What's next for Digital Origination? How AI is fueling real ROI

Delivering fast, frictionless digital banking experiences? That's table stakes. These processes continue to accelerate and become more personalized as AI-powered tools make their way into areas like home equity lending, refinance and debt consolidation. In this American Banker Leaders episode, Srini Venkatramani, Head of Product, Technology and Customer Operations at Blend, sits down with Michael Moeser, Senior Content Strategist at American Banker to provide a roadmap on where digital banking is heading and how financial institutions can take advantage to benefit their customer experiences (CX) and ROI. This strategic conversation will cover:

• How banks should rethink digital onboarding, loan origination and cross-selling
• Guidance on using unified, AI-driven platforms to reduce costs and increase productivity
• Real-world success stories from leading banks and credit unions
• Best practices on how to get started using AI-powered tools to enhance CX

Transcription:
Transcripts are generated using a combination of speech recognition software and human transcribers, and may contain errors. Please check the corresponding audio for the authoritative record.

Michael Moeser (00:09):
Greetings everyone. I'm Michael Moeser with American Banker and I'll be your host in today's session. What's next for digital origination? How AI's fueling real ROI. Today we'll be talking about where digital banking is headed and why banks and credit unions need to rethink digital onboarding, loan origination, and cross-selling to report real ROI. Joining me today is Rinni van, head of product and technology and customer operations at Blend. Before we start, let me do a quick introduction for rinni. He's a recent hire at Blend with a background of working at organizations ranging from startups to giant firms such as IBM with a blend, no pun intended, of consulting from Deloitte and my alma mater, McKinsey at Blend. He's responsible for innovation and client delivery, so he has a great view into what's happening today in the industry. One thing I want to mention about Srini is that he's very passionate about a book called Imaginate Forward. This was written by GE Vice Chair Beth Comstock. It's about confronting change and how to grapple with the challenges we face every single day. I guess you could say what worked yesterday, won't work today or even tomorrow. Welcome Rinni

Srini Venkatramani (01:33):
Michael. Thanks for having me.

Michael Moeser (01:35):
Well, I'd like to start off with, we've got, the first question is really about the landscape of digital banking, but I mentioned this book, imagine It Forward. Could you talk to the audience, talk to me about what is it about this book that really struck a chord with you? Why did it resonate so much?

Srini Venkatramani (01:54):
Great question. You talked about IBM, my background. Yes, the CEO Ty, I remember being in one of the all hands meeting,

Michael Moeser (02:02):
She

Srini Venkatramani (02:02):
Talked about comfort and growth does not coexist.

Michael Moeser (02:06):
So

Srini Venkatramani (02:07):
Thinking about how we evolve either as individuals or organizations getting out of the comfort zone and thinking about the future, imagining it forward.

Michael Moeser (02:16):
Got it,

Srini Venkatramani (02:17):
Is where the growth comes and getting out of that comfort zone makes a big difference. So when I read the book, imagine It Forward, really thinking about the future and being ready for it, it's something I take seriously both in my personal life and professional life.

Michael Moeser (02:31):
That makes a lot of sense. So let's move to the current state of digital banking. How do you view the current state? What's the overall landscape and how is it changing?

Srini Venkatramani (02:41):
Yeah, so great question. So when you look at banks and financial institutions, the initial focus was on providing a frictionless experience, but that is becoming more like a table stakes right now. All these financial institutions are dealing with what I call as a dual imperative paradox, tremendous pressure to increase operational efficiency, but they still need to be providing personalization at scale. That's one challenge and one thing that digital banks and banks need to be ready with. The second is advent of new players. These are not traditional players, these are nontraditional players, new age fintechs, tech native companies coming and changing what a B2C experience should look like in the financial world itself.

Michael Moeser (03:22):
Now that certainly causes challenges sometimes at the periphery, but sometimes also at the core of an institution's franchise. Can you talk to us about the challenges that banks are facing with these new fintechs that maybe siphon off the more profitable customers or the younger customers, the newer customers? What does that mean for a bank as they continue to pursue this dual paradox of reduction in expenses yet pursuing this frictionless experience?

Srini Venkatramani (03:54):
When I look at statistics in the market, the friction happens in 90% of the cases In the traditional onboarding experiences. When people are looking for origination or a deposit account, 90% of users face some kind of friction or the other and at least 50% or more people drop off. So conversion rate is a real challenge. So bringing frictionless personalized experience is critical. I'll give a personal example of what we do every day.

Michael Moeser (04:19):
Please.

