes of picking the software. And then eventually I recommend to use the inherent budgeting module inside the ERP system because it's free. And one that's free and two it's also because it has enough functionality. And usually

you know, the innate budgeting modules inside an ERP system. It does certain functions, but it's not robust. And that's why a lot of people go to outside vendors looking for a more robust budgeting process or software. But for us, I think, you know, the inherent budgeting module inside an ERP system just works just fine. And so we went towards that route and I began to design budgeting modules because those modules are like

It's like a template, right? In Excel, we utilize software and put all those information. What's a historical number looks like? And then in the middle, it's all the assumptions. What's driving those numbers, right? It's either by per loan basis or volume driven or sometimes it's a fixed income or fixed expenses or sometimes it's a variable expenses that varies with the production volume. So it has a assumptions in the middle and then on the right.

the budgeting model has three sections, right? The historical numbers shows what happened in the past. The middle section is the assumption piece, it shows what's driving the numbers, what's driving historical numbers. And then the right section is the budgeting for the next year. So that way, the budgeting is designed in such a way that it's kind of makes everything easier to budget into the future because it has all those pieces of information in one page.

Nancy Zhu (37:29.137)

So I designed all those budgeting templates and hand out to all the branches and regional contacts and all the ABAs and all the departments. And those are already done in a pretty automated way and has a lot of good information there. And then hand over to them. They will work on their budgeting individually and then hand it back to us. And then once we receive the budget, we upload it into the system, which is the ERP system that I just mentioned. has an inherent

budgeting module and I performed the whole process with fairly, you know, least amount of work involved because you know, budgeting, the bottoms up budgeting is a big endeavor for a lot of companies and a lot of companies, they, they probably it's taking about at least three months to finish the whole process. And it requires iteration sometimes because when the first initial draft, it's maybe not

Val Buresch (38:21.69)

yes.

Nancy Zhu (38:25.202)

good enough and then we, it's too high or too low. And then we maybe go back and renegotiate. We re-discuss and then come back again and then we submit the information to system. And then after that, we just pull the numbers using softwares and then compare actuals with budget and how we are doing. So that process, I designed that process for Cornerstone and we had a first ever bottoms up budgeting here.

so that they use other. Thank you.

Val Buresch (38:56.096)

Amazing. Yeah, no, amazing. And that makes such a big difference. What you've done is incredible. I truly appreciate it since we use a lot of data and I know how heavy that data pipeline can get and lenders, regulators report, make changes to those accounts and things change and you have to be on top of that. If you don't have a good

process of how to update your automation. Yeah. We use a lot of data engineering processes to get the data, then move it to our data warehouse. And even that process has to be automated. Yeah. Yeah.

Nancy Zhu (39:40.967)

Right, right. Yeah. So what you're dealing with is probably, so what I deal with is mostly financial data, but what you're dealing with is a lot of the industry data, right? So the data are different, but I believe the process wise, I mean, you your process is more, way more robust, you know, because it requires a lot of ETL, transform the data to be usable. That part is a lot, it's quite technical.

Val Buresch (39:50.701)

Yeah.

Val Buresch (39:56.107)

It's

Val Buresch (40:01.61)

It does. Yeah. Yeah, that's the key word there. Making the data analytics ready. If you want to use it to derive insights or use it for anything, it has to be analytics ready, which requires all this cleaning and putting it in a format that can be picked up by system, by BI system, financial systems, and things like that.

Nancy Zhu (40:25.168)

Right, yeah, absolutely.

Val Buresch (40:28.578)

Yeah, so as the senior vice president of financial reporting and an accountant in mortgage banking industry, what are the top two or three KPIs that every mortgage lender should be watching in today's market? Obsessively watching, actually. That's what I wanted to say. That's really the most important KPIs.

Nancy Zhu (40:50.097)

right.

Nancy Zhu (40:53.916)

Well, think so, of course, revenue, right? Revenue and related KPIs like gain on sale in terms of bips, gain on sale, the actual amount, The gain on sale mostly because the gain on sale is the biggest component of the revenue. Okay.

Val Buresch (41:11.318)

Let me stop you for a second here because I think and I know that some people who are listening to the CMBConnect podcast are not mortgage bankers. Explain what you mean by gain on sale.

Nancy Zhu (41:25.478)

Gain on sale is mostly, so the mortgage bankers mostly they originally loans using funds that came from mostly warehouse banks. They have a warehouse line of credit to borrow funds. And then they use the funds to fund the loans that borrowers requested. And then

they don't, mostly they don't hold on to the loans because a lot of typical depository banks, will hold on the loans, they will service the loans, they will collect principal and interest. And then that's the depository model. But for independent mortgage bankers, we don't hold on to the loans. We sell them into secondary marketing. And then when we sell into the market, we receive the money from the investors. And then we turn around,

we pay off to the warehouse bank where we owe them because we use their money to fund the loans from the first place. So the difference between what I sell into a secondary market to what I use the money from the warehouse banks, that difference is the profit that an IMB makes. And that's a major part of the gain on sale. It's really the premium, the difference between what we sell for

That's what we use the money for, for funding the loans. And there are other components of a gain on sale. Like sometimes we will charge a processing fees. Sometimes we pass through some of the expenses like credit report fees that we may collect from the customer, but we turn around. We also pay to the credit bureau the expenses. So these expenses are mostly passed through in nature.