Srini Venkatramani (04:20):
When we think about Uber as an app, when we need to go somewhere, we take the app, the app already knows where we are, my payment information is stored

Michael Moeser (04:29):
And

Srini Venkatramani (04:29):
I'm off in seconds, but compare and contrast it with what is happening with banking. I still need to enter information that the bank may already know about me and some of the documents may get rejected and I may not get answers for weeks. People are expecting B2C experiences in their banking experience, which some of the fintechs are doing right now,

Michael Moeser (04:51):
But doesn't. Some of that friction, I sometimes expect there's a degree of friction as sort of a feeling of security that it's asking me who I am or confirming my birthdate, my social security number, and there's a degree of I don't want it be so easy that somebody could sign up very easily. So is that, and I don't know if banks feel that way, that there needs to be a degree of friction or is that something that we should really consider going away from?

Srini Venkatramani (05:20):
It should be a balancing act. As you mentioned, taking privacy data security seriously is paramount, especially in a conservative and regulatory industry like finance. Having said that, reducing friction value can is very critical. For example, by just giving date of birth and your phone number and of course authenticating that is your phone number.

Michael Moeser (05:41):
A

Srini Venkatramani (05:41):
Lot of information is available right now where you live, your social security number and many other things can be pulled out. These are things that anyway need to be entered in an application form.

Michael Moeser (05:52):
So

Srini Venkatramani (05:52):
Instead of asking the borrower or a user to enter 20 fields, if they enter those two fields, everything else gets populated. And again, these are experiences people are now used to in the B2C world and they are expecting it as a baseline.

Michael Moeser (06:07):
Well, I know that many times people, whether they subscribe to Netflix or Apple and all of a sudden Apple tv, you're getting what's the next program I should watch? Or thinking about your purchasing from Amazon, you bought this, buy that. So similar things, but there's a feeling from the banking world that consumer expectation is rising so quickly. Are banks keeping up with that?

Srini Venkatramani (06:33):
Some of them are. Some of them are not. When I look at it, the companies that are keeping up are the ones that are taking a tech first mindset

Michael Moeser (06:40):
Instead

Srini Venkatramani (06:41):
Of being a bank at its core. Right? Think about being a tech first mindset.

Michael Moeser (06:44):
Well, what is that mindset? Could you maybe expand on that?

Srini Venkatramani (06:47):
So the mindset is around how do you think about customer experience? How do you think about a customer centric approach of how you think about your operations

Michael Moeser (06:55):
Instead

Srini Venkatramani (06:56):
Of a product centric approach? You talked about the example of Netflix.

Michael Moeser (07:00):
Netflix

Srini Venkatramani (07:01):
Has millions of movies and series available, but it kind of surfaces what is applicable for you.

Michael Moeser (07:08):
So

Srini Venkatramani (07:09):
As a bank customer, if I'm someone who is consistently saving money,

Michael Moeser (07:14):
Showing

Srini Venkatramani (07:14):
Me a high yield savings account is an example of how you could personalize as a bank. Oh, interesting. Similarly, someone is a freelancer and maybe giving them a tax free approach in terms of doing it, it's an approach. How do you really hyper, hyper-personalized is something that you can bring to the market.

Michael Moeser (07:32):
Now we talked earlier about rethinking the onboarding process and I believe that we rethought the onboarding process just a few years ago with the pandemic where banks realized if their system wouldn't allow you to open a checking account or apply for a mortgage loan because you had to go to the branch and the pandemic prevented that. So they had to rethink that. Why are banks needing to rethink banks and credit unions needing to rethink that onboarding process? Wasn't that solved or is it the issue of comfort and growth? Again,

Srini Venkatramani (08:05):
It's comfort and growth. What was best of breed and or best in class are becoming table stakes right now. Statistics again, say 90% or more of the people have at least two or more relationships with financial relationship with companies, and that means how do you really retain your customer and how do you make sure again, you reduce the friction becomes critical. If not, people are going to go and create those other financial relationships in terms

Michael Moeser (08:36):
Of is it because they're finding what they need from those other relationships

Srini Venkatramani (08:40):
As yes and or the existing provider is not actually maintaining even the relationship. So if you know about me,

Michael Moeser (08:48):
Srini,

Srini Venkatramani (08:49):
I am banking with someone, but you do not carry that information or you are forcing me to enter that information again, I might ask, we go somewhere else.

Michael Moeser (08:58):
That

Srini Venkatramani (08:59):
Is the frustrating part of it. That is the frustrating part of it. Yes.

Michael Moeser (09:01):
And talk to me about realizing real ROI. It's in the title of the session, what is real ROI and how do you realize that? Yeah, great question from AI specifically in that

Srini Venkatramani (09:14):
Great question. I would say three parts. When we think about the cost involved in any of the originations, mortgages today cost about $12,000

Michael Moeser (09:23):
To

Srini Venkatramani (09:23):
Originate. Even consumer lending products typically cost $500 or more. So there is a lot of money being spent by the financial institutions in supporting the lifecycle from the front office to the middle office to the back office. So that needs to become much more efficient. So when we think about application of ai, looking at how do you automate things, how do you make sure that you optimize the flow is going to be critical to reduce that cost. That's part one. The second part of this is around how do you increase the conversion rate? In many institutions we see the conversion rate is not that high, as high as it can be. For example, in many cases we are seeing that the conversion rate are as low as 20%. So 10 people start at the top of the funnel,

Michael Moeser (10:08):
Only

Srini Venkatramani (10:08):
Two people come and buy that product. So how do you increase the conversion rate? That is number two and three, all of these financial institutions have a lot of products. How do you cross-sell the products? Another study says that typically the ROI for you to retain and cross-sell and upsell to your customer is 10 times as acquiring a new customer. So how do you think about this? In the software world that we come from,

Michael Moeser (10:32):
There

Srini Venkatramani (10:32):
Is a term called as net revenue retention. This is how

Michael Moeser (10:35):
Net revenue retention,

Srini Venkatramani (10:36):
Retention, this is how you grow your relationship with your current customer in terms of giving them customer delight features so that they buy more from you. That's a metric that software companies take into account very seriously. Financial institutions, again, tech first mindset is what I talked about. If they can think about NRR, the net revenue retention that is going to be significant for them in terms of their top line growth and bottom line.

Michael Moeser (11:00):
Can you give me an example of the three things that you mentioned? The first one was really around efficiency in terms of reducing the cost of that $12,000 mortgage or the $500, I don't know,

Srini Venkatramani (11:14):
Consumer loans,

Michael Moeser (11:14):
Consumer loan. So how do you involve AI that really reduces that efficiency that speaks to that dual paradox that you're talking about.

Srini Venkatramani (11:24):
So when we look at the cost involved, I will even break it into two parts. There is about six to $7,000 being spent in the sales related activities, acquiring a prospect, getting them through the borrowing process.

Michael Moeser (11:37):
Then

Srini Venkatramani (11:38):
About four to $5,000 is being spent in just processing. And when you look at processing, it is the same set of documents. It's my income statement, my employment statement, my assets from banks. The same documents get STA and compared multiple times through

Michael Moeser (11:53):
Stared and compared

Srini Venkatramani (11:55):
Multiple times throughout the lifecycle. It is being looked at when I am submitting it,

Michael Moeser (11:59):
It's

Srini Venkatramani (12:00):
Looked at by a loan processor. It's being looked at by an underwriter, it's looked by someone before my fund gets released. It's done by post-close QC teams. It's done by the compliance teams. The same set of documents is being looked at multiple times. Yeah, it's become really, really good. Much better than human beings

Michael Moeser (12:18):
Enabled

Srini Venkatramani (12:18):
To compare the documents. Again, what a fan or Freddie or any other regulations say and significant cost savings and operational efficiency as an example. The second thing I talked about, increasing top line, when you are able to increase the conversion rate and reducing the friction, we have been able to do it within our own platform. By taking the friction out of the system, introducing ai, we are able to increase the conversion rate from 30% for customers overnight to 60%. That's amazing. Doubling down in terms of conversion rate. The third one is the ability to cross-sell. So again, through AA you can surface personalized offers. For example, someone is buying a car loan, but you may want to buy a gap insurance which you can offer.

Michael Moeser (13:05):
Okay, got it. Right.

Srini Venkatramani (13:06):
So you could cross sell products. So there are numerous examples of how this can be done

Michael Moeser (13:10):
Well, how do you advance the consumer experience then if you're talking about leveraging AI to improve the process, is that going to improve the consumer experience?

Srini Venkatramani (13:21):
And I'll also suggest that banks should think about this in broader scale, dream big, but start small in terms of where it is, where is right now as half a trillion dollars are being spent by the big firms, tech firms

Michael Moeser (13:34):
Really

Srini Venkatramani (13:35):
Building the plumbing and infrastructure of ai, even within the financial institutions, about 25 billion or so is being spent this year just application of AI. And studies say that it is going to grow to 300 billion plus in the next 78 years, 30% or more cagr. So a lot of money is going into AI overall and within banking, but start small capabilities which has advanced really well as document management. And I usually see it in three use cases where documents are classified. For example, instead of a human being saying that A, this is a W2 or not, AA can do it much more faster and with higher fidelity. Similarly extracting the data from a document and populating because that's what people do, read the document and they're entering it into a system anyway, you can automate that or you can generate documents. So anything related to document is fairly advanced right now. So in that bigger picture or the roadmap, I would say start small with anything related to documents so there's less stare and compare. That's right.