But sometimes we may have a difference like appraisal fee for example, we may collect from customer a higher amount or maybe you can pass around, know, I mean things like that. So there may be a premium in that process. So those are all part of the gain on sale and a direct net of direct costs.

Val Buresch (43:32.12)

Right, right.

Val Buresch (43:41.954)

Yeah, thank you. revenue, number one, gain on sale, two.

Nancy Zhu (43:45.929)

Right, so revenue to get on sale to in my opinion is it's just we can think about it in one bucket because they are the the the You know the driver right that the driver of the revenue is mostly it's gain on sale so that's to me the first ratio the second ratio is we need to look at expenses and Especially the biggest item for for Mochi banker is salary commission overrides

And that kind of expenses usually accounts for about at least 60, 70 percent depending on different companies. But it's a pretty big driver of the expenses for a IMB. So we watch for expenses very closely, especially we track what is the commission and compare with industry averages. what is the commission rate in terms of BIPs? What is the overrides?

Val Buresch (44:23.47)

Mm-hmm.

Nancy Zhu (44:43.464)

What is the fixed cost? Meaning fixed the salaries that we pay out as compared with, you know, revenue, right? What is the percentage? Or you can compare with the productions. Because in mortgage industry, we tend to compare all those numbers against the production volume and we calculate the bips. What is again on sale bips in terms of bips? Is again on sale 300 bips or 350 bips and then expenses.

You know, what is all those expenses, different categories of expenses, mostly salary and compensation expenses. What are the expenses in terms of production? So these two are the biggest, the most important driver of profitability. And we track that pretty much on a daily, on a monthly basis, obsessively, right? Right, right. Yeah.

Val Buresch (45:16.908)

Mm-hmm.

Val Buresch (45:31.042)

Yes.

Val Buresch (45:35.508)

obsessively, obsessively. Yeah.

Nancy Zhu (45:41.986)

And we also look at a lot of other stuff like on the hedging side, we look at, you know, what is the hedging effectiveness and we also on the production side, we look at the cycle time of pull through because pull through is a big indicator of future productions.

Val Buresch (45:51.0)

Mm-hmm.

Nancy Zhu (46:10.001)

Support through is something that we also look track very closely. We look at profitability by different channels, by different regions, because we need to compare which channel is more profitable versus less. we analyze. Yes, yes, like we have mostly retail channel, but we also have a lot of the affiliate company. So we we.

Val Buresch (46:27.02)

And you're talking about loan channel, Like broker, correspondent. Yeah.

Val Buresch (46:35.096)

Yes.

Nancy Zhu (46:39.241)

affiliate company, and we also have a small piece of consumer direct or maybe a CLD channel. we kind of put them into different perspectives and then try to analyze which channel has a higher profitability versus a lower profitability. And we also track productivity measures very, very closely. So we track how many loans.

Val Buresch (46:41.954)

Yes.

Val Buresch (46:49.122)

Yes.

Val Buresch (47:00.931)

Mm-hmm.

Nancy Zhu (47:08.905)

per loan officer produced and how many loans per processor, how many loans per underwriter, how many loans per total company employee count. And we compare with industry averages, Straml for example. So we want to see where we stand. What is the technology cost as compared with total cost?

Val Buresch (47:18.712)

Right, right.

Nancy Zhu (47:36.97)

And then also, know, lot of time we compare with industry averages to see where we are at. If we are having a high percentage of technology costs, for example, then why is that? And we dig into the numbers to see whether there's any cost savings there, you know, whether we should renegotiate contracts or things like that. So we do look at the different kinds of metrics just to see how we are doing.

Val Buresch (47:48.898)

Mm-hmm.

Val Buresch (48:02.274)

Yes. Yeah. Yeah. So, so to summarize on the financial side, revenue and expenses and revenue, the biggest part is gain on sale. Then on operational side, you're looking at the pull through rate and you're looking at your channel performance and then, and expenses and then on, on product, then you're looking at productivity per MLO and, the different functions in your, yeah.

Nancy Zhu (48:12.637)

Yes.

Nancy Zhu (48:20.135)

Right, and expenses, right.

Nancy Zhu (48:31.517)

Right, yeah.

Val Buresch (48:34.134)

Wonderful. Well, thank you. This is a great advice and hopefully it will spur people's thinking when they hear the podcast. But also I want to go back to the main theme of our podcast, which is the Certified Mortgage Banker Designation. And I know for sure that many people are studying currently to get their CMB designation. So I want to...

To finish with the question, what is your advice for mortgage professionals who are currently studying for their CMB? What should they be doing more or less of?

Nancy Zhu (49:15.081)

I think they, first of all, there's not really a shortcut. I know, at least from my perspective, I spent a lot of time preparing for the CMB exam. And I started January of this year. That was when I had a prep course. I attended a prep course.