Michael Moeser (14:36):
So I want to talk to you about before the session here we were chatting, you mentioned to me that you were not an athlete growing up and you recently took up marathon running and I think you were recently in the New York marathon. Talk to me about what that does for you from a personal perspective and how it translates into your business.

Srini Venkatramani (14:58):
Yeah, great. On the personal side, it showed me the value of resilience.

Michael Moeser (15:03):
Resilience

Srini Venkatramani (15:04):
In terms of what you need to do and having a plan. Having a plan, start small. The favorite code for me is the running the marathon is not actually start of the journey. It's the last 26.2 miles of the journey that you took. Even though it sometimes looks like a thousand miles, especially the last six weeks, the last 26.2, I will keep that in mind and running through Central Park the last two miles, while it was exhilarating, it was also tough, but I have taken that into my heart over the last dozen years with my running is in my personal life, endurance, resilience, having a plan, a mind over body in terms of how you think about it. Start small but think big. Have audacious goals.

Michael Moeser (15:45):
Audacious goals. I

Srini Venkatramani (15:46):
Like that. Same with professional life and even for the AA that we talked about, that's something that I mentioned to my customers, which is you know what, you can have an ambitious plan, but you don't need to worry about reaching that plan today, right? You need to think about what is the first step that you are taking document as an example, right? Start thinking about a small use case, start applying it, embed within your workflow

Michael Moeser (16:08):
And

Srini Venkatramani (16:08):
See the results. Again, go with the tech first mindset of iterative approach and good things will happen.

Michael Moeser (16:14):
Well talk to us about some recent integration examples. You guys have been deploying software out there into the banks, into the credit unions and you've had some good success stories. Can you talk to us about what you've recently done and how that's been impactful for your customers?

Srini Venkatramani (16:30):
Yeah, happy to. When I think about Blend, typically it's very well known for innovative and integrated customer experience. We bring a lot of elements together. For example, even within the application process, instead of you uploading documents, we automatically validate income or employment or other verifications that need to be done. We talked about entering just two fields. How can the entire set of fields can be filled? So that innovative integrated experience is something that blend typically brings. In the recent past, what we have done is we have taken this document capability and embedded it through the lifecycle

Michael Moeser (17:06):
Of

Srini Venkatramani (17:06):
An application process. For example, someone is applying for a mortgage and the bank is looking for the last two years. W2 in the past, people used to submit the documents and wait for the results to come.

Michael Moeser (17:19):
Right now you have to take a picture or submit it to a Dropbox, Dropbox or provide access to my W twos

Srini Venkatramani (17:28):
And it takes time. Sometimes it goes back and forth

Michael Moeser (17:30):
And you never know if it works.

Srini Venkatramani (17:32):
It may not work. And there are actually statistics that say that the back and forth, the follow-up is about 2.8 times

Michael Moeser (17:38):
For

Srini Venkatramani (17:38):
Each of the load

Michael Moeser (17:39):
That yes,

Srini Venkatramani (17:40):
Now all of this can be, it's

Michael Moeser (17:42):
Frustrating.

Srini Venkatramani (17:42):
It is, and you can cut all of this and even do it upfront, submit the document. For example, you submit employment letter, not your W2.

Michael Moeser (17:50):
Yeah,

Srini Venkatramani (17:51):
Yay would be gentle, but it'll say we were looking for W2 and then you submit your 2021 W2, you say, no, we were looking for 20, 23 and 24. Okay. Right, right. So that's a perfect example of how you can do all of this early in the process in a self-service manner and the customers are not dealing with that frustration instead of going back and forth. That three times that I talked about, you can just do it upfront. An example on the consumer lending side, on the auto loans,

Michael Moeser (18:19):
Typically

Srini Venkatramani (18:20):
A purchase agreement is provided by the lender and there are thousands of purchase agreements that are available. So it's not like a standard form which you can typically train an OCR, but when you have thousands of variations, OCR cannot take it, but Gena out of the box can read it, understand it, and can extract data from this. So we have integrated within our consumer lending platform. So you are a borrower, you are applying for an auto loan and you went to a dealership, you got the purchase agreement, you just take a picture and yay is able to read it, take information from that agreement, populate your application and you're done.

Michael Moeser (18:58):
Yeah, I've heard of the term trusted document and I guess as you once verified that it reduces the degree of stare and compare. Where does that play in terms of the workflow that you do in terms of you have this document, gen AI is reading it, and then does it really eliminate that stare and compare that you're having to do it a second time, third time, and then does it drive the value chain

Srini Venkatramani (19:25):
Forward? Quite a bit. Quite a bit. And usually if you look at human beings doing it, the industry statistics is the fidelity is about 97%. Now with generative ai, even with the first tab reading it for the first time, it is far exceeding that 97, but we are not even settling for it. We are looking at four nines or fin nines,

Michael Moeser (19:43):
99.999,

Srini Venkatramani (19:45):
Et cetera. And the way we are doing it is based on a consensus based model. So running the same question multiple times through the same LLM and make sure there is a consensus. If you're asking 10 times, if the model says all 10 times it is good, then we kind of take it or running the thing through multiple models so we can run it through Anthropic model or chat GPT model or Gemini models and make sure there is consensus around that. Models may hallucinate, but they don't typically hallucinate about the same thing. So we are fairly confident about the accuracy and fidelity. So what you just talked about the trusted information, how can you trust it's far exceeding what humans doing

Michael Moeser (20:28):
And you said humans get a 97% fidelity rating? That's

Srini Venkatramani (20:32):
Correct.

Michael Moeser (20:32):
Okay, okay. I mean, and I guess if you stare and compare it too many times, that probably goes way down.

Srini Venkatramani (20:39):
That's correct. And the risk and the liability for banks for the 3% is actually higher. Assume that you are a bank. Typically banks do not hold, especially on the Margate side. They don't hold it in their book. They sell it to sell it to the investors within the 90 days. And the investors are going to ask, did you follow the policy? And if you did not follow the policy, you're not able to sell it, it is going to sit on your book or you are going to sell say 60 cents to the dollar. So the liability by not following something is much higher. So increasing that fidelity is going to help the banks again, not just in operational efficiency but in their top line.

Michael Moeser (21:16):
So you're saying it's not just an imperative that they do this, but if they don't, there's significant consequences, negative consequences, and probably not just the regulatory side, but also competitors.

Srini Venkatramani (21:30):
That's correct.

Michael Moeser (21:31):
They come in and take customers. Yeah,

Srini Venkatramani (21:33):
That's correct. And that's a trend that we are seeing with all the vertical integration that's happening in the industry. A lot of these mortgage companies, the new age ones that have come,

Michael Moeser (21:43):
They

Srini Venkatramani (21:43):
Are becoming more tech first. So if I'm a traditional financial institution, a bank or a credit union, it is becoming more at table stakes and a game of survival at this point.

Michael Moeser (21:54):
Does that mean a difference for the consumer experience then if I'm going to, going from a traditional bank that is asking me to take pictures, upload PDFs, or provide access versus a new age bank that maybe I need this car loan mortgage, et cetera, and taking a different approach that is more tech first, it sounds like it would be a better experience.

Srini Venkatramani (22:19):
Amazingly different experience. So again, if you look at the statistics in the market, applications now can be started and finished in as less as 90 seconds. That's amazing. You don't need to wait for a lot of time. And Robinhood recently launched a product where you could even have your direct deposit happen in the matter of minutes. So a lot of these fintechs are coming up with fantastic customer experiences that banks and financial institutions need to be taking into account. A typical home equity loan used to take 45 days.

Michael Moeser (22:54):
Now

Srini Venkatramani (22:54):
For example, at Blend, what we have done is we have moved a lot of these things upfront. I talked about the pre-fill that

Michael Moeser (23:00):
Giving

Srini Venkatramani (23:01):
The two fields, a lot of things get filled. We also get the valuation of the particular property for which you are going to have home equity. You are also able to consolidate debt because we want to use the money instead of 45 days. We are able to do it as low as 10 days. That's

Michael Moeser (23:15):
Amazing.

Srini Venkatramani (23:16):
It's a significant experience and you as a borrower, you don't need to wait for days. You can literally get an answer within a couple of minutes and then you decide whether you want to do it or not.

Michael Moeser (23:25):
Right. That's amazing. So talk to us about how can banks stay ahead in this, because that sounds like a real game changer going from 45 to 10 days. So what can banks do if they want to stay ahead of the curve

Srini Venkatramani (23:37):
Growth mindset? I say get out of the comfort zone, have that growth mindset and have the tech first mindset, have a plan, but again, don't be worried about how they are going to achieve the plan. Think about that as small steps. And again, my marathon analogy is I think about when do you accelerate and when do you recover. There are certain areas where they can accelerate as AI capability exists today. So instead of waiting and being conservative, start today, especially the document area, do it right away, see how it works, iterate like

Michael Moeser (24:09):
How it document. So the document management, document handling, document verification, that's an area where you need to step on the accelerator. Now what about where you need to maybe dial back? You mentioned there are places where you need to, I guess in your marathon accelerate maybe downhill I guess where it's easy and then sort of conserve that energy. Where would a bank need to think about conserving that energy as they look at ai

Srini Venkatramani (24:34):
The last few miles? Anything right, downhill or uphill, right? Anything is a stretch, but looking at that they need to stop and recover. My view would be things around anything that involves sensitive data and where they may provide incorrect results or updates.