Val Buresch (49:20.78)

Ha ha ha ha.

Nancy Zhu (49:38.738)

And I finished by July of this year. So it took me seven months to go through the whole process. I honestly, I didn't have one day of break during that seven months. So meaning I wasn't taking breaks. wasn't I wasn't having weekends. I didn't have holidays. I was studying at night pretty much every night. So I didn't take any breaks for the seven months. So it's it's pretty hard.

journey in my opinion but it's all worth it. and my advice is to focus on understanding instead of memorizing because if you try to memorize everything it's not possible our brain is not designed for that. So focus on studying. I remember so when I talked to my sponsor Bob Mansur

Val Buresch (50:10.007)

Mm-hmm.

Nancy Zhu (50:33.363)

who was a wonderful, wonderful sponsor. He encouraged me along the way. He's always very, very encouraging and very positive. And I told him about my, I shared some concerns to him. I was like, I wasn't even sure, like the oral exam. I feel like, you know, my English is, I'm still learning my English and I don't think this is my forte, you know? And I'm-

Val Buresch (50:36.854)

Yes.

Nancy Zhu (51:02.247)

get nervous when I speak English and he just dismiss dissipates all my concerns and give me lot of encouragement. So encouragement is important, you know, because along the way, especially for me, for about two or three months, I was really, really trying to push through. I was almost on the verge of burnout because I tried and then I realized I'm not learning.

Val Buresch (51:26.518)

Nancy Zhu (51:29.917)

fast enough because there's too much too many materials to cover and it's just not going fast enough. I was being very ambitious and I thought if I give up my yoga class, if I give up my dance class, if I give up all the exercises and spend all the time I don't sleep you know I only keep the minimum amount of sleep. If I do that I can I can remember better, I can do more, I can learn better.

Val Buresch (51:52.088)

you

Nancy Zhu (51:59.091)

But I realized in the middle of it, I felt like I'm not going anywhere. know, I mean, there's just too much. So I got almost, you know, frustrated in the middle. My sponsor, Bob Mansur, encouraged me along the way and pushed me through. And he even gave me six sessions, five or six sessions of training to cover every topic.

every sections, think six sections. So he gave me six lessons with some other CMB candidates. He sacrifices a lot of his personal time for me, and plus the mentorship, the encouragement. So I'm very, grateful for him. My, you know, so I think that's my biggest advice is to focus on understanding, not trying to memorizing. I remember at the beginning of the

Val Buresch (52:38.008)

Wow.

Nancy Zhu (52:56.667)

of the process, Bob sent me a PDF. The PDF's name is 80 different ways to use Track GPT. And I benefit a lot from, although I did not go over the whole 80 different ways, but I begin to set up an account with Track GPT. And I utilize Track GPT a lot in my study. So when you're reading something and because you're not.

Val Buresch (53:06.894)

Ha

Val Buresch (53:14.441)

Yeah.

Nancy Zhu (53:26.311)

working that area so you don't understand what they're talking about. So you just make a copy and paste then into the Chai GPT and say, help me to understand or clarify or explain or give me examples, ask many, many whys. And Chai GPT does a good job in breaking the hard component into small, mini digestible parts. And I've benefited a lot from using Chai GPT. Of course, sometimes it can...

Val Buresch (53:29.708)

Mm-hmm.

Nancy Zhu (53:55.347)

hallucinate and sometimes after I'm asking many many why and explain or clarify or give me examples I I felt like I'm not going anywhere so I stopped and the next day I try one more time so just repeat that process the repetition eventually get me understand the certain topic so the more you understand the more you relieve yourself from memorizing it because even if you memorize it

Val Buresch (53:56.48)

Yes.

Val Buresch (54:06.22)

Yes.

Nancy Zhu (54:22.505)

One, it's not possible to memorize it. Even if you memorize it, by the time you go to the exam, no matter it's written exam or oral exam, just retrieving is gonna be a problem. And also, just memorizing does not help you to really digest and understand the content. So the questions change, or the format changes, or the perspective changes. It's very hard to really answer that question to the point.

Val Buresch (54:25.122)

Correct.

Val Buresch (54:51.918)

Yeah, thank you for sharing this advice. So what I heard is there are no shortcuts you have to learn, but focus on understanding rather than memorization. Encouragement, number two, encouragement is very important. And that was very touching, your story about Bob Mansur. I know Bob very well. I didn't know that he sacrifices so much of his time. And three, utilize modern technology. Why not?

Val Buresch (55:21.89)

Why not? But with some guardrails and with some thought process, yes, how you're going to do that because you can't trust it completely. Yeah, that's fantastic. And that's a wrap. I want to thank you so much, Nancy, for this amazing conversation. I learned from you and I'm actually inspired by you.

Nancy Zhu (55:27.219)

with a question mark.

Nancy Zhu (55:32.285)

Right, absolutely.

Nancy Zhu (55:47.331)

thank you. You are my example, you know. I told you before.

Val Buresch (55:52.87)

thank you. I appreciate that. And I want to thank all of our listeners today as well. Don't forget to follow us. This is CMBConnect, where we go behind the designation.

‍