Michael Moeser (24:51):
Give us an example.

Srini Venkatramani (24:52):
So for example, when you are thinking about making addition, when you are underwriting

Michael Moeser (24:57):
And

Srini Venkatramani (24:58):
You are making addition about whether to provide the loan or not, and at what rate? I would say get your data correct and clean first before you are putting this in front of the customers. So that's one way to look at it because when you're thinking about documents, you don't get it wrong, but when you're thinking about underwriting and others do it. So that final decision or

Michael Moeser (25:18):
If it's customer facing other than say a chatbot, but if it's customers facing that there's an impact. If they say yes or no, this could be the difference between Michael or Srini taking the loan or not taking the loan.

Srini Venkatramani (25:32):
Yeah, yeah, I would do that. And also that there are some still questions or question marks. Have human in the loop. Make sure that you are injecting the appropriate expertise in the workflow processes so that if there is a question, there is a doubt there is someone who is able to quickly verify.

Michael Moeser (25:53):
Can we dial down on that human in the loop? Because that's a term that's used often and often as it relates to AI because the human in the loop, there's often, and we do a lot of research and we see that there's, is there a concern about people losing their jobs to ai? And oftentimes we hear the answer is no, but that it'll change the jobs that we do and can AI be trusted to make that decision? Talk to us more about what that human in the loop in terms of where the bar is. Where do you really need to have that? Where you say AI can't make that decision.

Srini Venkatramani (26:29):
So even AI usually view it as augmented intelligence,

Michael Moeser (26:33):
Augmented intelligence,

Srini Venkatramani (26:34):
Not intelligence, not just the artificial intelligence. That's one part. The second thing I would say, think about this as more like a sous chef, not like a sous chef, not instead of a head chef where someone is helping cutting the onions, creating.

Michael Moeser (26:46):
So you're still Gordon Ramsey, but we've got the sous chef.

Srini Venkatramani (26:48):
That's right. What yay is doing is kind of automating some of the repetitive tasks. It is automating some of the things that may need more rigorous thinking in terms of what it happens, but you still are in charge. So that's the way I would think about it. Let's take the loan origination as an example,

Michael Moeser (27:05):
Please.

Srini Venkatramani (27:06):
Today, loan offices are spending time both wearing a sales hat and also an operational hat. And they're not necessarily spending time in building relationships.

Michael Moeser (27:16):
So

Srini Venkatramani (27:16):
If some of the operational elements can be automated through a loan officer could spend more time with their customers, building the relationship, bringing the right product to them, that's one. And second, with all the interest rate volatility that is happening, companies are also struggling to think about how do they staff for a surge of increase in volume that may come now AA could help them smoothen that elasticity in terms of how you staff for any of this surge and depths that happens in the mortgage volume or any other loan origination volume.

Michael Moeser (27:51):
Can you talk to us about customer engagement? What's the role of that customer engagement in this process? That sounds like it's very important, especially as you mentioned, relates to staffing too.

Srini Venkatramani (28:02):
Yeah, the dual imperative paradox I talked about is real. In this operational efficiency there is still an expectation. There is hyper-personalization

Michael Moeser (28:11):
And

Srini Venkatramani (28:12):
The hyper-personalization becomes a foundation for the customer engagement, knowing about the customer, and this is where the augmentation part can come, where me as someone, a banker or a loan officer working with my customer, if I have the insights about the particular individual when I am chatting with them, it becomes much easier. That's where the engagement comes. That's one. Second, if the systems and the banks and financial institutions know enough about me, if they can be more proactive and predictive in terms of what may be of interest for me

Michael Moeser (28:48):
Now, proactive and predictive, we all get a lot of reach outs. People will call you out on the phone whether email, et cetera. And sometimes these messages don't, they seem to fall flat, it's open a checking account, here's $600, but I may already have two checking accounts. Talk to me about the hyper-personalization where AI can really glean that information. And maybe it's from the documents that you talk about that it reviews,

Srini Venkatramani (29:16):
It's about the documents. So yay is also good in combining both unstructured and structured data.

Michael Moeser (29:22):
So

Srini Venkatramani (29:22):
You get unstructured data from the documents and you have structured data in the databases of these banks and yay is really good in combining both of them and coming back with an insight. So definitely something that it can hyper personal. The example that you talked about, instead of targeting you with a brand message about opening an account, you get $600, you already have two accounts, but make it somewhat more meaningful for you. For example, it can say that, you know what we have seen, Michael, you are doing these type of activities every month.

Michael Moeser (29:52):
If

Srini Venkatramani (29:53):
You have this type of an account, we are going to WA your charge every month because you seem to be a frequent user of this particular service. So being very hyper-personalized in terms of what may appeal to you is something that the banks could do better through ya.

Michael Moeser (30:09):
Where are banks in the hyper-personalization? I mean I realize that Netflix certainly has it at one level. Amazon has another, and it may not be fair to compare because if Netflix gets the wrong movie, recommended to me, who cares? But banks have compliance, they have regulators, and so it's a completely different environment in terms of what they can promise, what they can promote. But where are they in that hyper-personalization? Are they at par? Are they slightly behind or is it still very greenfield opportunity? For many banks,

Srini Venkatramani (30:45):
I would say mixed, but more early stage. Early

Michael Moeser (30:48):
Stage, okay.

Srini Venkatramani (30:49):
Early stage. For a lot of banks, from what we see, they do have the data. So it's not like they don't have the data about the customers. They do have the documents.

Michael Moeser (30:57):
So traditional banks actually have an advantage here. That's right, because they've got the data, they know the customers, it's just the fintechs have I guess out of the gate, come up with some new unique products that may resonate with some people. It's

Srini Venkatramani (31:10):
The customer-centric mindset,

Michael Moeser (31:11):
Customer-centric

Srini Venkatramani (31:12):
Mindset,

(31:13):
How people think about it. And then really making sure that the data is translating into insights. The other challenge a lot of traditional banks and financial institutions face is they have silos. They have siloed systems, they don't talk to each other. For example, an information about my income may be in system one and anything else that I'm doing in terms of my transactions day in, day out, maybe in another system, they don't talk to each other. So a bank cannot make an impression or an opinion about me because those systems do not talk to each other. So making sure that there is an integrated view of the data of the customer, customer centric mindset is going to be the key fundamental difference for banks and financial institutions.

Michael Moeser (31:57):
I'm curious in terms of, I know how banks are often set up, you may have a mortgage unit, it's running on its own system. Sometimes these are due to acquisitions. You have a credit card unit and you may have both products, yet neither can talk to each other or we can't get this unified view of Sini, this is what his financial life is like. Can the systems that you use the platforms sort of overcome those silos, overcome those barriers? It

Srini Venkatramani (32:26):
Does. And again, that's what Blend does. And the thing that we want to differentiate is really have a unified experience for the customer, the customer-centric view about various products. Me as a consumer may be getting from the back, whether it is mortgage, whether it's home equity, I may want to refinance, I may want to get an auto loan, I might want to get a personal loan, I want to get a credit card, but I also want to open a deposit account. I want to do a savings account.

Michael Moeser (32:52):
Supporting

Srini Venkatramani (32:53):
All these products within a single system allows people to bring the data together and also provide the cross-sell. At the moment of my purchase, as I mentioned, I'm buying an auto loan, can I offer a gap insurance? Swee is buying a credit card, can we ask him to kind of consolidate debt with someone else? So provide that opportunities at the moment of purchase increases the top line for the max.

Michael Moeser (33:18):
Can you talk to us about Blends plans and how they going forward, how you see that helping banks? And there was a product, we talked about doc ai. Can you talk about how that works and then the process? Yeah,

Srini Venkatramani (33:33):
We are very excited about what a can do for our customers, which are banks, financial institutions, and credit unions.

Michael Moeser (33:41):
So

Srini Venkatramani (33:41):
Very excited about what a can do in terms of simplifying the process, increasing with operational efficiency, increasing in top line. You talked about dock a a. We have embedded that within our workflow instead of being a separate product, we have incorporated that into how we kind of sell, deliver and implement it for our customers. So for example, within our point of sale system in mortgage, this document management and dock a capabilities already immersed in terms of how we do things. So if you're applying for a loan using Blend product, you won't be submitting document storage box, right? It'll be automatically read, et cetera. So that's an example of what we have done. But then yay, we are also looking at many other areas. For example, we want to think about intelligent orchestration. So especially when there is seasonality to the business, but the banks and financial institutions want to have a stable number of workforce.

Michael Moeser (34:38):
We

Srini Venkatramani (34:38):
Want to make sure that there is intelligent orchestration of work instead of it being fast and fussed out. We are looking at ways where the most complex of a loan goes to the most experienced person.

Michael Moeser (34:50):
So

Srini Venkatramani (34:50):
Again, an application of

Michael Moeser (34:51):
Intelligent

Srini Venkatramani (34:52):
Orchestration. Yeah, so that's one that we are doing. The other use case we are thinking about is can we think about decisioning and underwriting again within the right, appropriate boundaries in terms of building it. That's another area we are looking at. The other last area that we are looking at is how do we provide actionable and predictive and prescriptive insights either directly to the consumer of course with the financial institution's approval or provide that as a agent or a catalyst for the bankers to use. Give me an

Michael Moeser (35:23):
Example of how that would work. If you're got a bank account with you, credit card with you, maybe a mortgage and you're seeing how I'm going through my daily life, how will you use this information to predict what's next and potentially cross-sell a product to me? Yeah,

Srini Venkatramani (35:41):
I'll give you a couple of examples. You may have a mortgage, really good rate, it may be 3% even though the current interest rate is close to seven seven,

Michael Moeser (35:50):
But

Srini Venkatramani (35:51):
There may be value in your home equity. So by calculating what's your current interest and what could be a home equity, because we may want to monetize that providing you a very prescriptive recommendation, Michael, you may be able to leverage about 200 K in your home equity. Nice. Yeah, that's one. And then even go to the next step, which is taking this home equity, can you go and consolidate some debt? So Michael, we see you having three credit cards and an auto loan. All of them carry 14% interest.

Michael Moeser (36:26):
This

Srini Venkatramani (36:26):
Home equity is 8%. Why can't you take this home equity and then go and pay this debt? Sorry, that's another example.

Michael Moeser (36:34):
That's very interesting. I mean I think a lot of times we don't think about that in our daily lives and home equity is one of those hidden things that we just don't see and it continues to grow and grow and yeah, I think once you get the mortgage loan, you set it and forget it and if things go along and then you're focused maybe on just the credit card.

Srini Venkatramani (36:54):
That's an interesting statistic. Right now, the home value in the US it's about $35 trillion,

Michael Moeser (37:01):
35 trillion,

Srini Venkatramani (37:02):
Trillion dollars. So it has quadrupled in the last 15 years. And more than 75% of the people have a very attractive mortgage rate, less than 5%.

Michael Moeser (37:13):
So

Srini Venkatramani (37:13):
They are not going to refinance, but they are sitting on this money. Every household has about $315,000 in home equity. So people don't want to do anything with their first mortgage,

Michael Moeser (37:24):
But

Srini Venkatramani (37:24):
They want to do something with their equity and because they may have some higher interest debt. So home equity is an area where again, we recently launched a product called as Rapid Home Lending.

Michael Moeser (37:35):
Rapid Home lending

Srini Venkatramani (37:36):
That allows customers, again, frictionless, enter a couple of fields, we know the value of the house, we know what you would qualify in terms of interest rate. We are able to pull out what are the debts that you may have. So within two minutes you could take a hundred K off, you could pay off four different debt, and you are going to say a thousand dollars, that's

Michael Moeser (37:58):
Amazing.

Srini Venkatramani (37:58):
And you just click on a button, two minutes, it's all done. So this is a process where, which used to take again weeks. Now it's a matter of minutes and days.

Michael Moeser (38:07):
And I guess if I had that product, you made that offer to me. I would love to also have a savings product that taking the extra interest that I've been paying along and maybe build a savings program or pay down the debt. That would be amazing. That's

Srini Venkatramani (38:20):
Right. Again, cross-selling savings products or you are on auto loan that is at 14%, but even if you're not interested in consolidating, but you could qualify for 11% auto loan, so you could do an auto refinance. So there is a lot of opportunities. Again, that's where I mentioned that the RO is about 10 times

Michael Moeser (38:39):
In

Srini Venkatramani (38:39):
Cross-selling and upselling to your existing customer instead of even thinking about a new customer.

Michael Moeser (38:44):
I think there's a lot of value there and I really appreciate you spending time with us today and talking about this. So I guess in closing, could you share some thoughts with the audience in terms of going forward and imagining it forward, if you will?

Srini Venkatramani (38:59):
Yeah. First of all, thanks for having me. This has been a great chat. I would say do three things. One, think big, but start small, think big, start small, start small and think about agile approach in terms of how we do it. So that's number one. Second, take a tech first mindset,

Michael Moeser (39:15):
Tech first mindset

Srini Venkatramani (39:16):
And get into a growth mindset in terms of get out of the comfort zone. Try to do things that may not be typical to how a financial institution may work, but maybe something that's what the market is looking for in terms of merit. And then don't be worried about failing. There are going to be experiments where you are going to be successful and there are going to be experiments where you are going to fail, but take it in the stride. Got it. And learn from that and move

Michael Moeser (39:43):
On. Think big, start small tech first mindset. Don't worry about failure. Those are great things to think about. Well, thank you for spending time with us and I want to thank the audience for spending time with us today. Thank you. Goodbye. Thank you.